Method and system for industrial internet of things

Abstract

An embodiment of the invention discloses a system for data collection, processing and utilization, the system comprising a platform comprising a computing environment connected to a local data collection system having sensor signals obtained from a first machine in an industrial environment; the system also includes a first sensor in the local data collection system for connection to the first machine, a second sensor in the local data collection system; and a crosspoint switch located in the local data collection system having a plurality of inputs and a plurality of outputs, the plurality of inputs including a first input connected to the first sensor and a second input connected to the second sensor, wherein the plurality of outputs comprises a first output and a second output for switching between: using the first output for switching between the transmission of the first sensor signal and the transmission of the second sensor signal, and there is a contemporaneous transmission of the first sensor signal from the first output and the second sensor signal from the second output.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Patent Application No. 62 / 333,589, filed on May 9, 2016, and entitled "Powerful Industrial IoT Arrays"; U.S. Provisional Patent Application No. 62 / 350,672, filed on June 15, 2016, and entitled "High Sampling Rate Digital Recording of Measurement Waveform Data as Part of an Automated Sequential List of Streaming Long Duration and Gapless Waveform Data for Storage for More Flexible Post-Processing"; U.S. Provisional Patent Application No. 62 / 412,843, filed on October 26, 2016, and entitled "Methods and Systems for Industrial Internet of Things"; and U.S. Provisional Patent Application No. 62 / 427,141, filed on November 28, 2016, and entitled "Methods and Systems for Industrial Internet of Things." All of the above-referenced patent applications are hereby incorporated by reference in their entirety as if fully set forth herein. Technical Field

[0003] The present invention relates to methods and systems for data collection in industrial environments, and methods and systems for performing monitoring, remote control, autonomous actions and other activities in industrial environments using the collected data. Background Art

[0004] Heavy industrial environments, such as those used in large-scale manufacturing (e.g., aircraft, ships, trucks, automobiles, and large industrial machinery), energy production (e.g., oil and gas plants, renewable energy), energy extraction (e.g., mining, drilling), and construction (e.g., large building construction), involve highly complex machines, equipment, and systems, as well as highly intricate workflows. To optimize the design, development, deployment, and operation of diverse technologies for overall effectiveness, operators must be able to account for numerous parameters and metrics. Traditionally, data collection in heavy industrial environments has been done using dedicated data collectors, often recording multiple batches of specific sensor data on media such as tapes or hard drives for later analysis. These batches of data were then returned to headquarters for analysis, such as signal processing or other analysis performed on the data collected by the various sensors. The results of these analyses were then used to diagnose problems within the environment and / or identify operational improvements. Traditionally, this effort often took weeks or months and involved limited data sets.

[0005] The advent of the Internet of Things (IoT) has enabled the continuous connection of a wider range of devices and the continuous interconnection between these devices. Most of these devices are consumer devices, such as lights, thermostats, etc. This is more difficult to achieve in more complex industrial environments because the range of available data is often limited and the processing of data from multiple sensors is very complex, making it difficult to form "smart" solutions that are effective for the industrial sector. There is a need for improved methods and systems for data collection in industrial environments, as well as improved methods and systems for using the collected data to provide improved monitoring, control, intelligent problem diagnosis and intelligent operation optimization in a variety of heavy industrial environments. Summary of the Invention

[0006] Provided herein are methods and systems for data collection in industrial environments, as well as improved methods and systems for using the collected data to provide improved monitoring, control, intelligent problem diagnosis, and intelligent operational optimization in a variety of heavy industrial environments. These methods and systems include methods, systems, components, devices, workflows, services, processes, and the like that are applied in various configurations and locations, such as: (a) at the edge of the Internet of Things, such as in the local environment of a heavy industrial machine; (b) in a data transmission network that moves data between the local environment of a heavy industrial machine and other environments, such as other machines or remote controllers, such as the enterprise that owns or operates the machine or the facility where the machine operates; and (c) in cloud computing environments and on-premises computing environments where facilities are deployed to control the machine or its environment, such as in the enterprise that owns or controls the heavy industrial environment or the machines, devices, or systems deployed therein. These include a range of methods and systems for providing improved data collection, as well as methods and systems for deploying augmented intelligence at the edge of controllers in industrial environments, in networks, and in the cloud or on-premises.

[0007] Disclosed herein are methods and systems for continuous ultrasonic monitoring, including continuous ultrasonic monitoring of rotating elements and bearings of energy production facilities.

[0008] This article discloses a method and system for cloud-based machine pattern recognition based on remote simulated industrial sensor fusion.

[0009] Disclosed herein are methods and systems for cloud-based machine pattern analysis of multiple simulated industrial sensor state information to provide expected state information of an industrial system.

[0010] Disclosed herein are methods and systems for device-mounted sensors and data storage for industrial IoT devices, including device-mounted sensors and data storage for industrial IoT devices, wherein data from multiple sensors is multiplexed in a device for storing a fused data stream.

[0011] Disclosed herein are methods and systems for a self-organizing data marketplace for industrial IoT data, including a self-organizing data marketplace for industrial IoT data, wherein available data elements are organized in the marketplace for consumption by consumers based on training a self-organizing facility using a training set and feedback from marketplace success metrics.

[0012] Disclosed herein are methods and systems for self-organizing data pools, which may include self-organizing data pools based on utilization and / or revenue metrics, including utilization and / or revenue metrics tracked for multiple data pools.

[0013] Disclosed herein are methods and systems for training an AI (Artificial Intelligence) model based on industry-specific feedback, including training an AI model based on industry-specific feedback that reflects a measure of utilization, benefit, or impact, and wherein the AI ​​model operates on sensor data from an industrial environment.

[0014] Disclosed herein are methods and systems for self-organizing groups of industrial data collectors, including self-organizing groups of industrial data collectors that organize among the industrial data collectors based on the capabilities and status of group members to optimize data collection.

[0015] Disclosed herein are methods and systems for an Industrial Internet of Things distributed ledger, including a distributed ledger that supports tracking transactions performed on Industrial IoT data in an automated data marketplace.

[0016] Disclosed herein are methods and systems for self-organizing collectors, including self-organizing, multi-sensor data collectors that can optimize data collection, power, and / or revenue based on conditions in their environment.

[0017] Disclosed herein are methods and systems for network-aware collectors, including network condition-aware, self-organizing, multi-sensor data collectors that can optimize based on bandwidth, quality of service, pricing, and / or other network conditions.

[0018] Disclosed herein are methods and systems for remotely organizing a universal data collector that can power on and off sensor interfaces based on needs and / or conditions identified in an industrial data collection environment.

[0019] Disclosed herein are methods and systems for self-organizing storage of a multi-sensor data collector, including self-organizing storage of a multi-sensor data collector for industrial sensor data.

[0020] Disclosed herein are methods and systems for self-organizing network coding of a multi-sensor data network, including self-organizing network coding for a data network that transmits data from multiple sensors in an industrial data collection environment.

[0021] Disclosed herein are methods and systems for tactile or multisensory user interfaces, including wearable tactile or multisensory user interfaces for industrial sensor data collectors with vibration, thermal, electrical, and / or acoustic output.

[0022] Disclosed herein are methods and systems for an AR / VR industrial eyewear presentation layer, wherein heat map elements are presented based on patterns and / or parameters in collected data.

[0023] Disclosed herein are methods and systems for condition-sensitive, self-organizing adjustments to AR / VR interfaces based on feedback metrics and / or training in industrial environments.

[0024] In one embodiment, a system for collecting, processing, and utilizing data from signals from at least a first element in a first machine in an industrial environment includes a platform having a computing environment connected to a local data collection system, the local data collection system having first and second sensor signals obtained from at least the first machine in the industrial environment. The system includes a first sensor located in the local data collection system and a second sensor located in the local data collection system, the local data collection system being configured to connect to the first machine. The system also includes a crosspoint switch located in the local data collection system, the crosspoint switch having multiple inputs and multiple outputs, including a first input connected to the first sensor and a second input connected to the second sensor. The multiple outputs include a first output and a second output, each configured to switch between transmitting the first sensor signal and transmitting the second sensor signal, and transmitting the first sensor signal from the first output and the second sensor signal simultaneously from the second output. Each of the multiple inputs is configured to be individually assigned to any one of the multiple outputs. Unassigned outputs are configured to be turned off, thereby creating a high-impedance state.

[0025] In an embodiment, the first sensor signal and the second sensor signal are continuous vibration data associated with an industrial environment. In an embodiment, the second sensor in the local data collection system is configured to be connected to a first machine. In an embodiment, the second sensor in the local data collection system is configured to be connected to a second machine in the industrial environment. In an embodiment, the computing environment of the platform is configured to compare the relative phases of the first and second sensor signals. In an embodiment, the first sensor is a single-axis sensor and the second sensor is a three-axis sensor. In an embodiment, at least one of the multiple inputs of the crosspoint switch includes internal protocol front-end signal conditioning for improving signal-to-noise ratio. In an embodiment, the crosspoint switch includes a third input configured with a continuously monitored alarm having a predetermined trigger condition when the third input is not assigned to any of the multiple outputs.

[0026] In one embodiment, a local data collection system includes multiple multiplexing units and multiple data collection units that receive multiple data streams from multiple machines in an industrial environment. In one embodiment, the local data collection system includes distributed complex programmable hardware device (CPLD) chips, each chip dedicated to a data bus that logically controls the multiple multiplexing units and the multiple data collection units that receive multiple data streams from multiple machines in the industrial environment. In one embodiment, the local data collection system is configured to provide high current input capability using solid-state relays. In one embodiment, the local data collection system is configured to power off at least one of an analog sensor channel and a component board.

[0027] In one embodiment, the local data collection system includes an external voltage reference for an A / D zero reference that is independent of the voltages of the first and second sensors. In one embodiment, the local data collection system includes a phase-locked loop (PLL) bandpass tracking filter to acquire slow RPM and phase information. In one embodiment, the local data collection system is configured to digitally derive phase relative to at least one trigger channel and at least one input using an onboard timer. In one embodiment, the local data collection system includes a peak detector configured to perform peak detection using automatic scaling using a separate analog-to-digital converter. In one embodiment, the local data collection system is configured to route data from at least one of the raw and buffered trigger channels to at least one of the plurality of inputs. In one embodiment, the local data collection system includes at least one oversampling ADC configured to increase the input oversampling rate to reduce the output sampling rate and minimize anti-aliasing filtering requirements. In an embodiment, each distributed CPLD chip dedicated to a data bus for logically controlling a plurality of multiplexing units and a plurality of data acquisition units includes a high frequency crystal clock reference configured to be down-clocked by at least one distributed CPLD chip of at least one oversampling analog-to-digital converter to achieve a lower sampling rate without digital resampling.

[0028] In an embodiment, the local data collection system is configured to acquire long data blocks at a single relatively high sampling rate, rather than multiple data sets extracted at different sampling rates. In an embodiment, the single relatively high sampling rate corresponds to a maximum frequency of approximately 40 kHz. In an embodiment, the long data blocks have a duration exceeding 1 minute. In an embodiment, the local data collection system includes a plurality of data collection units, each data collection unit having an onboard card pack configured to store calibration information and maintenance history for the data collection unit in which the onboard card pack is located. In an embodiment, the local data collection system is configured to plan data collection paths based on a hierarchical template.

[0029] In one embodiment, the local data collection system is configured to manage data collection bands. In one embodiment, the data collection bands define specific frequency bands and at least one of a set of spectral peaks, true peak levels, crest factors derived from a time waveform, and complete waveforms derived from a vibration envelope. In one embodiment, the local data collection system includes a neural network expert system that utilizes intelligent management of the data collection bands. In one embodiment, the local data collection system is configured to create data collection paths based on hierarchical templates, each hierarchical template including a data collection band associated with a machine associated with the data collection path. In one embodiment, at least one of the hierarchical templates is associated with a plurality of interconnected components of a first machine. In one embodiment, at least one of the hierarchical templates is associated with similar components associated with at least the first machine and a second machine. In one embodiment, at least one of the hierarchical templates is associated with at least the first machine that is located adjacent to the second machine.

[0030] In an embodiment, the local data collection system includes a graphical user interface system configured to manage the data collection belt. In an embodiment, the graphical user interface system includes an expert system diagnostic tool. In an embodiment, the platform includes cloud-based machine pattern analysis of state information from multiple sensors to provide expected state information of the industrial environment. In an embodiment, the platform is configured to provide self-organization of the data pool based on at least one of utilization metrics and revenue metrics. In an embodiment, the platform includes a self-organizing group of industrial data collectors. In an embodiment, the local data collection system includes a wearable tactile user interface for the industrial sensor data collector, the user interface having at least one of vibration, thermal, electrical, and sound output.

[0031] In an embodiment, the multiple inputs of the crosspoint switch include a third input connected to a second sensor and a fourth input connected to the second sensor. The first sensor signal is from a single-axis sensor located at a constant position associated with the first machine. In an embodiment, the second sensor is a three-axis sensor. In an embodiment, the local data collection system is configured to simultaneously record gapless digital waveform data from at least the first input, the second input, the third input, and the fourth input. In an embodiment, the platform is configured to determine a change in relative phase based on the simultaneously recorded gapless digital waveform data. In an embodiment, the second sensor is configured to be movable to multiple positions associated with the first machine while acquiring the simultaneously recorded gapless digital waveform data. In an embodiment, the multiple outputs of the crosspoint switch include a third output and a fourth output. The second, third, and fourth outputs are collectively distributed to a series of three-axis sensors, each located at a different position associated with the machine. In an embodiment, the platform is configured to determine an operational deformation mode based on the change in relative phase and the simultaneously recorded gapless digital waveform data.

[0032] In an embodiment, the constant position is a position associated with a rotating shaft of the first machine. In an embodiment, the three-axis sensors in the series of three-axis sensors are respectively located at different locations on the first machine, but are respectively associated with different bearings in the machine. In an embodiment, the three-axis sensors in the series of three-axis sensors are respectively located at similar locations associated with similar bearings, but are respectively associated with different machines. In an embodiment, the local data collection system is configured to acquire simultaneously recorded gapless digital waveform data from the first machine while both the first machine and the second machine are operating. In an embodiment, the local data collection system is configured to characterize the contributions from the first machine and the second machine to the gapless digital waveform data simultaneously recorded for the first machine. In an embodiment, the simultaneously recorded gapless digital waveform data has a duration of more than 1 minute.

[0033] In an embodiment, a method for monitoring a machine having at least one shaft supported by a set of bearings includes monitoring a first data channel assigned to a single-axis sensor located at a constant position associated with the machine. The method also includes monitoring second, third, and fourth data channels assigned to the axes of the three-axis sensor, respectively. The method also includes simultaneously recording gapless digital waveform data from all data channels while the machine is operating, and determining a change in relative phase based on the digital waveform data.

[0034] In an embodiment, a triaxial sensor is placed at a plurality of locations associated with the machine while acquiring a digital waveform. In an embodiment, the second, third, and fourth channels are assigned together to a series of triaxial sensors located at different locations associated with the machine. In an embodiment, data is received from all sensors simultaneously. In an embodiment, the method includes determining an operating deformation mode based on relative phase change information and waveform data. In an embodiment, the constant position refers to a position associated with a shaft of the machine. In an embodiment, the triaxial sensors in the series of triaxial sensors are located at different locations and are associated with different bearings in the machine. In an embodiment, the constant position refers to a position associated with a shaft of the machine. The triaxial sensors in the series of triaxial sensors are located at different locations and are associated with different bearings supporting the shaft in the machine.

[0035] In one embodiment, the method includes monitoring a first data channel assigned to a single-axis sensor located at a constant position on a second machine. The method includes monitoring second, third, and fourth data channels, respectively assigned to axes of a three-axis sensor located at a position associated with the second machine. The method also includes simultaneously recording gapless digital waveform data for all data channels from the second machine while both machines are operating. In one embodiment, the method includes characterizing the contribution from each machine to the gapless digital waveform data simultaneously from the second machine.

[0036] In an embodiment, a method for data collection, processing, and utilization of signals by a platform monitoring at least a first element of a first machine in an industrial environment includes automatically acquiring at least a first sensor signal and a second sensor signal using a computing environment by monitoring a local data collection system of at least the first machine. The method includes connecting a first input of a crosspoint switch of the local data collection system to the first sensor and connecting a second input of the crosspoint switch to the second sensor in the local data collection system. The method includes switching between the following two situations: a situation where a first output of the crosspoint switch alternates between transmitting at least the first sensor signal and the second sensor signal, and a situation where the first sensor signal is transmitted from the first output of the crosspoint switch and the second sensor signal is transmitted from the second output of the crosspoint switch at the same time. The method also includes shutting down an unassigned output of the crosspoint switch to a high impedance state.

[0037] In an embodiment, the first sensor signal and the second sensor signal are continuous vibration data from an industrial environment. In an embodiment, the second sensor in the local data collection system is connected to the first machine. In an embodiment, the second sensor in the local data collection system is connected to a second machine in the industrial environment. In an embodiment, the method includes automatically comparing the relative phases of the first and second sensor signals using a computing environment. In an embodiment, the first sensor is a single-axis sensor and the second sensor is a three-axis sensor. In an embodiment, at least a first input of the crosspoint switch includes internal protocol front-end signal conditioning to improve signal-to-noise ratio.

[0038] In an embodiment, the method includes continuously monitoring at least one third input of the crosspoint switch with an alarm, the alarm having a predetermined trigger condition when the third input is not assigned to any output on the crosspoint switch. In an embodiment, the local data collection system includes a plurality of multiplexing units and a plurality of data acquisition units, the data acquisition units receiving a plurality of data streams from a plurality of machines in the industrial environment. In an embodiment, the local data collection system includes a distributed complex programmable hardware device (CPLD) chip, each chip being dedicated to a data bus for logic control of the plurality of multiplexing units and the plurality of data acquisition units receiving the plurality of data streams from the plurality of machines in the industrial environment. In an embodiment, the local data collection system uses solid-state relays to provide high current input capability.

[0039] In one embodiment, the method includes powering off at least one of an analog sensor channel and a component board of a local data collection system. In one embodiment, the local data collection system includes an external voltage reference for an A / D zero reference that is independent of the voltage of the first sensor and the second sensor. In one embodiment, the local data collection system includes a phase-locked loop bandpass tracking filter that captures slow RPMs and phase information. In one embodiment, the method includes digitally deriving phase relative to at least one trigger channel and at least one of a plurality of inputs on a crosspoint switch using an onboard timer.

[0040] In one embodiment, the method includes using a separate analog-to-digital converter to perform peak detection through automatic scaling with a peak detector. In one embodiment, the method includes routing at least one of the original and buffered trigger channels to at least one of the plurality of inputs on a crosspoint switch. In one embodiment, the method includes increasing the input oversampling rate using at least one oversampling analog-to-digital converter to reduce the output sampling rate and minimize the need for anti-aliasing filtering. In one embodiment, each distributed CPLD chip dedicated to a data bus for logic control of the plurality of multiplexing units and the plurality of data acquisition units includes a high-frequency crystal clock reference that is downsampled by at least one of the distributed CPLD chips for the at least one oversampling analog-to-digital converter to achieve a lower sampling rate without requiring digital resampling. In one embodiment, the method includes acquiring long data blocks at a single relatively high sampling rate, rather than multiple data sets extracted at different sampling rates. In one embodiment, the single relatively high sampling rate corresponds to a maximum frequency of approximately 40 kHz. In one embodiment, the long data blocks have a duration exceeding one minute. In an embodiment, the local data collection system includes a plurality of data collection units, each of which has an onboard card group that stores calibration information and maintenance history of the data collection unit in which the onboard card group is located.

[0041] In one embodiment, the method includes planning a data collection path based on a hierarchical template associated with at least a first element of a first machine in an industrial environment. In one embodiment, a local data collection system manages data collection bands that define specific frequency bands and at least one of a set of spectral peaks, true peak levels, a crest factor derived from a time waveform, and a complete waveform derived from a vibration envelope. In one embodiment, the local data collection system includes a neural network expert system that utilizes intelligent management of the data collection bands. In one embodiment, the local data collection system creates data collection paths based on hierarchical templates, each hierarchical template including a data collection band associated with a machine associated with the data collection path. In one embodiment, at least one of the hierarchical templates is associated with a plurality of interconnected elements of the first machine. In one embodiment, at least one of the hierarchical templates is associated with similar elements associated with at least the first machine and a second machine. In one embodiment, at least one of the hierarchical templates is associated with at least the first machine that is located adjacent to the second machine.

[0042] In an embodiment, the method includes a graphical user interface system for controlling a local data collection system to manage the data collection bands. The graphical user interface system includes an expert system diagnostic tool. In an embodiment, the computing environment of the platform includes cloud-based machine pattern analysis of state information from a plurality of sensors to provide expected state information of the industrial environment. In an embodiment, the computing environment of the platform provides self-organization of the data pool based on at least one of a utilization metric and a revenue metric. In an embodiment, the computing environment of the platform includes a self-organizing cluster of industrial data collectors. In an embodiment, each of the plurality of inputs of the crosspoint switch can be individually assigned to any one of the plurality of outputs of the crosspoint switch. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figures 1 to 5 are schematic diagrams respectively showing multiple parts of an overall view of an industrial IoT data collection, monitoring and control system according to the present invention;

[0044] Figure 6 is a schematic diagram of a platform including a local data collection system disposed in an industrial environment according to the present invention, the platform for collecting data from or relating to environmental elements such as machines, components, systems, subsystems, environmental conditions, states, workflows, processes, and other elements;

[0045] Figure 7 is a schematic diagram illustrating elements of an industrial data collection system for collecting analog sensor data in an industrial environment according to the present invention;

[0046] Figure 8 is a schematic diagram of a rotating or vibrating machine having a data acquisition module configured to collect waveform data according to the present invention;

[0047] Figure 9 is a schematic diagram of an exemplary three-axis sensor mounted to a motor bearing of an exemplary rotating machine according to the present invention;

[0048] Figure 10 and Figure 11 is a schematic diagram of an exemplary three-axis sensor and a single-axis sensor mounted to an exemplary rotating machine according to the present invention;

[0049] Figure 12 is a schematic diagram of a plurality of surveyed machines having a sensor set according to the present invention;

[0050] Figure 13 is a schematic diagram of a hybrid approach of relational metadata and binary storage according to the present invention;

[0051] Figure 14is a schematic diagram of components and interactions of a data collection architecture involving application of cognitive and machine learning systems to data collection and processing according to the present invention;

[0052] Figure 15 is a schematic diagram of components and interactions of a data collection architecture involving application of a platform having a cognitive data marketplace according to the present invention;

[0053] Figure 16 is a schematic diagram of components and interactions of a data collection architecture involving a self-organizing group of application data collectors according to the present invention;

[0054] Figure 17 is a schematic diagram of components and interactions of a data collection architecture involving application of a tactile user interface according to the present invention. DETAILED DESCRIPTION

[0055] Detailed embodiments of the present invention are disclosed herein; however, it should be understood that the disclosed embodiments are merely exemplary of the present invention, which can be embodied in a variety of forms. Therefore, the specific structural and functional details disclosed herein should not be interpreted as limiting, but merely as a basis for the claims and as a representative basis for guiding those skilled in the art to utilize the present invention in virtually any suitable specific configuration.

[0056] The terms "a" and "an," as used herein, are defined as one or more. The term "another," as used herein, is defined as at least a second or more. The terms "including" and / or "having," as used herein, are defined as comprising (i.e., open ended meanings).

[0057] Although only a few embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that many changes and modifications can be made to the present invention without departing from the spirit and scope of the invention as described in the following claims. All foreign and domestic patent applications and patents and all other publications cited herein are hereby incorporated in their entirety to the extent permitted by law.

[0058] Figures 1 to 5 Various parts of an overall view of an Industrial IoT data collection, monitoring and control system 10 are shown. Figure 2 Show Figure 1-5 The upper left portion of the schematic diagram of the industrial IoT system 10 is shown. Figure 2The diagram includes a mobile ad hoc network (MANET) 20 that can form secure, temporary network connections 22 (sometimes connected, sometimes isolated) with a cloud 30 or other remote networking system, allowing network functions to be performed within the MANET 20 within the environment without the need for an external network, but at other times, information can be sent and received from a central location. This allows industrial environments to take advantage of networking and control technologies while also providing security, such as protection against cyberattacks. The MANET 20 can use cognitive radio technology 40, including technologies that constitute the equivalent of the IP protocol, such as routers 42, MAC 44, and physical layer technologies 46. The diagram also shows network-sensitive or network-aware data transmission over the network to and from data collection devices or heavy industrial machinery.

[0059] Figure 3 Show Figure 1-5 The upper right portion of a schematic diagram of an industrial IoT system 10 is shown. This includes 50 intelligent data collection technologies deployed locally at the edge of the IoT deployment, where heavy industrial machinery is located. This includes various sensors 52, IoT devices 54, data storage capabilities 56 (including intelligent, self-organizing memory), sensor fusion (including self-organizing sensor fusion), and more. Figure 3 , interfaces for data collection are shown, including a multi-sensory interface, a tablet computer, a smart phone 58, etc. Figure 3 Also shown is a data pool 60 that can collect data published by machines or sensors detecting machine conditions, such as data for later consumption by local or remote intelligence. A distributed ledger system 62 can distribute storage to local storage of various environmental elements or more broadly throughout the system.

[0060] Figure 1 Show Figure 1-5 The middle part of the schematic diagram of the industrial IoT system shown in Figure 1. This includes the use of network coding (including self-organizing network coding and / or automated configuration), which can configure the network coding model based on feedback measurements, network conditions, etc., for efficient transmission of large amounts of data between the data collection system and the cloud across the network. Various intelligence, analysis, remote control, remote operation, remote optimization and other functions can be deployed in the cloud or on the premises of the enterprise owner or operator, including Figure 1 This includes various storage configurations, which may include distributed ledger storage for supporting system transaction data or other elements.

[0061] Figure 1 、 Figure 4 and Figure 5 Show Figure 1-5This includes a programmatic data marketplace 70, which can be a self-organizing marketplace, such as for providing data collected in an industrial environment, such as data collectors, data pools, distributed ledgers, and the like disclosed herein. Figure 1-5 Other components shown in . Figure 1 , Figure 4 and Figure 5 Also shown are on-device sensors 80, such as for storing device data from multiple simulated sensors 82 on the device, which can be analyzed locally or in the cloud by machine learning 84, including by training the machine based on an initial model created by humans and enhanced by providing feedback (e.g., based on successful measurements) while operating the methods and systems disclosed herein. Figures 1 to 5 Additional details on the various components and subcomponents of the .

[0062] In an embodiment, a method and system for a data collection, processing, and utilization system (referred to herein as platform 100) in an industrial environment is provided. Figure 6 The platform 100 may include a local data collection system 102 that may be located in an environment 104, such as an industrial environment, for collecting data from or related to elements of the environment, such as machines, components, systems, subsystems, environmental conditions, states, workflows, processes, and other elements. The platform 100 may be connected to or include Figure 1-51 . The platform 100 may include portions of an industrial IoT data collection, monitoring, and control system 10 shown in FIG. The platform 100 may include a network data transmission system 108, such as for transmitting data to and from a local data collection system 102 via a network 110, such as for transmitting data to a host processing system 112, e.g., a system located in a cloud computing environment or on an enterprise premises or comprised of distributed components that interact with each other to process data collected by the local data collection system 102. The host processing system 112, in some cases simply referred to as the host system 112, may include various systems, components, methods, processes, facilities, etc., for implementing automated or automation-assisted data processing, such as for monitoring one or more environments 104 or the network 110 or for remotely controlling one or more elements in the local environment 104 or in the network 110. The platform 100 may include one or more local autonomous systems 114, such as for implementing autonomous behavior, such as reflecting human or machine-based intelligence or for implementing automatic operations based on applying a set of rules or models to input data from the local data collection system 102 or from one or more input sources 116, which may include information feedback and input from various sources, including information feedback and input from the local environment 104, the network 110, the host system 112, or one or more external systems, databases, etc. The platform 100 may include one or more intelligent systems 118, which may be located in, integrated with, or serve as input to one or more components of the platform 100. Details of these and other components of the platform 100 are provided herein.

[0063] Intelligent systems may include cognitive systems 120, such as those that achieve a degree of cognitive behavior due to coordinated grid, peer, ring, serial, and other architectures of processing elements, where one or more node elements coordinate with each other to provide unified and coordinated behavior to assist in processing, communication, data collection, and the like. Figure 2The illustrated MANET 20 may also utilize cognitive radio technology, including technologies that constitute the equivalent of the IP protocol, such as routers 42, MACs 44, and physical layer technologies 46. In one example, the cognitive system technology stack may include several examples disclosed in U.S. Patent No. 8,060,017, issued to Schlicht et al. on November 15, 2011, and incorporated herein by reference as if fully set forth herein. The intelligent system may include a machine learning system 122, such as for learning from one or more data sets, such as information collected using local data collection system 102 or other information from input sources 116, to identify states, objects, events, patterns, conditions, etc., which are then processed by host system 112 as input to various components of platform 100 or industrial IoT data collection, monitoring, and control system 10. Learning may be manually supervised or fully automated, such as using one or more input sources 116 to provide data sets and information about the items to be learned. Machine learning can use one or more models, rules, semantic understanding, workflows, or other structured or semi-structured understandings of the world to automatically optimize the control of a system or process based on feedback or feedforward to a working model of the system or process. One such machine learning technique for semantic and scenario understanding, workflows, or other structured or semi-structured understanding is disclosed in U.S. Patent No. 8,200,775, issued to Moore on June 12, 2012. Machine learning can be used to improve the aforementioned techniques, for example, by adjusting one or more weights, structures, rules, etc. (e.g., modifying functions within a model) based on feedback (e.g., regarding the success of the model in a given situation) or based on iteration (e.g., in a recursive process). Machine learning can also be implemented in the absence of an underlying model if a full understanding of the underlying structure or behavior of the system is not yet known, insufficient data is available, or other solutions are preferred for various reasons; that is, input sources can be weighted, structured, etc. within the machine learning facility without any prior understanding of the structure, and the results (e.g., based on measurements of success in achieving various desired goals) can be continuously provided to the machine learning system, enabling it to learn how to achieve the designated goals. For example, the system can learn to identify failures, recognize patterns, develop models or functions, formulate rules, optimize performance, minimize failure rates, optimize revenue, optimize resource utilization, optimize processes (e.g., traffic flow), or optimize multiple other parameters that may be correlated with successful outcomes, such as outcomes in various environments. Machine learning can use genetic programming techniques, for example, to promote or demote one or more input sources, structures, data types, objects, weights, nodes, links, or other factors based on feedback so that successful elements emerge over a series of generations.For example, alternative available sensor inputs to the data collection system 102 may be arranged in alternative configurations and permutations, such that the system can use genetic programming techniques on a series of data collection events to determine a permutation that provides a successful outcome based on various conditions (e.g., conditions of platform 100 components, conditions of network 110, conditions of the data collection system 102, conditions of the environment 104), etc. In an embodiment, local machine learning can enable or disable one or more sensors in the multi-sensor data collector 102 in the permutation over time while tracking successful outcomes, such as contribution to successful prediction of failures, contribution to performance metrics (e.g., efficiency, effectiveness, return on investment, profitability), contribution to optimization of one or more parameters, identification of patterns (e.g., related to threats, failure modes, success patterns, etc.), etc. For example, the system can learn which groups of sensors should be enabled or disabled under given conditions to achieve the highest utilization of the data collector 102. In an embodiment, similar techniques can be used to handle data transmission optimization in the platform 100, such as in the network 110, such as by using genetic programming or other machine learning techniques to learn to configure network elements, such as configuring network transmission paths, configuring network coding types and architectures, configuring network security elements, etc.

[0064] In an embodiment, the local data collection system 102 may include a high-performance multi-sensor data collector with several novel features for collecting and processing analog and other sensor data. In an embodiment, the local data collection system 102 may be deployed to Figure 3 The local data collection system 102 can also be deployed to monitor other machines, such as Figure 9 and Figure 10 The machine 2300 shown in FIG. Figure 12 The machines 2400, 2600, 2800, 2950, ​​3000 and Figure 13 3204 are shown in the figures. The data collection system 102 may have onboard intelligence (e.g., learning to optimize the configuration and operation of the data collector, such as configuring the arrangement and combination of sensors based on scenarios and conditions). In one example, the data collection system 102 includes a crosspoint switch 130. The automated intelligent configuration of the local data collection system 102 may be based on various types of information, such as from various input sources, such as information based on available power, power requirements of sensors, the value of the collected data (e.g., based on feedback from other elements of the platform 100), the relative value of the information (e.g., based on the availability of other sources of the same or similar information), power availability (e.g., for powering sensors), network conditions, environmental conditions, operating status, operating scenarios, operating events, and others.

[0065] Figure 71 shows elements and subcomponents of a data collection and analysis system 1100 for sensor data (such as simulated sensor data) collected in an industrial environment. Figure 7 As shown, embodiments of the methods and systems disclosed herein may include hardware having several different modules, starting with a multiplexer (Mux) 1104. In an embodiment, the Mux 1104 is composed of a main board and an option board 1108. The main board is where the sensors connect to the system. These connections are located at the top for ease of installation. There are then many settings on the underside of this board and on the option board for the Mux, which connects to the main board via two connectors at each end of the board. In an embodiment, the option board for the Mux has a male connector that engages with the female connector on the main Mux board. This allows them to be stacked on top of each other, taking up less space.

[0066] In one embodiment, the master mux board is then connected to the mother analog board (e.g., via four simultaneous channels) and the daughter analog board 1110 (e.g., via four additional channels, for a total of eight channels) via cables, where some signal conditioning (e.g., hardware integration) occurs. These signals then move from the analog board 1110 to the anti-aliasing board, which removes some potential aliasing. The remainder of the aliasing is performed on the oversampling board 1112, which is connected via cables. The oversampling board 1112 provides further aliasing protection and other conditioning and digitization operations on the signals. Next, the data moves to the Jennic board 1114 for further digitization and transfer to a computer via USB or Ethernet. In one embodiment, the Jennic board 1114 can be replaced with a PIC board 1118 for more advanced and efficient data collection and communication. Once the data moves to the computer software 1102, the computer software 1102 can manipulate and process the data to display trends, spectra, waveforms, statistics, and analysis.

[0067] In embodiments, the system is designed to accept all types of data, from several volts to 4-20mA signals. In embodiments, open data storage and communication formats can be used. Analysis and reporting will include certain proprietary aspects of the system. In embodiments, smartband analysis is a way to break down data into easily analyzable components that can be combined with other smartbands to generate new, simpler, yet more comprehensive analyses. In embodiments, this unique information is extracted and graphically illustrated, as pictorial explanations are more helpful to users. In embodiments, complex programs and user interfaces are simplified to enable any user to manipulate data like an expert.

[0068] In embodiments, the system essentially operates in a single, large loop. It begins with a generic user interface in software. Most, but not all, online systems require the OEM to create or develop the GUI system 1124. In embodiments, fast-path creation can utilize hierarchical templates. In embodiments, a graphical user interface (GUI) is created so that any user can fill in information using a simple template. After creating the template, the user can copy and paste any required information. Furthermore, users can develop their own templates for future use and to institutionalize their knowledge. Once the user has entered all user information and connected all of their sensors, the user can start the system to acquire data. In some applications, rotating machinery can generate electrical charges that can harm electrical equipment. In embodiments, to mitigate the effects of this charge on equipment, unique electrostatic protection for trigger and vibration inputs is placed before the Mux and DAQ hardware to dissipate the charge as the signal passes from the sensor to the hardware. In embodiments, the Mux and analog boards can also utilize solid-state relays and design topologies to provide front-end circuitry and wider traces for high-current input capability, enabling the system to handle high-current inputs when necessary.

[0069] In this example, the important part before the mux is the front-end signal conditioning on the mux, which is used to improve the signal-to-noise ratio of the front-end signal conditioning. Most multiplexers are added later, and OEMs often don't pay attention to or even consider the signal quality coming from the mux. As a result, signal quality can degrade by as much as 30dB or more. Every system is only as strong as its weakest link, so whether the signal-to-noise ratio of a 24-bit DAQ is 110dB, signal quality is compromised by the mux. If the signal-to-noise ratio in the mux drops to 80dB, it's likely not much better than a 16-bit system from 20 years ago.

[0070] In addition to providing a better signal, multiplexers can also play a key role in enhancing the system. A truly continuous system can monitor every sensor at all times, but this is often costly. Multiplexer systems typically monitor only a certain number of channels at a time, switching between them individually within a larger set of sensors. This means that sensors not being collected are not monitored; if levels rise, the user may never know. In embodiments, a multiplexer continuous monitoring alarm feature provides continuous monitoring alarms for the multiplexer by providing circuitry on the multiplexer that measures input levels against known alarms even when the data acquisition ("DAQ") system is not monitoring the inputs. Essentially, this enables the system to monitor continuously, but it cannot capture data about issues as quickly as a true continuous system. In embodiments, this alarm feature is combined with adaptive scheduling technology for continuous monitoring, and the continuous monitoring system software adapts and adjusts the data collection sequence based on statistics, analytics, data alerts, and dynamic analysis. This allows the system to quickly collect dynamic spectrum data about alarming sensors immediately after an alarm is issued.

[0071] Another limitation of multiplexers is that they typically have a limited number of channels. In embodiments, the use of distributed complex programmable logic device (CPLD) chips, along with a dedicated bus for logic control of multiple muxes and data acquisition sections, enables the CPLD to control multiple muxes and DAQs, thereby removing the limit on the number of channels a system can handle. In embodiments, multiplexers and DAQs can be stacked together to provide additional input and output channels to the system.

[0072] In addition to a limited number of channels, multiplexers typically only collect data from sensors within the same group. This is a significant limitation for detailed analysis, as being able to simultaneously examine data from sensors on the same machine is extremely valuable. In an embodiment, this problem is addressed by using analog crosspoint switches to collect data from variable groups of vibration input channels. Crosspoint switches are commonly used in the telephone industry and provide a matrix circuit that enables the system to access any group of eight channels from the total number of input sensors.

[0073] In an embodiment, the system provides a phase-locked loop bandpass tracking filter method for acquiring slow RPMs and phase for balancing, thereby remotely balancing slow machinery at a site such as a paper mill and providing additional analysis based on its data.

[0074] In one embodiment, a distributed CPLD chip with a dedicated bus is used to control multiple multiplexers, providing logical control of multiple muxes and data acquisition sections. This capability can be enhanced by utilizing a hierarchical multiplexer, which allows multiple DAQs to collect data from multiple multiplexers. In one embodiment, this allows for faster data collection and more channels to be collected simultaneously for more complex analysis. In one embodiment, the mux can be modified to be portable and utilize the data acquisition residency feature, which can make the SV3X DAQ a protected system.

[0075] In an embodiment, once the signals leave the multiplexers and layered muxes, they are moved to analog boards with other enhancements. In an embodiment, the ability to power down analog channels when not in use, along with other power-saving measures including the ability to power down component boards, enables the system to power down channels on the main and daughter analog boards to save energy. In an embodiment, this can provide the same energy-saving benefits to protection systems, especially when operating on batteries or solar power. In an embodiment, to maximize signal-to-noise ratio and provide the best data, a peak detector routed to a separate A / D and used for automatic scaling provides the system with the highest peak in each set of data so it can quickly scale the data to that peak. In an embodiment, improved integration using both analog and digital methods results in innovative hybrid integration that will also improve or maintain the highest possible signal-to-noise ratio.

[0076] In an embodiment, a segment of the analog board allows the trigger channel (raw or buffered) to be routed to other analog channels. This allows the user to route the trigger to any channel for analysis and troubleshooting. In an embodiment, once the signals leave the analog board, they enter the oversampling board where a precise voltage reference for the A / D zero reference provides more accurate DC sensor data. The high speed of oversampling also allows the higher inputs of the oversampled A / D to be oversampled for the lower sample rate output to minimize anti-aliasing filter requirements, thereby oversampling the data at the higher inputs, which will minimize anti-aliasing requirements. In an embodiment, a CPLD can be used as a clock divider for the oversampled A / D to achieve lower sample rates without digital resampling, allowing the oversampled A / D to achieve lower sample rates without digitally resampling the data.

[0077] In embodiments, the data is then moved from the oversampling board to the Jennic board, where the phase relative to the input and trigger channels is digitally derived using onboard timers. In embodiments, the Jennic board can also store calibration data and system maintenance and repair history data in the onboard card deck. In embodiments, the Jennic board will enable the acquisition of long blocks of data at a high sampling rate, rather than multiple data sets extracted at different sampling rates, making it possible to stream data and acquire long blocks of data for later advanced analysis.

[0078] In one embodiment, after data passes through the Jennic board, it is transferred to a computer. Once installed on the computer, the software features numerous enhancements that improve system analysis capabilities. In one embodiment, hierarchical templates are used for rapid route creation, providing the ability to quickly create routes for all devices using simple templates, which also accelerates software deployment. In one embodiment, the software will be used to increase system intelligence. This will begin with defining the expert system's Smart Bands and diagnostics expert system graphical GUI approach. This will provide a graphical expert system with a simplified user interface, enabling anyone to develop complex analyses. In one embodiment, this user interface will be centered around the Smart Bands, providing a simplified path for average users to implement complex yet flexible analyses. In one embodiment, the Smart Bands will be paired with self-learning neural networks to enable even more advanced analytical methods. In one embodiment, the system will also utilize a hierarchical structure for additional analytical insights from the machine. A key component of predictive maintenance is the ability to learn from known information during maintenance or inspection. In one embodiment, a graphical back-calculation approach can improve the Smart Bands and correlations based on known faults or issues.

[0079] In embodiments, in addition to the detailed analysis performed by smart belts, bearing analysis methods are also provided. In recent years, there has been a strong drive for energy conservation, which has led to an influx of variable frequency drives. In embodiments, torsional vibration detection and analysis using transient signal analysis provides advanced torsional vibration analysis, enabling more comprehensive diagnosis of torsional-related machinery (such as machinery with rotating components). In embodiments, the system is capable of self-deploying multiple intelligent functions to achieve better data analysis and more comprehensive analysis. In embodiments, this intelligence begins with intelligent routing, where the software's intelligent routing simultaneously adjusts the sensor data it collects to gain additional correlation intelligence. In embodiments, intelligent operational data storage ("ODS") allows the system to selectively collect deformation modal analysis to further examine the condition of the machinery. In embodiments, in addition to changing routing, adaptive scheduling technology for continuous monitoring allows the system to change the scheduled data collected for full-spectrum analysis across multiple (e.g., eight) relevant channels. In embodiments, this intelligence provides data for continuous monitoring and extended statistical capabilities for analysis of localized environmental vibration, incorporating ambient and local temperature, as well as changes in vibration levels, to identify machinery issues.

[0080] Embodiments of the methods and systems disclosed herein may include a self-contained DAQ box. In embodiments, a data acquisition device can be controlled by a PC to implement desired data acquisition commands. In embodiments, the system can be self-contained and capable of data acquisition, processing, analysis, and monitoring independent of external PC control. Embodiments of the methods and systems disclosed herein may include secure digital ("SD") card storage. In embodiments, utilizing an SD card provides significant additional storage capacity, such as for cameras, smartphones, and the like. This is crucial for monitoring applications where critical data must be permanently stored. Furthermore, in the event of a power failure, the latest data can be stored even if it is not offloaded to another system. Embodiments of the methods and systems disclosed herein may include a data acquisition ("DAQ") system. The current trend is to enable DAQ systems to communicate with the outside world as much as possible, often in the form of networks, including wireless networks. While in the past, dedicated buses were typically used to control DAQ systems with microprocessors or microcontrollers / microprocessors paired with PCs, the need for networking is now greater, and this has deviated from the environment in which this new design prototype was created. In embodiments, multiple microprocessors / microcontrollers or dedicated processors can be utilized to enhance various aspects of DAQ functionality, with one or more processor units primarily focused on communication with the outside world. This eliminates the need to constantly interrupt key processes, including control of signal conditioning circuitry, raw data acquisition using A / D triggers, directing A / D output to appropriate onboard memory, and processing this data. In embodiments, a dedicated microcontroller / microprocessor is designated for all external communications. These include USB, Ethernet, and wireless, with the ability to provide one or more IP addresses for hosting web pages. All communication with the outside world is accomplished using simple text-based menus. A series of commonly used commands (actually over 100) are provided, such as InitializeCard (card initialization), AcquireData (data acquisition), StopAcquisition (stop acquisition), RetrieveCalibrationInfo (retrieve calibration information), etc. In embodiments, intensive signal processing activities, including resampling, weighting, filtering, and spectral processing, can be performed by dedicated processors, including one or more field-programmable gate arrays (FPGAs), digital signal processors (DSPs), microprocessors, and microcontrollers. In one embodiment, this subsystem can communicate with the communication processing segment via a dedicated hardware bus. Dual-port memory, signal logic, and the like will facilitate this goal. This embodiment can significantly improve not only efficiency but also processing capabilities, including data streaming and other high-end analysis techniques.

[0081] Embodiments of the methods and systems disclosed herein may include sensor overload identification. A monitoring system is required to identify when a sensor is overloaded. A monitoring system can identify when its system is overloaded, but in some embodiments, the system can look at the voltage of a sensor to determine if the overload is coming from that sensor, which can help the user obtain another sensor more appropriate for the situation or attempt to recollect data. There are often situations involving high-frequency inputs that would saturate a standard 100mv / g sensor (the most commonly used in the industry), and the ability to sense the overload improves data quality for better analysis.

[0082] Embodiments of the methods and systems disclosed herein can include RFID on accelerometers and RFID on inclinometers or other sensors, whereby the sensors can tell the system / software which machine / bearing and orientation they are connected to, and can automatically configure themselves in the software to store data without the user having to tell them. In embodiments, a user can place the system on any machine or machines, and it will automatically configure itself and be ready to collect data within seconds.

[0083] Various embodiments of the methods and systems disclosed herein may include ultrasonic online monitoring by placing ultrasonic sensors inside transformers, MCCs, circuit breakers, and the like, wherein the system monitors the acoustic spectrum, continuously looking for patterns that identify arcing, corona, and other electrical issues that indicate a fault or problem. In embodiments, an analysis engine is used in ultrasonic online monitoring and identification of other faults by combining this data with other parameters such as vibration, temperature, pressure, heat flux, magnetic field, electric field, current, voltage, capacitance, inductance, and combinations thereof (e.g., single ratio).

[0084] Embodiments of the methods and systems disclosed herein may include the use of analog crosspoint switches for collecting variable sets of vibration input channels. For vibration analysis, it is beneficial to simultaneously acquire multiple channels from vibration transducers mounted in different orientations on different parts of a machine (or machines). By acquiring readings simultaneously, for example, the relative phases of these inputs can be compared to diagnose various mechanical faults. Additionally, other types of cross-channel analysis, such as cross-correlation, transfer functions, and operating deformation modes (ODS), can be performed. Current systems using conventional fixed-group multiplexers can only compare a limited number of channels assigned to a specific group at installation time (based on the number of channels in each group). The only way to provide some flexibility is to overlap channels or incorporate significant redundancy in the system, both of which can significantly increase overhead (in some cases, the cost increases exponentially with the flexibility). The simplest mux design selects one of multiple inputs and routes it to a single output line. A grouped design consists of a set of simple building blocks, each processing a fixed set of inputs and routing them to its corresponding output. Typically, these inputs do not overlap, so inputs from one mux group cannot be routed to another. Unlike conventional mux chips, which typically switch one or more fixed groups of channels to a single output—groups of 2, 4, 8, and so on—crosspoint muxes allow users to assign any input to any output. Previously, crosspoint multiplexers were used for specialized purposes, such as RGB digital video, and were too noisy for analog applications like vibration analysis; however, recent advances in the technology now make this possible. Another advantage of crosspoint muxes is the ability to disable outputs by placing them in a high-impedance state. This is ideal for output busses, allowing multiple mux cards to be stacked and their output buses connected together without the need for bus switches.

[0085] Embodiments of the methods and systems disclosed herein may include front-end signal conditioning on the Mux to improve the signal-to-noise ratio. Various embodiments may perform signal conditioning (e.g., range / gain control, integration, filtering, etc.) on vibration and other signal inputs before the Mux switches to achieve the highest signal-to-noise ratio.

[0086] Embodiments of the methods and systems disclosed herein may include a data acquisition dwell feature. In embodiments, continuous monitoring of the Mux bypass provides a mechanism to continuously monitor for significant alarms on channels not currently being sampled by the Mux system using a filtered peak hold circuit or similarly functioning multiple trigger conditions, thereby cost-effectively delivering these alarms to the monitoring system using hardware interrupts or other means.

[0087] Various embodiments of the methods and systems disclosed herein may include the use of distributed CPLD chips with a dedicated bus to perform logic control for multiple MUXs and data acquisition components. Interfacing with various types of predictive maintenance and vibration transducers requires extensive switching operations. This includes AC / DC coupling, 4-20 interfaces, IEPE (transducer power), channel power reduction (for switching op amp power), and single-ended or differential grounding options. Furthermore, digital port control is required for range and gain control, hardware-integrated switching, AA filtering, and triggering. This logic can be performed by a series of CPLD chips strategically positioned for the tasks being controlled. A single large CPLD requires long circuit paths, and these circuit paths are densely packed within the large chip. In embodiments, distributed CPLDs not only address these issues but also provide considerable flexibility. A bus is created in which each CPLD has its own unique device address with fixed assignments. For multiple boards, such as multiple MUX boards, jumpers are provided to set multiple addresses. In another example, three bits allow up to eight boards to be cross-configured. In embodiments, a bus protocol is defined so that each CPLD on the bus can be addressed individually or as a group.

[0088] Embodiments of the methods and systems disclosed herein may include high current input capability using solid state relays and design topologies. Typically, vibration data collectors are not designed to handle large input voltages because they are expensive and generally not needed. As technology improves and monitoring costs drop significantly, these data collectors will need to acquire many different types of PM data. In an embodiment, one approach is to use mature OptoMOS technology rather than the more conventional spring relay approach, which allows for switching of front-end high voltage signals. Many of the historical issues regarding nonlinear zero crossings or other nonlinear solid-state behavior have been eliminated with respect to the passage of weak buffered analog signals. Additionally, in an embodiment, the PCB is topologically arranged to place all single channel input circuitry as close to the input connector as possible.

[0089] Embodiments of the methods and systems disclosed herein include the ability to power down unused analog channels, power down component boards, and implement other power-saving measures. In embodiments, powering down analog signal processing op amps for unselected channels and powering down component boards and other hardware via low-level firmware in the DAQ system makes high-level application control relatively easy compared to power-saving capabilities. Explicit hardware control is possible but not required by default.

[0090] Embodiments of the methods and systems disclosed herein can include unique electrostatic protection for trigger and vibration inputs. In many critical industrial environments where significant electrostatic forces may be generated, such as low-speed balancing using large conveyors, appropriate transducer and trigger input protection is required. In embodiments, a low-cost yet effective method for such protection is described that eliminates the need for external supplemental equipment.

[0091] Embodiments of the methods and systems disclosed herein may include a precise voltage reference for the A / D zero reference. Some A / D chips provide their own internal zero voltage reference to serve as a mid-scale value for external signal conditioning circuitry, ensuring that the A / D and external op amp use the same reference. While this seems reasonable in theory, practical complications arise. In many cases, these references are themselves based on the supply voltage using resistor dividers. For many current systems, particularly those powered from a PC via USB or similar buses, this provides an unreliable reference, as the supply voltage often varies significantly with load. This is particularly true for oversampling A / D chips that require extensive signal processing. While the offset may drift with load, this can create problems if digital calibration of the reading is required. Typically, the voltage offset represented by the A / D counter is digitally modified to compensate for the DC offset. However, in this case, if the calibration offset is determined to be appropriate for one set of load conditions, it will not be applicable for other conditions. The absolute DC offset represented by the counter is no longer applicable. Calibration for all load conditions becomes complex, unreliable, and ultimately unmanageable. In an embodiment, an external voltage reference is used, which is independent only from the supply voltage, to serve as a zero offset.

[0092] Embodiments of the methods and systems disclosed herein may include a phase-locked loop (PLL) bandpass tracking filter approach for acquiring slow PRMs and phase for balancing purposes. For balancing purposes, balancing at very slow speeds is sometimes necessary. Typical tracking filters can be constructed based on a phase-locked loop (PLL) or PLL design. However, stability and speed range are paramount considerations. In an embodiment, multiple digitally controlled switches are used to select appropriate RC and damping constants. Switching can be fully automated after measuring the frequency of an incoming tachometer (tach) signal. Embodiments of the methods and systems disclosed herein may include using onboard timers to digitally derive phase relative to input and trigger channels. In an embodiment, digital phase derivation uses a digital timer to determine the precise delay from a trigger event to the precise start of data acquisition. This delay or offset is then further refined using interpolation to obtain a more precise offset, which is then applied to the phase determined by analysis of the acquired data so that the phase is "essentially" absolute phase and has a precise mechanical meaning useful for primary balancing, alignment analysis, and the like.

[0093] Embodiments of the methods and systems disclosed herein may include a peak detector for autoscaling, which is routed to a separate A / D converter. Many currently used microprocessors have built-in A / D converter functionality. For vibration analysis purposes, these often lack sufficient bit count, channel count, or sampling frequency without significantly reducing the microprocessor speed. Despite these limitations, using them for autoscaling can be beneficial. In embodiments, a separate, less expensive A / D converter with reduced functionality can be used. For each channel's input, after buffering the signal (typically using the appropriate coupling: AC or DC) but before signal conditioning, the signal is fed directly to the microprocessor or a low-cost A / D converter. Unlike adjusting the signal for range, gain, and filter switch positions, the switches are not changed. This allows for simultaneous sampling of autoscaling data while the input data is signal-conditioned, fed to a more powerful external A / D, and directed to onboard memory using direct memory access (DMA), which allows access to memory without the CPU. This significantly simplifies the autoscaling process without the need to toggle switch positions and allow for setup time, which significantly slows down the autoscaling process. Furthermore, the data are collected simultaneously, which ensures an optimal signal-to-noise ratio. The reduced number of bits and other characteristics are usually more than sufficient for automatic scaling purposes.

[0094] Embodiments of the methods and systems disclosed herein may include routing a trigger channel (raw or buffered) to other analog channels. Many systems may have trigger channels to determine the relative phase between multiple input data sets or to obtain important data without having to repeat undesired inputs. In embodiments, a digitally controlled relay is used to switch the raw or buffered trigger signal to one of these input channels. Often times, it is very useful to check the quality of the trigger pulse as it is often corrupted for a variety of reasons, including inadequate placement of the trigger sensor, wiring problems, incorrect setup issues (such as dirty reflective tape when using a light sensor), etc. The ability to view the raw or buffered signal provides an excellent diagnostic or debugging tool. It can also provide some improved phase analysis capabilities by using the recorded data signal for various signal processing techniques, such as variable speed filtering algorithms.

[0095] Embodiments of the methods and systems disclosed herein may include oversampling the higher input of an oversampled A / D for a lower sample rate output to minimize anti-aliasing filter requirements. In embodiments, a higher input oversampling rate is used for the oversampled A / D to obtain lower sample rate output data, thereby minimizing anti-aliasing filtering requirements. Lower oversampling rates can be used for higher sample rates. For example, a third-order AA filter with a minimum sample rate requirement of 256 Hz (Fmax of 100 Hz) is sufficient for an Fmax range of 200 to 500 Hz. Another higher-cutoff AA filter can be used for an Fmax range starting at 1 kHz and higher (where the auxiliary filter starts at the highest sample rate of 256 x 128 kHz). Embodiments of the methods and systems disclosed herein may include using a CPLD as a clock divider for the oversampled A / D to achieve lower sample rates without the need for digital resampling. In embodiments, a CPLD can be used as a programmable clock divider to downsample a high-frequency crystal reference to a lower frequency. The downsampled lower frequency can be even more accurate than the original source relative to its longer time period. This also minimizes or eliminates the need for digital resampling.

[0096] Embodiments of the methods and systems disclosed herein may include signal processing firmware / hardware. In some embodiments, long blocks of data are acquired at a high sampling rate, rather than extracting multiple data sets at varying sampling rates. Typically, in modern path acquisition processes for vibration analysis, data is collected at a fixed sampling rate and a specified data length. Depending on the specific mechanical analysis requirements at hand, the sampling rate and data length may vary from one path point to another. For example, a motor may require a relatively slow sampling rate and high resolution to distinguish operating speed harmonics from line frequency harmonics. However, the practical trade-off is that achieving this higher resolution requires increased acquisition time. In contrast, some high-speed compressors or gear sets require much higher sampling rates to measure the amplitude of relatively high-frequency data, although precise resolution may not necessarily be required. Ideally, however, it would be better to collect data with very long sample lengths at very high sampling rates. When digital acquisition equipment first became commonplace in the early 1980s, A / D sampling, digital storage, and computational power were not anywhere near the levels available today, resulting in a trade-off between the time required for data collection and the desired resolution and accuracy. Because of these limitations, some analysts in the field have even refused to abandon their analog tape recording systems, which were not significantly impacted by these digitization drawbacks. Several hybrid systems have been employed, digitizing playback of recorded analog data at various sampling rates and lengths as needed, though these systems undoubtedly have a lower degree of automation. As previously mentioned, a more common approach balances data collection time with analysis capability and digitally acquires blocks of data at various sampling rates and lengths, storing these blocks individually and digitally. In embodiments, data with long data lengths can be collected and stored at the highest practical sampling rate, such as 1024 kHz, corresponding to a 40 kHz Fmax. This long block of data can be acquired in the same amount of time as the shorter lengths at lower sampling rates employed by prior methods, thus adding no significant delay to the samples at the measurement point, which is always a concern during path collection. In embodiments, the data from the analog tape recordings is digitized to a precision that ensures it can be considered practically continuous or "analog" for many purposes, including those of the present invention, unless the context indicates otherwise.

[0097] Embodiments of the methods and systems disclosed herein may include storing calibration data and maintenance history on an onboard card set. Many data acquisition devices that rely on interfacing with a PC to work store their calibration coefficients on the PC. This is especially true for complex data acquisition devices that have many signal paths and therefore calibration tables that can be very large. In an embodiment, the calibration coefficients are stored in a flash memory that permanently remembers this data or any other important information about the matter for all practical purposes. This information may include nameplate information, such as the serial number of a single component, firmware or software version number, maintenance history, and calibration tables. In an embodiment, the DAQ box remains calibrated and continues to store all such important information, regardless of which computer the box is ultimately connected to. A PC or external device can poll this information at any time for implantation or information exchange purposes.

[0098] Embodiments of the methods and systems disclosed herein may include fast path creation using hierarchical templates. In the field of vibration monitoring and general parameter monitoring, data monitoring points need to be established in a database or functionally equivalent location. These points are associated with various attributes, including the following categories: transducer attributes, data collection settings, mechanical parameters, and operating parameters. Transducer attributes will include probe type, probe mounting type, and probe mounting direction or axis orientation. Data collection attributes associated with the measurement will include sampling rate, data length, IEPE (probe power) and coupling requirements, hardware integration requirements, 4-20 or voltage interface, range and gain settings (if applicable), filter requirements, etc. Mechanical parameter requirements relative to a specific point will include the following: operating speed, bearing type, and bearing parameter data corresponding to rolling bearings, including the middle diameter, number of balls, inner ring diameter, and outer ring diameter; for tilting pad bearings, this parameter data will include the number of bearing pads, etc. For measurement points on equipment such as gearboxes, required parameters would include, for example, the number of teeth on each gear; for induction motors, the number of rotor bars and posts; for compressors, the number of blades and / or vanes; for fans, the number of vanes; for belt / pulley systems, the number of belts and the associated belt pass frequency calculated based on pulley size and pulley center distance; for measurements near couplers, the coupler type and number of teeth in the gear coupler may be essential, among others. Operating parameter data would include workload, which could be expressed in megawatts, flow rate (gas or liquid), percentage, horsepower, feet per minute, etc. Operating temperature (both ambient and operating), operating pressure, humidity, etc. may also be relevant. As can be seen, the setup information required for a single measurement point can be quite extensive. Performing any reasonable analysis of this data is also crucial. Machinery, equipment, and bearing-specific information is essential for identifying fault frequencies and predicting various types of specific faults. Transducer properties and data collection parameters are crucial for correctly interpreting the data and providing constraints for the appropriate type of analysis technique. The traditional way of entering this data is to enter it manually, usually at the lowest hierarchical level, for example, the bearing level for mechanical parameters or the transducer level for data collection setup information, and the operation is quite cumbersome. However, it cannot be overemphasized that the importance of the hierarchical relationships required to organize the data for analysis and interpretation as well as the storage and movement of data. Here, we mainly emphasize the storage and movement of data. In essence, the setup information discussed above is extremely redundant at the lowest hierarchical level. However, because it has a strong hierarchical nature, it can be stored efficiently in this form. In an embodiment, the hierarchical nature can be utilized when copying data in the form of templates. For example, a hierarchical storage structure suitable for many purposes is defined from the general to the specific of the company, plant or site, unit or process, machine, equipment, shaft element, bearing and transducer.Copying data associated with a specific machine, device, shaft element or bearing is much easier than copying the data only at the lowest transducer level. In an embodiment, the system not only stores data in this hierarchical manner, but also uses hierarchical templates to robustly support the rapid duplication of data. The similarity of components at a specific hierarchical level contributes to the efficient storage of data in a hierarchical format. For example, many machines have common components such as motors, gearboxes, compressors, belts, fans, etc. More specifically, many motors can be easily classified as induction, DC, fixed speed or variable speed motors. Many gearboxes can be divided into common groupings such as input / output, input pinion / intermediate gear / output pinion, 4 cylinders (poster), etc. In a factory or company, there are many similar types of equipment that are purchased and standardized for cost and maintenance reasons. This results in a large number of similar types of equipment being repeated, and therefore provides a great opportunity to utilize a hierarchical template approach.

[0099] Embodiments of the methods and systems disclosed herein may include smart bands. Smart bands can be applied to any processed signal feature derived from any dynamic input or group of inputs in order to analyze data and achieve a correct diagnosis. Furthermore, smart bands can even include miniature or relatively simple diagnostics to implement more robust and complex smart bands. Historically, in the field of mechanical vibration analysis, alarm bands have been used to define relevant spectral bands for analyzing and / or trending significant vibration modes. Alarm bands typically consist of a region of the spectrum (amplitude plotted against frequency) defined between a low-frequency limit and a high-frequency limit. The amplitudes between these limits are summed in the same manner used to calculate the total amplitude. Smart bands offer greater flexibility because they can be applied not only to a specific frequency band, but also to a set of spectral peaks, such as harmonics of a single peak, true peak level or crest factor derived from a time waveform, the overall waveform derived from the vibration envelope spectrum, or other specialized signal analysis techniques or logical combinations (AND, OR, XOR, etc.) of these signal properties. Furthermore, countless other types of parametric data (including system load, motor voltage and phase information, bearing temperature, flow rate, etc.) can also be used as the basis for forming additional smart bands. In an embodiment, smart belt symptoms can be used as building blocks for an expert system whose engine uses these inputs to derive diagnostic information. Some of these micro-diagnostics can then in turn be used as smart belt symptoms (smart belts can even include diagnostics) for a broader range of diagnostics.

[0100] Embodiments of the methods and systems disclosed herein may include a neural network expert system that utilizes smart bands. Typical vibration analysis engines are rule-based, i.e., they utilize an expert rule list engine that triggers a specific diagnosis when conditions are met. In contrast, a neural approach utilizes weighted triggering of multiple input stimuli into smaller analysis engines or neurons, which in turn feed a simplified weighted output to other neurons. The outputs of these neurons can also be classified into smart bands that in turn feed other neurons. This results in a more hierarchical approach to expert diagnosis than the one-shot approach of a rule-based system. In an embodiment, the expert system utilizes this neural approach using smart bands; however, the system does not preclude the reclassification of rule-based diagnostics into smart bands as additional stimuli for the expert system to utilize. In this sense, although it is essentially a neural approach at the highest level, it can be summarized as a hybrid approach.

[0101] Embodiments of the methods and systems disclosed herein may include the use of a database hierarchy in the analysis. Smart belt fault phenomena and diagnostics may be assigned to multiple hierarchical database levels. For example, a smart belt may be called "Looseness" at the bearing level, trigger "Looseness" at the equipment level, and trigger "Looseness" at the machine level. Another example would be performing a smart belt diagnostic called "Horizontal Plane Phase Flip" on a coupling and generating a smart belt diagnostic called "Vertical Coupling Misalignment" at the machine level.

[0102] Embodiments of the methods and systems disclosed herein may include an expert system GUI. In one embodiment, the system uses a graphical approach to define smart bands and diagnostics for the expert system. Entering fault phenomena, rules, or broader smart bands for creating specific machine diagnostics can be tedious and time-consuming. One approach to making this process more convenient and efficient is to provide a graphical approach using wiring. The proposed graphical interface consists of four main components: a fault phenomenon component library, a diagnostic library, a tool library, and a graphical wiring area ("GWA"). The fault phenomenon component library includes a variety of spectra, waveforms, envelopes, and any type of signal processing features or feature groupings (such as spectrum peaks, spectrum harmonics, waveform true peaks, waveform crest factors, spectrum warning bands, etc.). Each component can be assigned additional attributes. For example, a spectrum peak component can be assigned the frequency or order (or multiple orders) of the operating speed. Some components can be predefined or user-defined, such as 1×, 2×, or 3× the operating speed, 1×, 2×, or 3× the gear mesh, 1×, 2×, or 3× the blade passes, or the number of motor rotor bars × the operating speed.

[0103] In one embodiment, a diagnostic library contains various predefined and user-defined diagnostics, such as misalignment, imbalance, looseness, and bearing failure. Similar to components, diagnostics can also be used as building blocks for more complex diagnostics. In one embodiment, the tool library includes logical operations such as AND, OR, and XOR, or other methods that combine the various components listed above, such as maximization, minimization, interpolation, averaging, and other statistical operations. Various components, tools, and diagnostics are represented by icons that are graphically connected in the desired manner. Embodiments of the methods and systems disclosed herein may include an expert system graphical GUI for defining expert system smart bands and diagnostics. Entering symptoms, rules, or broader smart bands for creating specific machine diagnostics can be tedious and time-consuming. One approach to making this process more convenient and efficient is to provide a graphical approach using routing. The proposed graphical interface consists of four main components: a symptom parts bin, a diagnostics bin, a tools bin, and a graphical routing area ("GWA"). In an embodiment, the fault phenomenon component library contains various spectra, waveforms, envelopes, and any type of signal processing features or feature groupings, such as spectral peaks, spectral harmonics, waveform true peaks, waveform crest factors, spectral warning bands, etc. Each component can be assigned additional attributes; for example, a spectral peak component can be assigned the frequency or order (or multiple orders) of the operating speed. Some components can be predefined or user-defined, such as 1×, 2×, or 3× operating speed, 1×, 2×, or 3× gear mesh, 1×, 2×, or 3× blade passes, number of motor rotor bars × operating speed, etc. In an embodiment, the diagnostic library contains various predefined and user-defined diagnostics, such as misalignment, imbalance, looseness, bearing faults, etc. Similar to components, diagnostics can also be used as components to build more complex diagnostics. The tool library includes logical operations such as AND, OR, and XOR, or other ways to combine the various components listed above, such as maximization, minimization, interpolation, averaging, and other statistical operations. GWA can generally include components from the component library or diagnostics from the diagnostic library, which are connected together using tools to create diagnostics. The various components, tools, and diagnostics are represented by icons that are simply graphically connected together in the required manner.

[0104] Embodiments of the methods and systems disclosed herein may include graphical methods for backpropagation definitions. In embodiments, the expert system also provides the system with learning opportunities. If a unique set of stimuli or smart bands is known to correspond to a specific fault or diagnosis, then a set of coefficients can be backpropagated that, when applied to a future set of similar stimuli, will produce the same diagnosis. In embodiments, if multiple sets of data are present, then a best fit method can be used. Unlike the smart band GUI, this embodiment will generate the wiring diagram itself. In embodiments, the user can customize the backpropagation method settings and use a database browser to match a specific data set with the desired diagnosis. In embodiments, the smart band GUI can be used to create or customize the desired diagnosis. In embodiments, the user can then press the GENERATE button and a dynamic wiring of the fault phenomenon to the diagnosis may appear on the screen when the algorithm is completed to achieve the best fit. In embodiments, various statistics will be displayed after the above operations are completed, detailing the extent to which the mapping process has progressed. In some cases, mapping may not be possible, for example if the input data is all zero or erroneous data (erroneous allocation), etc. Embodiments of the methods and systems disclosed herein may include a bearing analysis method. In an embodiment, the bearing analysis method may combine computer-aided design ("CAD"), predictive deconvolution, minimum variance distortionless response ("MVDR"), and spectral harmonic summation.

[0105] Embodiments of the methods and systems disclosed herein may include torsional vibration detection and analysis using transient signal analysis, a significant trend with the recent popularity of variable-speed machinery. Due primarily to the decreasing cost of motor speed control systems and the increasing cost and awareness of energy usage, the potentially significant energy savings of load control have become more economically justifiable. Unfortunately, one design aspect of this problem that is often overlooked is vibration. If a machine is designed to operate at only one speed, the physical structure is correspondingly much simpler to design, thereby avoiding structural and torsional mechanical resonances, each of which can significantly degrade the machine's mechanical health. This can include structural features such as the type of material used, its weight, the need for and placement of reinforcements, bearing type and location, and base support constraints. Even when a machine operates at a single speed, designing a structure to minimize vibration can still be a daunting task, potentially requiring computer modeling, finite element analysis, and field testing. With the addition of variable speeds, in many cases, designing for all required speeds becomes impossible. The problem then becomes one of minimization, such as through speed avoidance. This is why many modern motor controllers are typically programmed to skip or rapidly progress through specific speed ranges or speed bands. Embodiments include determining speed ranges within a vibration monitoring system. Non-torsional structural resonance is generally fairly easy to detect using conventional vibration analysis techniques. However, this is not the case for torsional vibration. A particular area of ​​current concern is the increasing incidence of torsional resonance problems, apparently due to the increased torsional stresses associated with speed variations and the operation of equipment at torsional resonant speeds. Unlike non-torsional structural resonance, which typically exhibits a significantly increased effect of housing or external vibrations, torsional resonance is generally free of such effects. In the case of shaft torsional resonance, the torsional motion caused by the resonance can only be discerned by looking for changes in velocity and / or phase. Current standard methods for analyzing torsional vibrations involve the use of specialized instrumentation. The methods and systems disclosed herein allow for the analysis of torsional vibrations without the use of such specialized instrumentation. This approach can consist of shutting down the machine and using strain gauges and / or velocity encoder plates and / or other specialized fixtures such as gears. Friction wheels are another alternative, but they are typically manual and require specialized analysis personnel. Generally, these techniques can be prohibitively expensive and / or cumbersome. Continuous vibration monitoring systems are becoming increasingly popular due to decreasing costs and increasing convenience (e.g., remote access). In an embodiment, torsional velocity and / or phase changes can be discerned using only the vibration signal.In an embodiment, transient analysis techniques can be used to distinguish torsional induced vibrations from velocity changes due to process control.In embodiments, factors used for discrimination may focus on one or more of the following: the rate of speed change caused by variable speed motor control is relatively slow, continuous, and steady; torsional speed changes tend to be brief, impulsive, and not continuous; torsional speed changes tend to be fluctuating, likely decaying exponentially, while process speed changes do not; and small speed changes associated with torsional relative to the rotational speed of the shaft indicate that monitored phase behavior will exhibit fast or brief speed bursts, in contrast to the slow phase changes historically associated with increasing or decreasing machine speed (as represented by Bode or Nyquist plots).

[0106] Embodiments of the methods and systems disclosed herein may include improved integration using both analog and digital methods. When integrating signals digitally using software, the amplitude of the frequency data at the low end of the spectrum is essentially multiplied by a function that rapidly becomes infinite as it approaches zero, creating what is known in the industry as a "ski-slope" effect. The amplitude of the ski-slope is essentially the instrument's noise floor. A simple remedy for this is a traditional hardware integrator, which can operate at a much higher signal-to-noise ratio than the signal already digitized. Furthermore, the amplification factor can be limited to a reasonable level, essentially prohibiting multiplication by very large numbers. However, at higher frequencies, where the frequency increases, an initial amplitude well above the noise floor may be multiplied by a very small number (1 / f), which reduces the amplitude to a level well below the noise floor. Hardware integrators have a fixed noise floor, although this lower limit does not decrease with the current lower-amplitude high-frequency data. In contrast, the same digital multiplication of the digitized high-frequency signal still proportionally reduces the noise floor. In an embodiment, hardware integration may be used below the unity gain point (at a value typically determined based on the gain being unity and / or the desired signal-to-noise ratio) and software integration may be used above the unity gain value to produce the desired results. In an embodiment, the integration is performed in the frequency domain. In an embodiment, the resulting mixed data may then be converted back into a waveform that should be far superior to either the hardware integrated or software integrated data in terms of signal-to-noise ratio. In an embodiment, the advantages of hardware integration are used in conjunction with the advantages of digital software integration to achieve the maximum signal-to-noise ratio. In an embodiment, a first order progressive hardware integrator high pass filter with curve fitting allows some relatively low frequency data to pass while reducing or eliminating noise, thereby allowing very useful analytical data to be recovered that was eliminated by the improper filter.

[0107] Embodiments of the methods and systems disclosed herein may include adaptive scheduling techniques for continuous monitoring. Continuous monitoring is typically performed using a pre-mux, with the goal of selecting a few channels from among many data channels to feed the hardware signal processing, A / D, and processing components of the DAQ system. This is primarily due to practical cost considerations. A compromise is to not monitor all points continuously (instead, monitoring a smaller range using alternative hardware methods). In embodiments, multiple scheduling levels are provided. In embodiments, at the lowest level, which is the most continuous, all measurement points are cycled through in a round-robin fashion. For example, if acquiring and processing a measurement point takes 30 seconds and there are 30 measurement points, each point will be serviced every 15 minutes. However, if a point should alarm based on user-selected criteria, its priority can be increased so that it is serviced more frequently. Because each alarm can have multiple levels of severity, there may be multiple levels of priority in monitoring. In embodiments, more severe alarms are monitored more frequently. In embodiments, various additional advanced signal processing techniques can be applied at less frequent intervals. Embodiments may take advantage of the increased processing power of the PC, and the PC may be able to temporarily pause the round-robin route collection (with multiple collection layers) process and stream the desired amount of data for a point of its choice. Embodiments may incorporate various advanced processing techniques such as envelope processing, wavelet analysis, and many other signal processing techniques. In embodiments, after collecting this data, the DAQ card group will continue its route at the point where it was interrupted. In embodiments, the various PC scheduled data collections will follow their own schedule, which will be less frequent than the DAQ card routes. This can be set by the hour, day, number of route cycles (e.g., once every 10 cycles), and also add schedules based on their alarm severity priority or measurement type (e.g., motors can be monitored differently than fans).

[0108] Embodiments of the methods and systems disclosed herein may include a data collection resident feature. In embodiments, a data collection box used for route collection, real-time analysis, and generally as a collection tool can be separated from its PC (tablet or other device) and powered by an external power source or a suitable battery. In embodiments, the data collector still retains continuous monitoring capabilities, and its onboard firmware can perform dedicated monitoring functions for extended periods of time or can be remotely controlled for further analysis. Embodiments of the methods and systems disclosed herein may include extended statistical capabilities for continuous monitoring.

[0109] Embodiments of the methods and systems disclosed herein may include environmental sensing, local sensing, and vibration for analysis. In embodiments, ambient temperature and pressure, sensored temperature and pressure, and long-term / medium-term vibration analysis can be combined with long-term / medium-term vibration analysis to predict any condition or characteristic range. Variations can include infrared sensing, infrared thermography, ultrasound, and many other sensor types and input types, combined with vibration or with each other. Embodiments of the methods and systems disclosed herein may include intelligent routing. In embodiments, the software of the continuous monitoring system adapts / adjusts the data collection sequence based on statistics, analysis, data alerts, and dynamic analysis. Typically, routing is set based on the channels to which the sensors are connected. In embodiments, a Mux can couple any input Mux channel to (e.g., eight) output channels via crosspoint switches. In embodiments, as channels enter an alarm state or the system identifies a critical deviation, the Mux will suspend the normal routing configured in the software to collect specific, synchronized data from channels sharing critical statistical variations for more advanced analysis. Embodiments include implementing intelligent ODS or intelligent transfer functions.

[0110] Embodiments of the methods and systems disclosed herein may include smart ODS and transfer functions. In embodiments, due to the system's multiplexers and crosspoint switches, ODS, transfer functions, or other specialized tests can be performed on all vibration sensors connected to a machine / structure, showing how various points on the machine are interconnected. In embodiments, data streams at 40-50 kHz and longer (e.g., at least one minute) can reveal information different from what a normal ODS or transfer function would display. In embodiments, the system will be able to determine smart line signatures based on the data / statistics / analysis to be used. Smart line signatures deviate from standard routes and perform ODS on a machine, structure, or multiple machines and structures that may exhibit correlations due to the conditions / data guiding them. In embodiments, for transfer functions, an impact hammer can be used on one channel to compare it to other vibration sensors on the machine. In embodiments, the system can perform transfer functions using changes in machine or system conditions such as load, speed, temperature, or other changes. In embodiments, different transfer functions can be compared to each other over time. In an embodiment, different transfer functions can be strung together like a movie that can show how a mechanical failure progresses, for example, a bearing can show how it goes through four stages of bearing failure, etc. Embodiments of the methods and systems disclosed herein may include a layered mux. In an embodiment, the layered mux can allow for modular output of more (16, 24, or more) channels to multiples of an 8-channel card set, which allows for multiple channels of data to be collected simultaneously for more complex analysis and faster data collection. Disclosed herein are methods and systems for continuous ultrasonic monitoring, including providing continuous ultrasonic monitoring of rotating elements and bearings in energy production facilities.

[0111] See also Figure 8 The present invention generally involves digitally collecting or streaming waveform data 2010 from a machine 2020, which can operate at speeds varying from relatively slow rotational or oscillatory speeds to relatively fast speeds under different circumstances. The waveform data 2010 on at least one machine can include data from a single-axis sensor 2030 mounted at a constant reference location 2040 and from a three-axis sensor 2050 mounted at varying locations (or at multiple locations), including location 2052. In an embodiment, the waveform data 2010 can be vibration data acquired simultaneously from each sensor 2030, 2050 in a gap-free format over a duration of several minutes, with a maximum resolvable frequency sufficient to capture periodic and transient impact events. By way of example, the waveform data 2010 can include vibration data that can be used to generate operational deformation modes. If desired, this data can also be used to diagnose vibrations, based on which machine maintenance plans can be prescribed.

[0112] In an embodiment, the machine 2020 may further include a housing 2100, which may include a drive motor 2110 capable of driving a shaft 2120. The shaft 2120 may be supported for rotation or oscillation via a set of bearings 2130 (e.g., including a first bearing 2140 and a second bearing 2150). A data collection module 2160 may be connected to (or resident on) the machine 2020. In one example, the data collection module 2160 may be located and accessed via a cloud network facility 2170, collect waveform data 2010 from the machine 2020, and transmit the waveform data 2010 to a remote location. The working end 2180 of the drive shaft 2120 of the machine 2020 may drive a windmill, a fan, a pump, a drill, a gear system, a drive system, or other working element, as the technology described herein can be applied to a wide range of machines, equipment, tools, etc. that include rotating or oscillating elements. In other cases, a generator may replace the motor 2110, and the working end of the drive shaft 2120 may direct the rotational energy to the generator to produce electricity, rather than consume electricity.

[0113] In an embodiment, waveform data 2010 can be acquired using a predetermined path format based on the layout of machine 2020. Waveform data 2010 can include data from a single-axis sensor 2030 and a three-axis sensor 2050. The single-axis sensor 2030 can serve as a reference probe with one data channel and be fixed at a constant location 2040 on the machine being surveyed. The three-axis sensor 2050 can serve as a three-axis probe with three data channels (e.g., three orthogonal axes) and can be moved from one test point to the next according to a predetermined diagnostic path format. In one example, sensors 2030 and 2050 can be manually installed on machine 2020 and, in certain service instances, can be connected to a separate portable computer. The reference probe can remain in one location, while the user can move the three-axis vibration probe along a predetermined path on the machine (e.g., bearing to bearing). In this example, the user is instructed to position the sensors at predetermined locations to complete the survey of the machine (or portion thereof).

[0114] See also Figure 9 , shows a portion of an exemplary machine 2200 according to the present invention having a three-axis sensor 2210 mounted to a location 2220 associated with a bearing of a motor of the machine 2200 having an output shaft 2230 and an output member 2240. Figure 9 and Figure 10 , shows an exemplary machine 2300 according to the present invention having a triaxial sensor 2310 and a uniaxial vibration sensor 2320, which is used as a reference sensor connected to a constant position on the machine 2300 during a vibration survey. The triaxial sensor 2310 and the uniaxial vibration sensor 2320 can be connected to a data collection system 2330.

[0115] In other examples, the sensors and data collection modules and devices may be integrated into or reside on the rotating machine. In these examples, the machine may include multiple single-axis sensors and multiple three-axis sensors at predetermined locations. The sensors may be installed as pre-installed equipment and provided by the original equipment manufacturer (OEM), or installed at a different time in a retrofit application. Data collection module 2160, etc., may select and use the single-axis sensors and, while moving to each of the three-axis sensors, acquire data only from the single-axis sensors during waveform data collection 2010. Data collection module 2160 may reside on machine 2020 and / or be connected via cloud network infrastructure 2170.

[0116] See also Figure 8Various embodiments include collecting waveform data 2010 by digitally recording locally or streaming over a cloud network facility 2170. Waveform data 2010 can be collected continuously to ensure no gaps and, in some aspects, can be similar to analog recorded waveform data. Waveform data 2010 from all channels can be collected over one to two minutes, depending on the rotational or oscillatory speed of the monitored machine. In embodiments, the data sampling rate can be relatively high relative to the operating frequency of the machine 2020.

[0117] In an embodiment, a second reference sensor may be used and data may be collected on a fifth data channel. Thus, a single-axis sensor may be the first channel, and three-axis vibration may occupy the second, third, and fourth data channels. This second reference sensor, like the first sensor, may be a single-axis sensor, such as an accelerometer. In an embodiment, the second reference sensor may remain in the same location on the machine as the first reference sensor for the entire vibration survey. The location of the first reference sensor (i.e., the single-axis sensor) may be different from the location of the second reference sensor (i.e., the other single-axis sensor). In some instances, a second reference sensor may be used when the machine has two axes that operate at different speeds, with the two reference sensors being located on two different axes. Following this example, other single-axis reference sensors may be used at other constant locations associated with the rotating machine.

[0118] In an embodiment, the waveform data can be transmitted electronically in a gapless format at a relatively high rate over a relatively long period of time. In one example, the period is 60 to 120 seconds. In another example, the sampling rate is 100 kHz and the maximum resolvable frequency (Fmax) is 40 kHz. According to the present invention, it should be understood that the waveform data can be displayed so as to more closely resemble some of the data that can be obtained from the previous example of analog recording of waveform data.

[0119] In an example, sampling, frequency band selection, and filtering techniques can allow one or more portions of a long data stream (i.e., one to two minutes in duration) to be undersampled or oversampled to achieve varying effective sampling rates. To this end, interpolation and decimation can be used to further achieve varying effective sampling rates. For example, oversampling can be applied to a frequency band close to the rotational or oscillatory operating speed of the sampled machine or to its harmonics, because the vibration effects at these frequencies may often be more pronounced within the machine's operating range. In an embodiment, the digital sample data set can be decimated to produce a lower sampling rate. According to the present invention, it should be understood that decimation herein can be the opposite of interpolation. In an embodiment, decimating the data set can include first applying a low-pass filter to the digital sample data set and then undersampling the data set.

[0120] In one example, a 100 Hz sample waveform can be undersampled at every tenth of the digital waveform to produce an effective sampling rate of 10 Hz, but the remaining nine-tenths of the waveform are effectively discarded and not included in the modeling of the sample waveform. Furthermore, this type of unmodified undersampling may generate spurious frequencies due to the undersampling rate (i.e., 10 Hz) associated with the 100 Hz sample waveform.

[0121] Most hardware used for analog-to-digital conversion uses a sample-and-hold circuit that charges a capacitor for a given period of time in order to determine the average value of a waveform relative to a specific time. According to the present invention, it should be understood that the value of a waveform relative to a specific time is not a linear value, but rather more similar to a cardinal sine ("sinc") function; therefore, it can be shown that waveform data at the center of a sampling interval may be given greater importance, with the cardinal sine signal exponentially decaying away from the center of the sampling interval.

[0122] With the above example, a 100 Hz sample waveform can be hardware sampled at 10 Hz, so each sample point is averaged over 100 milliseconds (e.g., each point of a 100 Hz sampled signal can be averaged over 10 milliseconds). In contrast to effectively discarding nine-tenths of the data points of the sampled waveform as discussed above, the present invention can include weighting adjacent data. Adjacent data can refer to the previously discarded sample points and the one remaining point that is retained. In one example, a low-pass filter can be used to find the average of adjacent sample data in a linear manner, that is, determining the sum of every ten points and then dividing the sum by ten. In another example, the adjacent data can be weighted using a sinc function. The process of weighting the initial waveform using a sinc function can be called an impulse function, or in the time domain, it can be called a convolution.

[0123] The present invention is applicable not only to digitizing waveform signals based on detected voltages, but also to digitizing waveform signals based on current waveforms, vibration waveforms, and image processing signals, including rasterization of video signals. In one example, decimation can be performed for window resizing on a computer screen, albeit in at least two directions. In these further examples, it will be appreciated that insufficient undersampling can be demonstrated. To this end, insufficient oversampling or upsampling can be demonstrated in a similar manner, so that interpolation can be used as with decimation, rather than simply undersampling itself.

[0124] According to the present invention, it should be understood that interpolation, as used herein, may refer to first applying a low-pass filter to digitally sampled waveform data and then upsampling the waveform data. According to the present invention, it should be understood that real-world scenarios may often require the use of non-integer factors for decimation, interpolation, or both. To this end, the present invention encompasses sequential interpolation and decimation to achieve non-integer factor ratios for interpolation and decimation. In one example, sequential interpolation and decimation can be defined as applying a low-pass filter to the sample waveform, then interpolating the waveform after low-pass filtering, and then decimating the waveform after interpolation. In one embodiment, the vibration data can be looped to purposefully simulate the looping of a conventional tape recorder, using digital filtering techniques in conjunction with efficient splicing to allow for longer-term analysis. According to the present invention, it should be understood that the above techniques do not preclude waveform, spectral, and other types of analysis that can be processed and displayed using a user GUI at the time of collection. According to the present invention, it should be understood that the new system can allow this functionality to be performed in parallel with the high-performance collection of raw waveform data.

[0125] Regarding the time to collect data, it's important to understand that the time savings achieved by compromising data resolution by collecting data at varying sampling rates and data lengths may not actually be as great as anticipated. Consequently, each time the data acquisition hardware is stopped and started, especially when performing hardware autoscaling, latency issues may arise. The same can be true for data retrieval of line information (i.e., test locations), which is typically performed in a database format and at very slow speeds. Storing raw data to disk (whether solid-state or otherwise) can also be quite slow.

[0126] In contrast, many embodiments include digital streaming waveform data 2010 as disclosed herein, while also enjoying the benefit of loading route parameter information while only having to set up the data acquisition hardware once. Since the waveform data 2010 is streamed to a file, there is no need to open and close the file or use storage media to switch between loading and writing operations. It can be shown that collecting and storing waveform data 2010 as described herein can produce relatively more meaningful data in a significantly shorter time than traditional batch data acquisition methods. In view of the above, an example includes a motor. To achieve a sufficiently high resolution, a data length of 4K points (i.e., 4096) can be used to collect relevant waveform data for the motor in order to (in particular) distinguish electrical sideband frequencies. For fans or blowers, a reduced resolution of 1K (i.e., 1024) can be used. In some cases, 1K can be the minimum waveform data length requirement. The sampling rate can be 1280Hz, which is equivalent to an Fmax of 500Hz. According to the present invention, it should be understood that an industry-standard oversampling factor of 256 satisfies the two-fold (2×) oversampling required by the Nyquist criterion, with some additional margin to accommodate anti-aliasing filter rolloff. The time to collect these waveform data is 1024 points at 1280 Hz, or 800 milliseconds.

[0127] To improve accuracy, the waveform data can be averaged. Eight averages can be used with a fifty percent overlap (for example). This would extend the time from 800 milliseconds to 36 seconds, equivalent to 800 msec × 8 averages × 0.5 (overlap ratio) + 0.5 × 800 msec (unoverlapped head and tail). After collecting waveform data at Fmax = 500 Hz, a higher sampling rate can be used. In one example, ten times (10×) the previous sampling rate can be used, with Fmax = 10 kHz. With this example, eight averages can be used with a fifty percent (50%) overlap to collect waveform data at a higher rate (which may be equivalent to a collection time of 360 msec or 0.36 seconds). According to the present invention, it should be understood that it may be necessary to read the hardware collection parameters for the higher sampling rate from the route list and allow the hardware to automatically scale or reset other necessary hardware collection parameters, or both. To this end, a delay of several seconds can be added to accommodate the change in sampling rate. In other cases, the introduction of delays can accommodate automatic scaling of the hardware and changes in hardware collection parameters that may be required when using the lower sampling rates disclosed herein. In addition to accommodating changes in sampling rates, additional time is required to read route point information from the database (i.e., the monitored location and the next location to be monitored), display the route information, and process the waveform data. In addition, the display of the waveform data and / or associated spectrum may also consume a significant amount of time. Based on the above, 15 to 20 seconds may have passed while the waveform data is being obtained at each measurement point.

[0128] In other examples, additional sampling rates can be added, but this may make the total amount of time for the vibration survey longer because the time is added up to the conversion time from one sampling rate to another and the time to obtain additional data at the different sampling rates. In one example, a lower sampling rate is used, such as a sampling rate of 128 Hz, where Fmax = 50 Hz. With this example, in addition to the other conditions mentioned above, at this sampling rate, the vibration survey will require an additional 36 seconds for the first set of averaged data, and therefore the total time spent at each measurement point will increase even more significantly. Other examples include similar digital streaming of gapless waveform data disclosed herein for wind turbines and other machines that may have relatively low-speed rotating or oscillating systems. In many examples, the collected waveform data may include long data samples at a relatively high sampling rate. In one example, the sampling rate may be 100 kHz and the sampling duration may be two minutes across all recorded channels. In many examples, one channel may be used for a single-axis reference sensor, and the other three data channels may be used for a three-channel sensor with three axes. According to the present invention, it will be appreciated that longer data lengths can be displayed to facilitate detection of very low-frequency phenomena. Furthermore, longer data lengths can be displayed to accommodate the inherent speed variations in wind turbine operation. Longer data lengths can also be displayed to provide the opportunity to use numerous averages, as discussed herein, to achieve very high spectral resolution and enable tape looping for certain spectral analyses. Many advanced analysis techniques are available today, as these techniques can utilize available, long, uninterrupted lengths of waveform data in accordance with the present invention.

[0129] According to the present invention, it should also be appreciated that simultaneously collecting waveform data from multiple channels can facilitate the implementation of transfer functions between the multiple channels. Furthermore, simultaneously collecting waveform data from multiple channels can facilitate establishing phase relationships across the machine, thereby enabling the use of more complex correlations by relying on the fact that waveforms are collected from each channel simultaneously. In other examples, more channels can be used in data collection to reduce the time it takes to complete an entire vibration survey by allowing waveform data to be collected simultaneously from multiple sensors (which would otherwise have to be collected in a sequential manner, moving from sensor to sensor during the vibration survey).

[0130] The present invention incorporates the use of at least one single-axis reference probe on a channel to allow for relative phase comparisons between channels. A reference probe can be an accelerometer or other type of transducer that does not move during a vibration survey of a machine and is therefore fixed in a constant position. Multiple reference probes can be deployed at appropriate fixed positions (i.e., constant locations) throughout the vibration data collection for the vibration survey. In some instances, up to seven reference probes can be deployed, depending on the capacity of the data collection module 2160. Using transfer functions or similar techniques, the relative phases of all channels can be compared across all selected frequencies. By maintaining one or more reference probes in a constant position while moving or monitoring other three-axis vibration sensors, it can be shown that the entire machine can be mapped in terms of amplitude and relative phase. This is true even if there are more measurement points than data collection channels. Using this information, operational deformation modes can be generated to represent the dynamic motion of the machine in 3D, providing a valuable diagnostic tool. In some embodiments, one or more reference probes can provide relative phase rather than absolute phase. While it should be understood that relative phase may not be as valuable as absolute phase for some applications, it can still be shown that relative phase information is very useful, according to the present invention.

[0131] In embodiments, the sampling rate used during a vibration survey can be digitally synchronized to a predetermined operating frequency, which may be correlated to a relevant parameter of the machine, such as rotational or oscillation speed. This allows for the use of synchronous averaging techniques to extract more information. According to the present invention, it should be appreciated that this can be achieved without using key phasors or reference pulses from rotating shafts, which are typically unavailable for route-collected data. This allows for the removal of non-synchronous signals from complex signals without requiring the use of key phasors to implement synchronous averaging. This can be particularly effective when analyzing a specific pinion in a gearbox, or any component typically found in complex mechanical mechanisms. In many cases, key phasors or reference pulses are rarely available for route-collected data, but the methods disclosed herein can overcome this lack. In embodiments, the machine being analyzed may contain multiple shafts operating at different speeds. In some cases, each shaft may have a single-axis reference probe. In other cases, only a single-axis reference probe at a constant position on one shaft may be used to correlate the phase of one shaft with another. In embodiments, variable-speed equipment may be more easily analyzed using relatively longer data durations than single-speed equipment. Vibration surveys can be performed at several machine speeds in the same continuous vibration data set using the same techniques disclosed herein. These techniques can also allow for studying changes in the relationship between vibration and the rate of change of speed, which was not previously possible.

[0132] In embodiments, because raw waveform data can be captured in the gapless digital format disclosed herein, a multitude of analysis techniques are available. The gapless digital format facilitates multiple avenues for analyzing waveform data in a variety of ways after a specific problem has been identified. Vibration data collected according to the techniques disclosed herein can provide analysis of transient, semi-periodic, and very low-frequency phenomena. Waveform data collected according to the present invention can comprise a relatively long stream of raw, gapless waveform data that can be conveniently replayed as needed and subjected to a variety of complex analysis techniques. A wide variety of such techniques can provide various forms of filtering to extract low-amplitude modulations from transient impulse data, which can be contained within the relatively long stream of raw, gapless waveform data. In accordance with the present invention, it will be appreciated that in past data collection practices, these types of phenomena were often discarded during the averaging process of spectral processing algorithms because previous data acquisition modules were designed solely for periodic signals; or because large portions of the original raw signal were often discarded when it was known they would not be used, thereby archiving these phenomena through size reduction methods.

[0133] In an embodiment, there is a method for monitoring vibration in a machine having at least one shaft supported by a set of bearings. The method includes monitoring a first data channel assigned to a single-axis sensor at a constant location associated with the machine. The method also includes monitoring second, third, and fourth data channels assigned to a three-axis sensor. The method further includes simultaneously recording gapless digital waveform data from all data channels while the machine is operating; and determining a change in relative phase based on the digital waveform data. The method also includes a three-axis sensor located at multiple locations associated with the machine when the digital waveform is obtained. In an embodiment, the second, third, and fourth channels are assigned together to a series of three-axis sensors, each located at a different location associated with the machine. In an embodiment, data is received simultaneously from all sensors on all channels thereof.

[0134] The method further includes determining an operating deformation mode based on changes in relative phase information and waveform data. In an embodiment, the constant position of the reference sensor refers to a position associated with a shaft of the machine. In an embodiment, the three-axis sensors in a series of three-axis sensors are located at different locations and are associated with different bearings in the machine. In an embodiment, the constant position refers to a position associated with a shaft of the machine, and wherein the three-axis sensors in the series of three-axis sensors are located at different locations and are associated with different bearings supporting the shaft in the machine. Various embodiments include methods for sequentially monitoring vibration or similar process parameters and signals from multiple channels (which may be referred to as a set) of rotating or oscillating machines or similar processing machinery. In various examples, a set may include one to eight channels. In other examples, a set may represent a logical grouping of measurements on the monitored equipment, whether the measurement locations are temporary for measurement, provided by the original equipment manufacturer, retrofitted, or one or more combinations thereof.

[0135] In one example, a collection can monitor bearing vibration in a single direction. In another example, the collection can monitor three different directions (e.g., orthogonal directions) using a triaxial sensor. In other examples, the collection can monitor four or more channels, where the first channel can monitor a single-axis vibration sensor, and the second, third, and fourth channels can each monitor three directions of the triaxial sensor. In other examples, the collection can be affixed to a group of adjacent bearings on the same equipment or related shafts. Various embodiments provide methods comprising strategies for collecting waveform data from various collections deployed in vibration studies, etc., in a relatively more efficient manner. These methods also include simultaneously monitoring a reference channel, which is assigned to a constant reference position associated with the collection of monitored machines. Collaboration with the reference channel can be shown to support more complete correlation of the collected waveforms with the collection. The reference sensor on the reference channel can be a single-axis vibration sensor or a phase-referenced sensor that can be triggered by, for example, a reference position on a rotating shaft. As disclosed herein, the above methods can further include simultaneously recording gapless digital waveform data from all channels in each collection at a relatively high sampling rate to include all frequencies deemed necessary for proper analysis of the monitored machinery during operation. The data in the collection can be streamed to the storage medium for subsequent processing in a seamless manner. The storage medium can be connected to cloud network facilities, local data links, Bluetooth TM connection, cellular data connection, etc.

[0136] In embodiments, the methods disclosed herein include strategies for collecting data from various sets, including digital signal processing techniques that can then be applied to the data from the sets to highlight or better isolate specific frequency or waveform phenomena. This may be contrasted with current methods that collect multiple sets of data at different sampling rates or that include integrated different hardware filtering configurations, which (referred to as a priori hardware configurations) offer relatively low post-processing flexibility due to their constraints. Furthermore, it can be shown that these hardware configurations increase the time of vibration surveys due to the delays associated with configuring the hardware for each individual test. In embodiments, the methods including strategies for collecting data from various sets include data labeling techniques for classifying portions of the streaming data as homogeneous data belonging to a specific set. In one example, the categories can be defined as operating speed. In this way, numerous sets can be generated from a set that conventional systems collect as only one set. Many embodiments include post-processing analysis techniques for comparing the relative phases of all relevant frequencies between each channel of the collected sets and between all channels of all monitored sets, as appropriate.

[0137] refer to Figure 12 Many embodiments include a first machine 2400 having a rotating or oscillating component 2410, or both, each supported by a set of bearings 2420, including a bearing set 2422, a bearing set 2424, a bearing set 2426, and more bearing sets as needed. The first machine 2400 can be monitored by a first sensor set 2450. The first sensor set 2450 can be configured to receive signals from sensors initially installed on (or added later to) the first machine 2400. The sensors on the machine 2400 can include single-axis sensors 2460, such as single-axis sensor 2462, single-axis sensor 2464, and more single-axis sensors as needed. In many examples, the single-axis sensor 2460 can be positioned within the machine 2400 to allow sensing of the position of one of the rotating or oscillating components 2410 of the machine 2400.

[0138] Machine 2400 may also include three-axis (e.g., orthogonal) sensors 2480, such as three-axis sensor 2482, three-axis sensor 2484, and more (as needed). In many examples, three-axis sensors 2480 may be positioned within machine 2400 to allow sensing of the position of each of multiple bearing sets 2420 associated with rotating or oscillating components of machine 2400. Machine 2400 may also include temperature sensors 2500, such as temperature sensor 2502, temperature sensor 2504, and more (as needed). Machine 2400 may also include tachometer sensors 2510 or more (as needed) that each detail the RPM of one of its rotating components. Using the above example, first sensor set 2450 may survey the above sensors associated with first machine 2400. To this end, first sensor set 2450 may be configured to receive eight channels of data. In other examples, first sensor set 2450 may be configured to include more or fewer than eight channels, as needed. In this example, the eight channels include two channels that can each monitor a single-axis reference sensor signal and three channels that can monitor a triaxial sensor signal. The remaining three channels can monitor two temperature signals and a signal from a tachometer. In one example, according to the present invention, first set 2450 can monitor single-axis sensor 2462, single-axis sensor 2464, triaxial sensor 2482, temperature sensor 2502, temperature sensor 2504, and tachometer sensor 2510. During a vibration survey of machine 2400, first set 2450 can monitor triaxial sensor 2482 and then triaxial sensor 2484.

[0139] According to the present invention, after monitoring triaxial sensor 2484, first set 2450 may monitor other triaxial sensors on machine 2400 as needed, which sensors are part of a predetermined route list associated with a vibration survey of machine 2400. During the vibration survey, first set 2450 may continue to monitor single-axis sensor 2462, single-axis sensor 2464, two temperature sensors 2502, 2504, and tachometer sensor 2510, while first set 2450 may continue to monitor multiple triaxial sensors 2480 as per the predetermined route plan for the vibration survey.

[0140] refer to Figure 12Many embodiments include a second machine 2600 having a rotating or oscillating component 2610, or both, each supported by a set of bearings 2620, including a bearing set 2622, a bearing set 2624, a bearing set 2626, and more bearing sets as needed. The second machine 2600 can be monitored by a second sensor set 2650. The second sensor set 2650 can be configured to receive signals from sensors initially installed (or added later) on the second machine 2600. The sensors on the machine 2600 can include single-axis sensors 2660, such as single-axis sensor 2662, single-axis sensor 2664, and more single-axis sensors as needed. In many examples, the single-axis sensor 2660 can be positioned within the machine 2600 to allow sensing of the position of one of the rotating or oscillating components 2610 of the machine 2600.

[0141] Machine 2600 may also have three-axis (e.g., orthogonal) sensors 2680, such as three-axis sensor 2682, three-axis sensor 2684, three-axis sensor 2686, three-axis sensor 2688, and more as needed. In many examples, three-axis sensor 2680 may be positioned on machine 2600 to allow sensing of each of the plurality of bearing sets 2620 associated with a rotating or oscillating component of machine 2600. Machine 2600 may also have temperature sensors 2700, such as temperature sensor 2702, temperature sensor 2704, and more as needed. Machine 2600 may also have tachometer sensor 2710 or more tachometer sensors, such as needed, that each detail the revolutions per minute of one of its rotating components.

[0142] Using the above example, second sensor set 2650 can survey the above sensors associated with second machine 2600. To this end, second sensor set 2650 can be configured to receive data from eight channels. In other examples, second sensor set 2650 can be configured to have more or fewer than eight channels, as needed. In this example, the eight channels include one channel that can monitor a single-axis reference sensor signal and six channels that can monitor two triaxial sensor signals. The remaining channels can monitor temperature signals. In one example, second sensor set 2650 can monitor single-axis sensor 2662, triaxial sensor 2682, triaxial sensor 2684, and temperature sensor 2702. During a vibration survey of machine 2600 according to the present invention, second sensor set 2650 can first simultaneously monitor triaxial sensor 2682 and triaxial sensor 2684, and then simultaneously monitor axis sensor 2686 and triaxial sensor 2688.

[0143] According to the present invention, after monitoring triaxial sensor 2680, second set 2650 may monitor other triaxial sensors on machine 2600 (in pairs simultaneously) as needed, which are part of a predetermined route list associated with a vibration survey of machine 2600. During the vibration survey, second set 2650 may continuously monitor single-axis sensor 2662 and temperature sensor 2702 at their constant locations, while second set 2650 may continuously monitor multiple triaxial sensors according to the predetermined route plan for the vibration survey.

[0144] Continue to refer Figure 12 Many embodiments include a third machine 2800 having a rotating or oscillating component 2810, or both, each supported by a set of bearings 2820, including a bearing set 2822, a bearing set 2824, a bearing set 2826, and further bearing sets as needed. Third machine 2800 can be monitored by a third sensor set 2850. Third set 2850 can be configured with a single-axis sensor 2860 and two three-axis (e.g., orthogonal) sensors 2880, 2882. In many examples, single-axis sensor 2860 can be user-secured to machine 2800, allowing for sensing the position of one rotating or oscillating component of machine 2800. Three-axis sensors 2880, 2882 can also be user-positioned to machine 2800, allowing for sensing the position of each bearing in a set of bearings, each associated with a rotating or oscillating component of machine 2800. Third set 2850 can also include a temperature sensor 2900. Unlike the first set 2450 and the second set 2650, the third set 2850 and its sensors can be moved to other machines.

[0145] Many embodiments also include a fourth machine 2950 having a rotating or oscillating assembly 2960, or both, each supported by a set of bearings 2970 comprising a bearing set 2972, a bearing set 2974, a bearing set 2976, and further bearing sets as needed. When a user moves a third sensor set 2850 to the fourth machine 2950, ​​the fourth machine 2950 can also be monitored by the third sensor set 2850. Many embodiments also include a fifth machine 3000 having a rotating or oscillating assembly 3010, or both. While fifth machine 3000 may not be explicitly monitored by any sensor or sensor set during operation, it can generate vibration or other impulse energy of sufficient magnitude to be recorded in data associated with any of machines 2400, 2600, 2800, or 2950 in a vibration survey.

[0146] Many embodiments include monitoring a first sensor set 2450 on a first machine 2400 along a predetermined route as disclosed herein. Many embodiments also include monitoring a second sensor set 2650 on a second machine 2600 along a predetermined route. The location of machine 2400 in proximity to machine 2600 can be included in the contextual metadata for both vibration surveys. A third sensor set 2850 can be moved between machine 2800, machine 2950, ​​and other suitable machines. Machine 3000 is not equipped with onboard sensors but can be monitored by third sensor set 2850 when needed. Machine 3000 and its operating characteristics can be recorded in metadata associated with vibration surveys of other machines to indicate its contribution due to its proximity.

[0147] Many embodiments incorporate hybrid database adaptation for reconciling relational metadata and streaming raw data formats. According to the present invention, it will be appreciated that, unlike older systems that utilize traditional database structures to associate nameplates and operating parameters (sometimes considered metadata) with discrete and relatively simple individual data measurements, more modern systems can collect relatively large amounts of raw streaming data at higher sampling rates and higher resolutions. At the same time, according to the present invention, it will be appreciated that the metadata network used to link and access this raw data, or to associate with this raw data, or both, is expanding at an ever-increasing rate.

[0148] In one example, a single overall vibration level can be collected as part of a route or prescribed list of measurement points. The collected data can then be correlated with database measurement location information for a point on the bearing housing surface on a particular machine that is vertically adjacent to the coupler. Mechanical analysis parameters relevant to the appropriate analysis can be associated with the point located on the surface. Examples of mechanical analysis parameters relevant to the appropriate analysis can include the operating speed of the shaft passing through the measurement point on the surface. Other examples of mechanical analysis parameters relevant to the appropriate analysis can include one or a combination of the following: the operating speed of all component shafts of a piece of equipment and / or machine; the type of bearing being analyzed, such as plain bearings or roller bearings; the number of gear teeth on the gears (if a gearbox is present); the number of poles in the motor; the differential and line frequencies of the motor; the size of the roller bearing elements; the number of fan blades, etc. Examples of mechanical analysis parameters relevant to the appropriate analysis can further include machine operating conditions, such as the load on the machine and whether the load is expressed in percentage, wattage, airflow, head pressure, horsepower, etc. Other examples of mechanical analysis parameters include information about neighboring machines that may affect the data obtained during the vibration study.

[0149] According to the present invention, it should be understood that a wide variety of equipment and machinery types can support many different categories, each of which can be analyzed in significantly different ways. For example, it can be shown that some machines, such as screw compressors and hammer mills, operate noisily and can be expected to vibrate more significantly than other machines. It can be shown that machines known to vibrate more significantly require a change in vibration levels that might be considered acceptable for quieter machines.

[0150] The present invention further encompasses hierarchical relationships discovered within the collected vibration data that can be used to support appropriate analysis of the data. An example of hierarchical data includes the interconnections of machine components (e.g., bearings) being measured in a vibration survey, the relationships between the bearings (including how a bearing is connected to a specific shaft, a specific pinion is mounted within a specific gearbox on a specific shaft), and the relationships between the shafts, pinions, and gearboxes. The hierarchical data can further include at which specific point within the machine's gear train the monitored bearing is positioned relative to other components in the machine. The hierarchical data can also detail whether the measured bearing in the machine is connected to another machine whose vibrations could affect what is being measured in the machine being studied.

[0151] Analysis of the layered data, which is correlated with each other, vibration data from bearings or other components, can use table lookups, correlation searches between frequency patterns derived from the raw data, and specific frequencies from the machine metadata. In some embodiments, the above data can be stored in a relational database and retrieved from the database. In an embodiment, the National Instruments Technical Data Management Solution (TDMS) file format can be used. The TDMS file format can be optimized for streaming various types of measurement data (i.e., binary digital samples of waveforms) and for manipulating layered metadata.

[0152] Many embodiments include a hybrid relational metadata binary storage approach (HRM-BSA). The HRM-BSA may include a relational database engine based on Structured Query Language (SQL). The relational database engine based on Structured Query Language may also include a raw data engine that may be optimized for throughput and storage density of uniform and relatively unstructured data. According to the present invention, it will be appreciated that benefits may be demonstrated in terms of collaboration between hierarchical metadata and an SQL relational database engine. In one instance, tagging techniques and indicator benchmarks may be used to correlate a raw database engine with an SQL relational database engine. Three examples of correlation between a raw database engine and an SQL relational database engine link include: indicators from an SQL database to raw data; indicators from auxiliary metadata tables or similar raw data groupings to an SQL database; and independent storage tables outside of the SQL database or raw data technology domain.

[0153] refer to Figure 13 The present invention may include indicators for Group 1 and Group 2, which may include: associated file names; path information; table names; database key fields used with existing SQL database technology, which can be used to associate specific database segments or locations; asset attributes of specific measurement raw data streams; records with associated timestamps / date stamps; or associated metadata, such as operating parameters, panel conditions, etc. By way of example, device 3200 may include a first machine 3202, a second machine 3204, and many other machines within device 3200. First machine 3202 may include a gearbox 3210, a motor 3212, and other components. Second machine 3204 may include a motor 3220 and other components. A number of waveforms 3230, including waveform 3240, waveform 3242, waveform 3244, and other waveforms (as needed), may be collected from machines 3202 and 3204 within device 3200. Waveforms 3230 may be associated with a local tag link table 3300 and a linked raw data table 3400. The machines 3202, 3204 and their components may be associated with a linked table with a relational database 3500. The linked table raw data table 3400 and the linked table with the relational database 3500 may be associated with a linked table with an optional independent storage table 3600.

[0154] The present invention may include markers that can be applied to time stamps or sample lengths within the raw waveform data. Markers generally fall into two categories: preset markers or dynamic markers. Preset markers can be associated with preset or existing operating conditions, such as load, head pressure, airflow in cubic feet per minute, ambient temperature, RPM, etc. These preset markers can be fed directly into the data acquisition system. In some cases, preset markers can be collected on a data channel in parallel with waveform data, such as waveforms for vibration, current, voltage, etc. Alternatively, the values ​​for the preset markers can be manually entered.

[0155] For dynamic tagging, such as trending data, comparing similar data can be crucial. For example, comparing vibration amplitude and pattern to a set of repeatable operating parameters. One embodiment of the present invention includes a parallel channel input, which is a key phasor trigger pulse from the operating shaft, providing RPM information at the instant of collection. In this example of dynamic tagging, a portion of the collected waveform data can be tagged with the appropriate speed or speed range.

[0156] The present invention may also include dynamic tags that can be associated with data that can be obtained from post-processing and analysis of the sample waveforms. In other embodiments, dynamic tags can also be associated with parameters derived after collection (including RPM) and other metrics derived from operation (such as alarm conditions such as maximum RPM). In some instances, many modern devices that are candidates for vibration surveys with portable data collection systems as described herein do not include tachometer information. This may be because even though the measurement of RPM may be most important for vibration surveys and analysis, it is not always practical or cost-effective to add a tachometer. It should be understood that for fixed speed machinery, especially when the approximate speed of the machine can be determined in advance, obtaining an accurate RPM measurement is not as important; however, variable speed drives are becoming more and more common. According to the present invention, it should also be understood that various signal processing techniques can allow RPM to be derived from the raw data without the need for a dedicated tachometer signal.

[0157] In many embodiments, RPM information can be used to tag segments of raw waveform data in the collection history. Other embodiments include techniques for collecting instrument data along a prescribed route for a vibration study. Dynamic tagging can enable analysis and trending software to use multiple segments of the collection time interval indicated by the tag (e.g., two minutes) as multiple historical collection sets, rather than just one set as in previous systems where the route collection system would have stored data for only one RPM setting. As previously described, this in turn can be extended to any other operating parameter, such as load setting, ambient temperature, etc. However, dynamic tags, which can be placed in a type of index file pointing to the raw data stream, can categorize portions of the stream in a homogeneous entity, making it easier to compare with previously collected portions of the raw data stream.

[0158] Many embodiments include a hybrid relational metadata binary storage method that can fully utilize existing technologies for relational data streams and raw data streams. In embodiments, the hybrid relational metadata binary storage method can match data with various tagged links. Tagged links can allow for rapid searches of relational metadata and more efficient analysis of raw data using conventional SQL techniques and existing technologies. This can enable the use of many features, links, compatibilities, and extensions that conventional database technologies cannot provide.

[0159] Tag linking can also allow the use of conventional binary storage and data compression techniques to quickly and efficiently store raw data. This can mean that many of the functions, links, compatibilities, and extensions provided by conventional raw data technologies, such as TMDS (National Instruments), UFF (Universal File Format, such as UFF58), etc., can be utilized. Tag linking can further allow the use of tag technology links, where a richer data set from the collection can be accumulated at the same collection time as a more conventional system. The richer data set from the collection can store data snapshots associated with predetermined collection criteria, and the proposed system can utilize tag technology to derive multiple snapshots from the collected data stream. This can mean that a relatively more comprehensive analysis of the collected data can be achieved. One of the benefits of this can include more trending vibration points at specific frequencies or orders of operating speed relative to RPM, load, operating temperature, flow rate, etc., which can be collected in a time similar to the time it takes to collect data using a conventional system.

[0160] In an embodiment, the platform 100 may include a local data collection system 102 deployed in an environment 104 to monitor signals from machines, machine components, and the environment in which the machines are located. These machines may include heavy machinery deployed at a local work site or distributed work sites under common control. Heavy machinery may include earthmoving equipment, heavy-duty on-highway industrial vehicles, heavy-duty off-road industrial vehicles, and industrial machines deployed in various environments, such as turbines, turbomachinery, generators, pumps, pulley systems, manifold and valve systems, etc. In an embodiment, heavy industrial machinery may also include earthmoving equipment, compacting equipment, hauling equipment, lifting equipment, transport equipment, aggregate production equipment, equipment used for concrete construction, and piling equipment. In an example, earthmoving equipment may include excavators, backhoes, loaders, bulldozers, skid steers, trenchers, motor graders, automatic scrapers, crawler loaders, and wheeled loader forks. In an example, construction vehicles may include dump trucks, tank trucks, dump trucks, and trailers. In an example, material handling equipment can include cranes, transporters, forklifts, and cranes. In an example, construction equipment can include tunnel loading and unloading equipment, rollers, concrete mixers, hot mix equipment, road builders (ramming machines), stone crushers, pavers, slurry sealers, sprayers, and heavy pumps. Other examples of heavy industrial equipment can include different systems for implementing traction, structure, drive trains, controls, and information. Heavy industrial equipment can include many different drive trains and combinations thereof to provide mobile power and provide power for accessories and onboard functions. In each such example, the platform 100 can deploy the local data collection system 102 to an environment 104 in which machines, motors, pumps, etc. operate and are directly connected and integrated into each machine, motor, pump, etc.

[0161] In an embodiment, the platform 100 may include a local data collection system 102 deployed in an environment 104 to monitor data from machines in operation and machines under construction (e.g., turbines and generators, such as Siemens TM SGT6-5000F TM Gas turbine, SST-900 TM Steam turbine, SGen6-1000A TM Generator and SGen6-100A TMIn embodiments, a local data collection system 102 may be deployed to monitor a steam turbine as it rotates in an electric current caused by hot water vapor that may be directed through the turbine, but which may also be generated from different sources such as a gas furnace, a nuclear core, a molten salt ring, etc. In these systems, the local data collection system 102 may monitor the turbine as well as water or other fluid in a closed loop cycle in which the water condenses and is then heated until it evaporates again. The local data collection system 102 may monitor the steam turbine and the fuel source deployed to heat the water into steam separately. In embodiments, the operating temperature of the steam turbine may be between 500°C and 650°C. In many embodiments, the steam turbine array may be arranged and configured for high, medium, and low pressures so that it can convert the corresponding steam pressures into rotational motion in an optimal manner.

[0162] The local data collection system 102 can also be deployed in a gas turbine arrangement, thereby monitoring not only the operating turbine but also the hot combustion gases fed into the turbine, which can reach temperatures exceeding 1500°C. Because these gases are much hotter than those in a steam turbine, the blades can be cooled using air that can be blown through smaller openings to form a protective film or boundary layer between the exhaust gas and the blades. This temperature profile can be monitored by the local data collection system 102. Unlike a typical steam turbine, a gas turbine engine includes a compressor, a combustor, and a turbine, all of which are journaled to rotate with the rotating shaft. The construction and operation of each of these components can be monitored by the local data collection system 102.

[0163] In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from a hydro turbine used as a rotary engine that can capture energy from flowing water and use it to generate electricity. The type of hydro turbine or hydroelectric power plant selected for a project can be based on the site's still water height (commonly referred to as head) and flow rate (or volume of water). In this example, a generator can be placed on top of a shaft connected to the hydro turbine. When the turbine captures the naturally flowing water in its blades and rotates, the turbine sends rotational power to the generator to produce electricity. In this way, the platform 100 can monitor signals from the generator, turbine, local water supply system, and flow controls (such as dam windows and sluice gates). In addition, the platform 100 can monitor the local conditions of the power grid, including load, predicted demand, frequency response, etc., and include this information in the monitoring and control deployed by the platform 100 in these hydroelectric environments.

[0164] In an embodiment, the platform 100 may include a local data collection system 102 deployed in an environment 104 to monitor signals from energy production environments such as thermal, nuclear, geothermal, chemical, biomass, carbon-based fuels, and hybrid energy recycling equipment. Many of these devices may use various forms of energy acquisition equipment powered by heat from nuclear, gas, sunlight, and molten salt heat sources, such as wind turbines, hydro turbines, and steam turbines. In an embodiment, the elements in such systems may include transmission lines, heat exchangers, desulfurization scrubbers, pumps, coolers, recuperators, chillers, etc. In an embodiment, certain embodiments of turbomachinery, turbines, scroll compressors, etc. may be configured as array controls to monitor large facilities that generate electricity for consumption, provide refrigeration, generate steam for local manufacturing and heating, etc., and the array control platform may be provided by a supplier of industrial equipment (such as Honeywell and its Experion TM PKS platform). In embodiments, platform 100 can specifically communicate with and integrate local manufacturer-specific controls and can allow devices from one manufacturer to communicate with other devices. Furthermore, platform 100 allows local data collection systems 102 to collect information from systems from many different manufacturers. In embodiments, platform 100 can include local data collection systems 102 deployed in environment 104 to monitor data from offshore industrial equipment, marine diesel engines, shipbuilding facilities, oil and gas plants, refineries, petrochemical plants, ballast water treatment solutions, marine pumps and turbines, and the like.

[0165] In an embodiment, platform 100 may include a local data collection system 102 deployed in environment 104 to monitor signals from heavy industrial equipment and processes, including monitoring one or more sensors. By way of example, a sensor may be a device that can detect or respond to some type of input in a physical environment, such as an electrical signal, a thermal signal, or an optical signal. In an embodiment, local data collection system 102 may include, but is not limited to, a plurality of sensors including, for example, temperature sensors, pressure sensors, torque sensors, flow sensors, heat sensors, smoke sensors, arc sensors, radiation sensors, position sensors, acceleration sensors, strain sensors, pressure cycle sensors, pressure sensors, air temperature sensors, and the like. The torque sensor may include a magnetic torsion angle sensor. In one embodiment, the torsion and velocity sensors in local data collection system 102 may be similar to those discussed in U.S. Patent No. 8,352,149 to Meachem, issued on January 8, 2013 and incorporated herein by reference as if fully set forth herein. In an embodiment, one or more sensors may be provided, such as tactile sensors, biosensors, chemical sensors, image sensors, humidity sensors, inertial sensors, and the like.

[0166] In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from sensors. These sensors may provide signals for fault detection, including excessive vibration, improper materials, improper material properties, accuracy of proper dimensions, accuracy of proper shapes, proper weights, and accuracy of balance. Other fault sensors include those used for inventory control and inspection to confirm that parts are packaged as planned and that parts have planned tolerances, sensors indicating packaging damage or crushing, and sensors that may indicate impact or damage in transit. Other fault sensors may include those detecting insufficient lubrication, excessive lubrication, the need to clean sensor detection windows, the need for maintenance due to low lubrication, and the need for maintenance due to blockage or reduced flow in lubrication areas.

[0167] In an embodiment, the platform 100 may include a local data collection system 102 deployed in an environment 104, which includes aircraft operations and manufacturing, and may include monitoring signals from sensors used for specific applications, such as sensors used in an aircraft's attitude and heading reference system (AHRS), such as gyroscopes, accelerometers, and magnetometers. In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from image sensors, such as semiconductor charge-coupled devices (CCDs) in complementary metal-oxide semiconductor (CMOS) or N-type metal-oxide semiconductor (NMOS, active MOS) technology, active pixel sensors, and the like. In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from sensors, such as infrared (IR) sensors, ultraviolet (UV) sensors, touch sensors, and proximity sensors. In an embodiment, the platform 100 may include a local data collection system 102 deployed in an environment 104 to monitor signals from sensors configured for optical character recognition (OCR), reading barcodes, detecting surface acoustic waves, detecting transponders, communicating with home automation systems, medical diagnostics, health monitoring, etc.

[0168] In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from sensors such as micro-electromechanical systems (MEMS) sensors, such as ST Microelectronic's TM The LSM303AH smart MEMS sensor includes a 3D digital linear acceleration sensor and a 3D digital magnetic sensor in an ultra-low-power, high-performance system-in-package.

[0169] In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from turbines, windmills, industrial vehicles, robots, and other large machines. These large machines include multiple components and elements that provide multiple subsystems on each machine. To this end, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from individual components such as axles, bearings, belts, barrels, gears, shafts, gearboxes, cams, brackets, camshafts, clutches, brakes, rollers, generators, feeders, flywheels, pads, pumps, jaws, robot arms, seals, sockets, sleeves, valves, wheels, actuators, motors, servomotors, and the like. Many machines and their components may include servomotors. The local data collection system 102 can monitor the motors, rotary encoders, and potentiometers of the servomechanisms to provide three-dimensional details of the position, placement, and progress of the industrial process.

[0170] In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from gear drives, powertrains, transfer cases, multi-speed shafts, transmissions, direct drives, chain drives, belt drives, shaft drives, magnetic drives, and similar meshing mechanical drives. In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor signals from industrial machine fault conditions, which may include overheating, noise, gear grinding, gear locking, excessive vibration, oscillation, under-inflation, over-inflation, and the like. Operational faults, maintenance indications, and maintenance or operational issues from interactions with other machines may occur during operation, installation, and maintenance. Faults may occur not only within the industrial machine's mechanism, but also within the infrastructure supporting the machine, such as its wiring and local mounting platform. In an embodiment, large industrial machines may face different types of fault conditions, such as overheating, noise, gear grinding, excessive vibration of machine components, fan vibration issues, and issues with rotating components of the machine.

[0171] In an embodiment, the platform 100 may include a local data collection system 102 deployed in an environment 104 to monitor signals from industrial machinery, including faults caused by premature bearing failures that may occur due to contamination or loss of bearing lubricant. In another example, mechanical defects such as bearing misalignment may occur. Many factors may cause failures, such as metal fatigue, so the local data collection system 102 can monitor cyclic and local stresses. Using this example, the platform 100 can monitor improper operation of machine components, lack of component maintenance and repair, corrosion of important machine components such as couplings or gearboxes, misalignment of machine components, etc. Although it is impossible to completely prevent the occurrence of failures, many industrial failures can be mitigated to reduce operational and economic losses. The platform 100 provides real-time monitoring and predictive maintenance in many industrial environments, which may be shown to be more cost-effective than a scheduled maintenance process that replaces components based on fixed time periods rather than the actual load and wear on the component or machine. To this end, the platform 10 can provide reminders for preventive measures or perform some preventive measures, such as complying with the machine's operating manual and model instructions, properly lubricating and maintaining machine parts, minimizing or eliminating machine overruns that exceed the machine's limited capacity, replacing worn but still usable parts as needed, and providing training to relevant personnel on the use of the machine.

[0172] In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104 to monitor multiple signals that may be carried by multiple physical, electronic, and symbolic formats or signals. The platform 100 may utilize signal processing, which encompasses various mathematical, statistical, computational, heuristic, and linguistic representations and processing of signals, as well as the various operations required to extract useful information from signal processing operations (e.g., techniques for representing, modeling, analyzing, synthesizing, sensing, acquiring, and extracting information from signals). In practice, signal processing can be performed using a variety of techniques, including but not limited to transformations, spectral estimation, statistical operations, probabilistic and random operations, numerical analysis, data mining, and the like. The processing of various types of signals forms the basis of many electrical or computational processes. Therefore, signal processing is applicable to nearly all disciplines and applications in industrial environments, such as audio and video processing, image processing, wireless communications, process control, industrial automation, financial systems, feature extraction, quality improvement such as noise reduction, and image enhancement. Signal processing for images can include pattern recognition for manufacturing inspection, quality control, and automated operational inspection and maintenance. Platform 100 can utilize a variety of pattern recognition techniques, including those that classify input data into multiple categories based on key features in order to identify patterns or regularities in the data. Platform 100 can also implement pattern recognition processes with machine learning capabilities and can be used in applications such as computer vision, speech and text processing, radar processing, handwriting recognition, and CAD systems. Platform 100 can utilize both supervised and unsupervised classification. Supervised learning classification algorithms can be used to create classifiers for image or pattern recognition based on training data from different object categories. Unsupervised learning classification algorithms can operate by using advanced analytical techniques such as segmentation and clustering to find hidden structures in unlabeled data. For example, some analytical techniques used for unsupervised learning include K-means clustering, Gaussian mixture models, and hidden Markov models. The algorithms used in both supervised and unsupervised learning pattern recognition methods enable the use of pattern recognition in a variety of high-precision applications. Platform 100 can utilize pattern recognition in applications related to face detection, such as security systems, tracking, sports-related applications, fingerprint analysis, medical and forensic applications, navigation and guidance systems, vehicle tracking, public infrastructure systems such as transportation systems, and license plate monitoring.

[0173] In an embodiment, the platform 100 may include a local data collection system 102 deployed in the environment 104, which uses machine learning to implement inference-based learning outcomes from computers without programming. Thus, the platform 100 can learn and make decisions from a set of data by making data-driven predictions and adapting to the data set. In an embodiment, machine learning may involve the machine learning system performing multiple machine learning tasks, such as supervised learning, unsupervised learning, and reinforcement learning. Supervised learning may involve providing a set of example inputs and required inputs to the machine learning system. Unsupervised learning may involve the learning algorithm itself structuring its inputs through methods such as pattern detection and / or feature learning. Reinforcement learning may involve the machine learning system executing in a dynamic environment and then providing feedback on correct and incorrect decisions. In an embodiment, machine learning may include multiple other tasks based on the output of the machine learning system. In an embodiment, tasks may also be categorized as machine learning problems, such as classification, regression, clustering, density estimation, dimensionality reduction, anomaly detection, etc. In an embodiment, machine learning may include a variety of mathematical and statistical techniques. In an example, many types of machine learning algorithms can include decision tree-based learning, association rule learning, deep learning, artificial neural networks, genetic learning algorithms, inductive logic programming, support vector machines (SVMs), Bayesian networks, reinforcement learning, representation learning, rule-based machine learning, sparse dictionary learning, similarity and metric learning, learning classifier systems (LCS), logistic regression, random forests, K-means, gradient boosting and adaptive boosting, K-nearest neighbor (KNN), prior algorithms, etc. In an embodiment, certain machine learning algorithms (such as genetic algorithms defined to solve constrained and unconstrained optimization problems, which may be based on natural selection, i.e., the process of promoting biological evolution) can be used. With this example, genetic algorithms can be deployed to solve a variety of optimization problems that are not very suitable for standard optimization algorithms, including problems where the objective function is discontinuous, non-differentiable, random, or highly nonlinear. In one example, genetic algorithms can be used to solve mixed integer programming problems where some components are restricted to integer values. Genetic algorithms and machine learning techniques and systems can be used in computational intelligence systems, computer vision, natural language processing (NLP), recommendation systems, reinforcement learning, building graphical models, etc. In this instance, machine learning systems can be used to perform intelligent computing-based control and respond to tasks in a variety of systems, such as interactive websites and portals, brain-computer interfaces, online security and fraud detection systems, medical applications such as diagnostic and treatment assistance systems, and DNA sequence classification. In practical applications, machine learning systems can be used in advanced computing applications such as online advertising, natural language processing, robotics, search engines, software engineering, speech and handwriting recognition, pattern matching, game theory, computational anatomy, and bioinformatics systems. In practical applications, machine learning can also be used in financial and marketing systems, such as user behavior analysis, online advertising, economic evaluation, and financial market analysis.

[0174] The following provides information on combining Figures 1 to 6 Other details of the methods, systems, devices, and components described herein. In an embodiment, disclosed herein are methods and systems for cloud-based machine pattern recognition based on remote simulated industrial sensor fusion. For example, data streams from vibration, pressure, temperature, acceleration, magnetic field, electric field, and other simulated sensors can be multiplexed or otherwise fused, relayed over a network, and fed to a cloud-based machine learning facility that can use one or more models related to the operating characteristics of industrial machines, industrial processes, or their components or elements. The model can be created by a person with relevant experience in industrial environments and can be associated with a training data set, such as a model created by manual or machine analysis of data collected by sensors in the environment or other similar environments. The learning machine can then operate on other data, initially using a set of rules or elements of the model to provide various outputs, such as classifying the data, identifying certain patterns (such as patterns indicating the presence of a fault, or patterns indicating operating conditions such as fuel efficiency, power generation, etc.). The machine learning facility can receive feedback, such as one or more inputs or success metrics, so that it can train or improve its initial model, such as by adjusting weights, rules, parameters, etc. based on the feedback. For example, a fuel consumption model for an industrial machine might include physical model parameters characterizing weight, motion, drag, momentum, inertia, acceleration, and other factors indicative of consumption, as well as chemical model parameters, such as those predicting energy produced and / or consumed through combustion, chemical reactions during battery charging and discharging, and so on. The model can be refined by feeding it data from sensors located within the machine's environment, within the machine itself, and data indicating actual fuel consumption, so that the machine can provide increasingly accurate sensor-based fuel consumption estimates and can also provide outputs indicating changes that could be made to increase fuel consumption (e.g., changing the machine's operating parameters or other elements of the environment, such as ambient temperature or the operation of nearby machines). For example, if a resonance effect between two machines adversely affects one of the machines, the model can account for this and automatically provide outputs that alter the operation of one of the machines, such as reducing the resonance or increasing the fuel efficiency of one or both machines. By continuously adjusting parameters to align outputs with actual conditions, a machine learning facility can self-organize to provide highly accurate models of environmental conditions, such as those used to predict failures or optimize operating parameters. This can be used to increase fuel efficiency, reduce wear, increase output, extend operating life, avoid failure conditions, and many other purposes.

[0175] Figure 14 Illustrate the components and interactions of a data collection architecture that involves applying cognitive and machine learning systems to data collection and processing. Figure 14The data collection system 102 can be set up in an environment, such as an industrial environment that manufactures, assembles, or operates one or more complex systems of electromechanical systems and machines. The data collection system 102 can include onboard sensors and can receive input from one or more sensors (such as any type of analog or digital sensor disclosed herein) and from one or more input sources 116 (such as sources available via WiFi, Bluetooth, NFC, or other local network connections or via the Internet) through one or more input interfaces or ports 4008. Multiple sensors can be combined and multiplexed for processing, such as using one or more multiplexers 4002. Data can be cached or buffered in a cache / buffer 4022 and made available to external systems, such as the remote host processing system 112 described elsewhere herein (which can include a broad processing architecture 4024, including any elements described in conjunction with other embodiments described herein and in the figures) through one or more output interfaces and ports 4010 (which, in embodiments, can be separate from or identical to the input interfaces and ports 4008). The data collection system 102 can be configured to obtain inputs from the main processing system 112, such as inputs from the analysis system 4018, which can operate on data from the data collection system 102 and data from other input sources 116 to provide analysis results, which in turn can be provided to the data collection system as learning feedback input 4012 to assist in the configuration and operation of the data collection system 102.

[0176] The combination of inputs (including selecting which sensors or input sources to turn on or off) can be performed under the control of machine-based intelligence using a local cognitive input selection system 4004, an optional remote cognitive input selection system 4114, or a combination of both. The cognitive input selection systems 4004, 4014 can utilize the intelligence and machine learning capabilities described elsewhere herein, such as using detected conditions (e.g., conditions reported by input sources 116 or sensors), state information (including state information determined by a machine state recognition system 4020 that can determine states), such as those related to operating states, environmental states, states within a known process or workflow, states involving fault or diagnostic conditions, or many other states. This can include optimizing input selection and configuration based on learning feedback from a learning feedback system 4012, which can include providing training data (e.g., from the host processing system 112 or from other data collection systems 102 directly or through the host 112) and providing feedback metrics, such as success metrics calculated within an analysis system 4018 of the host processing system 112. For example, if a data stream consisting of a particular sensor and input combination produces positive results under a given set of conditions (e.g., provides improved pattern recognition, improved prediction, improved diagnosis, improved yield, improved return on investment, improved efficiency, etc.), metrics related to the results from the analysis system 4018 can be provided to the cognitive input selection systems 4004, 4014 via the learning feedback system 4012 to help configure future data collection to select combinations under those conditions (e.g., by de-energizing other sensors, allowing other input sources to be deselected). In embodiments, under the control of one or more cognitive input selection systems 4004, the selection and deselection of sensor combinations may occur in an automatically evolving manner, such as using genetic programming techniques, so that over time, based on learning feedback 4012 from the analysis system 4018, etc., effective combinations for a given state or set of conditions are promoted and less effective combinations are demoted, allowing the local data collection system to be gradually optimized and adapted for each unique environment. Thus, an automatically adapting multi-sensor data collection system is provided in which cognitive input selection is used with feedback to improve the effectiveness, efficiency, or other performance parameters of the data collection system in its specific environment. The performance parameters may be related to overall system metrics (e.g., financial benefits, process optimization results, energy production or use, etc.), analytical metrics (e.g., success in identifying patterns, making predictions, classifying data, etc.), and local system metrics (e.g., bandwidth utilization, memory utilization, power consumption, etc.). In an embodiment, the host's analysis system 4018, state system 4020, and cognitive input selection system 4114 may receive data from multiple data collection systems 102 so that optimization (including optimization of input selection) can be performed by coordinating the operation of multiple systems 102.For example, cognitive input selection system 4114 can understand that if one data collection system 102 is already collecting X-axis vibration data, the X-axis vibration sensor of another data collection system may be turned off, which may facilitate obtaining Y-axis data from another data collector 102. Thus, through the coordinated collection of the master cognitive input selection system 4114, the activities of multiple collectors 102 on numerous different sensors can provide a rich data set for the master processing system 112 without wasting energy, bandwidth, storage space, etc. As described above, optimization can be based on overall system success metrics, analytical success metrics, local system metrics, or a combination of the two.

[0177] Disclosed herein are methods and systems for cloud-based machine pattern analysis of multiple industrial sensor state information to provide expected state information for an industrial system. In an embodiment, machine learning can utilize a state machine, such as tracking the states of multiple analog and / or digital sensors, feeding the states to a pattern analysis facility, and determining the expected state of the industrial system based on historical data about the sequence of state information. For example, in the case where the temperature state of an industrial machine exceeds a certain threshold and then a fault condition occurs (such as a set of bearings failing), the temperature state can be tracked by a pattern recognizer, which can generate an output data structure indicating an expected bearing fault state whenever a high temperature input state is identified. Various measurements and expected states of temperature, pressure, vibration, acceleration, momentum, inertia, friction, heat, heat flux, current state, magnetic field state, electric field state, capacitance state, charge and discharge state, motion, position, and many other parameters can be managed by a state machine. The state can include a combined state, where the data structure includes a series of states, each state represented by a position in a byte-like data structure. For example, an industrial machine may be characterized by a genetic structure, such as a structure that provides pressure, temperature, vibration, and acoustic data, where the measurements of the structure occupy a position in the data structure, such that the combined state can be operated on a byte-like structure, such as for succinctly describing the current combined state of the machine or environment or succinctly describing the expected state. Such a byte-like structure can be used by a state machine for machine learning, such as pattern recognition that operates on the structure to determine patterns that affect the combined effects of multiple conditions. Many such structures can be tracked and used, such as in machine learning, to represent various combinations of various lengths of different elements that can be sensed in an industrial environment. In embodiments, the byte-like structure can be used in genetic programming techniques, such as by replacing different types of data or data from different sources and tracking the results over time, so as to generate one or more advantageous structures based on the success of these structures when used in real-world situations, such as successful predictions indicating the expected state or achieving successful operational results, such as increased efficiency, successful routing of information, and achieving profit growth. That is, by varying the data types and sources used in a byte-like structure (for machine optimization over time), a genetic programming-based machine learning facility can "evolve" a set of data structures that include a favorable mix of data types (e.g., pressure, temperature, and vibration) from a favorable mix of data sources (e.g., temperature from sensor X, vibration from sensor Y) for a given purpose. Different desired outcomes can result in different data structures that best support the efficient achievement of those outcomes over time through genetic programming with applied machine learning and upgrading of structures with favorable results (for the desired outcome of interest).The upgraded data structure can provide concise and efficient data for various activities described throughout the present invention, including storage in a data pool (which can be optimized by storing favorable data structures that provide the best operating results for a given environment), presentation in a data marketplace (such as presentation as the most efficient structure for a given purpose), and other activities.

[0178] In an embodiment, a platform is provided that includes cloud-based machine pattern analysis of state information from multiple simulated industrial sensors to provide expected state information for industrial systems. In an embodiment, a host processing system 112, such as one located in the cloud, may include a state system 4020 that can be used to infer or calculate the current state or to determine an expected future state related to the data collection system 102 or an aspect of the environment in which the data collection system 102 is located, such as a machine state, component, workflow, process, event (e.g., whether an event has occurred), object, person, condition, function, etc. Maintaining state information allows the host processing system 112 to perform analysis in one or more analysis systems 4018 to determine contextual information, apply semantic and conditional logic, and perform many other functions implemented by the processing architecture 4024 described throughout the present invention.

[0179] In one embodiment, a platform is provided that features a cloud-based policy automation engine for IoT (creating, deploying, and managing IoT devices). In one embodiment, the platform 100 includes (or is integrated with, or contains) a host processing system 112 (e.g., located on a cloud platform) and a policy automation engine 4032 for automating the creation, deployment, and policy management of IoT devices. Policies, including access policies, network usage policies, storage usage policies, bandwidth usage policies, device connection policies, security policies, rule-based policies, role-based policies, and other policies, may be required to govern the use of IoT devices. For example, because IoT devices may have many different network and data communications than other devices, policies may be required to indicate which devices a given device can connect to, the data it can transmit, and the data it can receive. Given the expected deployment of billions of devices with countless potential connections in the near future, configuring policies for IoT devices on a connection-by-connection basis is not feasible. Therefore, the intelligent policy automation engine 4032 may include cognitive features for creating, configuring, and managing policies. The policy automation engine 4032 may utilize information about possible policies from, for example, a policy database or library, which may contain one or more public sources of available policies. These policies can be written in one or more conventional policy languages ​​or scripts. Policy automation engine 4032 can apply policies based on one or more models, such as those based on characteristics of a given device, machine, or environment. For example, a large machine (such as one used for power generation) may include a policy that only authenticated local controllers can change certain parameters of the power generation, thereby preventing remote "takeover" by hackers. This can be achieved by automatically discovering and applying security policies that prevent the machine's control infrastructure from connecting to the internet, such as by requiring access authentication. Policy automation engine 4032 may include cognitive features, such as those that change the application and configuration of policies, such as those based on state information from state system 4020. Policy automation engine 4032 may receive feedback from learning feedback system 4012 based on one or more analysis results from analysis system 4018, overall system results (such as the extent of security vulnerabilities, policy violations, etc.), local results, and analysis results. By making changes and selections based on such feedback, policy automation engine 4032 can learn over time to automatically create, deploy, configure, and manage policies across a large number of devices, such as policies used to configure connections between IoT devices.

[0180] Disclosed herein are methods and systems for device-based sensors and data storage for industrial IoT devices, including device-based sensors and data storage for industrial IoT devices, wherein data from multiple sensors is multiplexed in a device for storing a fused data stream. For example, pressure and temperature data can be multiplexed into a data stream combining pressure and temperature in a time series, such as in a byte-like structure (where time, pressure, and temperature are bytes in the data structure, so that pressure and temperature remain correlated in time without the need for separate stream processing by an external system), or by addition, division, multiplication, subtraction, etc., so that the fused data can be stored on the device. Any sensor data type described throughout the present invention can be fused in this manner and stored in a local data pool, memory, or on an IoT device, such as a data collector, machine component, etc.

[0181] In one embodiment, a platform with on-device sensors and data storage for industrial IoT devices is provided. In one embodiment, a cognitive system is used in a self-organizing storage system 4028 of the data collection system 102. Sensor data, especially simulated sensor data, can consume significant storage capacity, particularly when the data collector 102 has multiple onboard sensor inputs or sensor inputs from the local environment. Simply storing all data indefinitely is generally not an option. Even transmitting all data may exceed bandwidth limitations or bandwidth allowances (e.g., exceeding the capacity of a cellular data plan). Therefore, storage strategies are required. These strategies often include capturing only portions of the data (e.g., snapshots), storing data for a limited period of time, or storing portions of the data (e.g., in an intermediate or abstracted form). There are many possible options among these and other options, and determining an appropriate storage strategy can be extremely complex. In one embodiment, the self-organizing storage system 4028 can utilize cognitive systems based on learning feedback 4012, such as using various metrics from the analysis system 4018 or another system in the primary cognitive input selection system 4114, such as overall system metrics, analytical metrics, and local performance indicators. The self-organizing storage system 4028 can automatically change storage parameters such as storage location (including local storage on the data collection system 102, storage on nearby data collection systems 102 (e.g., using peer organization), and remote storage, such as network-based storage), storage volume, storage duration, the type of data stored (including individual sensor or input source 116 data, as well as various combined or multiplexed data, such as data selected under the cognitive input selection systems 4004, 4014), storage type (e.g., use of RAM, flash memory, or other short-term storage relative to available hard drive space), storage organization (e.g., in raw form, hierarchical, etc.), and other parameters. Changes to parameters can be made based on feedback so that over time, the data collection system 102 adapts its data storage to optimize itself based on the conditions of its environment (e.g., a particular industrial environment) so that it stores the correct amount of data needed and the correct type of data available to users.

[0182] In embodiments, the local cognitive input selection system 4004 can organize and fuse data from various onboard sensors, external sensors (e.g., in the local environment), and other input sources 116 to the local collection system 102 into one or more fused data streams, thereby creating various signals using multiplexers 4002 and the like, such signals representing combinations, permutations, blends, layering, decimations, data-metadata combinations, etc. of the source analog and / or digital data processed by the data collection system 102. The selection of a particular sensor fusion can be determined locally by the cognitive input selection system 4004 based on learning feedback (e.g., various overall system, analysis system, and local system results and metrics) from the learning feedback system 4012. In embodiments, the system can learn a particular combination and permutation of fused sensors to best achieve appropriate state expectations, etc., as indicated by feedback from the analysis system 4018 regarding its ability to predict future states (e.g., various states processed by the state system 4020). For example, the input selection system 4004 can direct the selection of a subset of sensors from a larger set of available sensors, and can combine the inputs from the selected sensors, for example, by placing the inputs into bytes of a defined multi-bit data structure (e.g., by acquiring the signals at a given sampling rate or time and placing the results into a byte structure, then collecting and processing the bytes over time), by multiplexing in the multiplexer 4002 (e.g., by combining by additive mixing of consecutive signals), etc. Any of a variety of signal processing and data processing techniques for combining and fusion can be used, including convolution techniques, auto-conversion techniques, transformation techniques, etc. The particular fusion in question can be adapted to a given context through cognitive learning, such as by having the cognitive input selection system 4004 learn based on feedback 4012 from the results (e.g., transmitted by the analysis system 4018), so that the local data collection system 102 performs context-adaptive sensor fusion.

[0183] In embodiments, analysis system 4018 may be adapted to employ any of a variety of analysis techniques, including statistical and econometric techniques (e.g., linear regression analysis, using similarity metrics, heatmap-based techniques, etc.), reasoning techniques (e.g., Bayesian reasoning, rule-based reasoning, inductive reasoning, etc.), iterative techniques (e.g., feedback, recursive, feedforward, and other techniques), signal processing techniques (e.g., Fourier and other transforms), pattern recognition techniques (e.g., Kalman and other filtering techniques), search techniques, probabilistic techniques (e.g., random walks, random forest algorithms, etc.), simulation techniques (e.g., random walks, random forest algorithms, linear optimization, etc.), and other techniques. This may include computing various statistics or metrics. In embodiments, analysis system 4018 may be at least partially located on data collection system 102, such that the local analysis system may compute one or more metrics, such as metrics related to any of the items discussed throughout this disclosure. For example, metrics of efficiency, power utilization, memory utilization, redundancy, entropy, and other factors may be computed on-board, enabling data collection 102 to implement the various cognitive and learning functions discussed throughout this disclosure without relying on a remote (e.g., cloud-based) analysis system.

[0184] In embodiments, the host processing system 112, the data collection system 102, or both may include, be connected to, or be integrated with a self-organizing network system 4020, which may include a cognitive system for providing machine-based intelligence or organizing network utilization for data in the data collection system (e.g., processing analog data and other sensor data or other sources between one or more local data collection systems 102 and the host system 112, etc.). This may include organizing network utilization for source data transmitted to the data collection system, feedback data (analytic data provided to or provided by the learning feedback system 4012), data used to support the market (such as data described in conjunction with other embodiments), and output data provided from one or more data collection systems 102 via output interfaces and ports 4010.

[0185] Disclosed herein are methods and systems for a self-organizing data marketplace for industrial IoT data, including organizing available data elements in a marketplace for consumption by consumers based on training a self-organizing facility using a training set and feedback from marketplace success metrics. A marketplace can initially be established to provide available data collected from one or more industrial environments, such as by type, source, environment, machine, one or more modes, and the like, such as a menu or hierarchy. The marketplace can then modify the data collected, the organization of the data, the presentation of the data (including pushing the data to external sites, providing links, configuring APIs for accessing the data, and the like), and the pricing of the data, based on machine learning and other factors, which can change any of the various parameters previously described. The machine learning facility can manage all of these parameters through self-organization, for example, by changing parameters over time (including changing elements of the data types presented), the source data used to obtain each data type, the data structure presented (such as a byte-like structure, a fused or multiplexed structure (such as representing multiple sensor types), and a statistical structure (such as various mathematical results representing sensor information), data pricing, where the data is presented, how the data is presented (such as through an API, a link, a push message, etc.), how the data is stored, how the data is acquired, etc. As the parameters change, feedback can be obtained on success metrics such as number of views, revenue per visit (e.g., price paid), total revenue, unit profit, total profit, etc. and many other metrics, and the self-organizing machine learning facility can promote configurations that improve the success metric and demote configurations that do not improve, so that over time the market is gradually configured to present favorable data type combinations (e.g., combinations that provide stable predictions of the expected state of a given type of specific industrial environment) from favorable sources (e.g., reliable, accurate, and low-priced sources) at effective pricing (e.g., pricing that tends to provide higher overall profits from the market). The market may include spiders, web crawlers, etc. that find input data sources, such as data pools that publish potentially relevant data, connected IoT devices, etc. These can be trained by human users and improved by machine learning in a manner similar to that described elsewhere in this invention.

[0186] In an embodiment, a platform is provided that has a self-organizing data marketplace for industrial IoT data. Figure 15In an embodiment, a platform is provided having a cognitive data marketplace 4102 (in some cases referred to as a self-organizing data marketplace) for data collected by one or more data collection systems 102 or from other sensors or input sources 116 located in various data collection environments (e.g., industrial environments). In addition to the data collection systems 102, this data may include data collected, processed, or exchanged by IoT devices, including cameras, monitors, embedded sensors, mobile devices, diagnostic devices and systems, instrumentation systems, telematics systems, and the like, for monitoring various parameters and characteristics of machines, devices, components, parts, operations, functions, conditions, states, events, workflows, and other elements (collectively, the term "state") within the aforementioned environments. The data may also include metadata about any of the foregoing, such as metadata describing the data, indicating its origin, indicating elements regarding identity, access, roles, and permissions, providing a summary of the data, or extracting or otherwise supplementing one or more data items to enable further processing, such as for extracting, transforming, loading, and processing the data. This data (a term that includes metadata unless the context indicates otherwise) can be extremely valuable to third parties, either as individual elements (e.g., data about environmental conditions that can be used as an example of a condition in a process) or as an aggregate (e.g., data collected from many systems and devices in different environments that can be selectively used to develop behavioral models, train learning systems, etc.). With the deployment of billions of IoT devices (using countless connections), the amount of available data will explode. To enable access and utilization of data, cognitive data marketplace 4102 enables users to supply, find, consume, and trade various components, features, services, and processes in the form of data packets (e.g., batch data, data streams (including event streams), data from various data pools 4120, etc.). In embodiments, cognitive data marketplace 4102 can be included in, connected to, or integrated with one or more other components of the main processing architecture 4024 of the main processing system 112 (e.g., a cloud-based system), and can be connected to various sensors, input sources 115, data collection systems 102, etc. The cognitive data marketplace 4102 may include a marketplace interface 4108, which may include one or more supplier interfaces through which data suppliers may provide available data and one or more consumer interfaces through which data may be found and collected. The consumer interface may include an interface to a data marketplace search system 4118, which may include features that enable users to indicate the type of data they wish to retrieve (e.g., by entering keywords in a natural language search interface featuring data or metadata). The search interface may utilize various search and filtering techniques, including keyword matching, collaborative filtering (e.g., using a consumer's known preferences or characteristics to match results of other similar consumers and other consumers' past results), and ranking techniques (e.g., ranking based on various metrics of past results' success as described in conjunction with other embodiments of the present invention).In embodiments, a provisioning interface may allow data owners or suppliers to offer data through cognitive data marketplace 4102 in one or more packages (e.g., packaging batches of data, data streams, etc.). Suppliers may pre-package data by providing data from a single input source 116, a single sensor, etc., or by providing combinations, permutations, etc. (e.g., multiplexed analog data, mixed bytes of data from multiple sources, extract, load, and transform results, convolution results, etc.), as well as metadata associated with any of the foregoing. Packaging may include pricing on a per-batch, streaming basis (e.g., subscription to an event feed or other feed or stream), per-item basis, revenue share basis, or other basis. For data subject to pricing, data trading system 4114 may track orders, deliveries, and utilization, including order fulfillment. Trading system 4114 may include rich trading features, including digital rights management, for example, by managing keys that control access to purchased data and control usage (e.g., allowing a limited set of users or roles to use the data for a limited time, within a limited domain, or for a limited purpose). Transaction system 4114 may manage payments by processing credit cards, wire transfers, debits, and other forms of consideration, among other things.

[0187] In embodiments, cognitive data packaging system 4012 of marketplace 4102 may utilize machine-based intelligence to package data, for example, by automatically configuring data packets into batches, streams, pools, and the like. In embodiments, packaging may be based on one or more rules, models, or parameters, for example, by packaging or aggregating data that is likely to complement or supplement an existing model. For example, operational data from a group of similar machines (such as one or more industrial machines described herein) may be aggregated based on metadata indicating the data type or by identifying features or characteristics in the data stream that indicate the nature of the data. In embodiments, packaging may be performed using machine learning and cognitive capabilities, for example, by learning which combinations, permutations, blends, or layers of input sources 116, sensors, information from data pool 4120, and information from data collection system 102 are likely to meet user needs or generate success metrics. Learning may be based on learning feedback 4012, such as metrics determined by analytics system 4018, such as system performance metrics, data collection metrics, analytics metrics, and the like. In embodiments, success metrics may be associated with marketplace success metrics, such as package views, package engagement, package purchases or licenses, and payments made for packages. These metrics can be calculated in analysis system 4018, comprise specific feedback measure and search term and other input association, make cognitive packaging system 4110 can find and configure design and be used for providing higher value to consumer and provide higher return package for data supplier.In an embodiment, cognitive data packaging system 4110 can use learning feedback 4012 to change packaging automatically, for example, use different combinations, permutations, mixing etc. and change the weight applied to given input source, sensor, data pool etc., to promote favorable package and no longer emphasize less favorable package.This can use genetic programming and the similar technology of comparing the result of different packages to carry out.Feedback can comprise state information (as about various running states etc.) from state system 4020, and information about market conditions and state (as pricing and availability information of other data sources).Therefore, provide adaptive cognitive data packaging system 4110, its automatic adaptation condition thereby provides favorable data packet for market 4102.

[0188] In embodiments, a cognitive data pricing system 4112 may be provided to set pricing for data packages. In embodiments, data pricing system 4112 may use a set of rules, models, and the like, such as those based on supply conditions, demand conditions, pricing from various available sources, and the like, to set pricing. For example, pricing for a package may be configured to be based on the sum of the prices of the constituent elements (e.g., input sources, sensor data, etc.), or based on a rule-based discount on the sum of the prices of the constituent elements, and the like. Rules and conditional logic may be applied, such as rules that consider cost factors (e.g., bandwidth and network usage, peak demand factors, scarcity factors, etc.), rules that consider usage parameters (e.g., the purpose, domain, user, role, duration, etc. of the package), and many other rules. In embodiments, cognitive data pricing system 4112 may include extensive cognitive intelligence features using genetic programming, etc., such as automatically changing pricing and tracking feedback on results (e.g., based on metrics from various financial domains, utilization metrics, etc., which may be provided by metrics calculated in analytics system 4018 on data from data trading system 4114).

[0189] Disclosed herein are methods and systems for self-organizing data pools, which can include self-organization of data pools based on utilization and / or revenue metrics, including utilization and / or revenue metrics tracked for multiple data pools. The data pool can initially include an unstructured or loosely structured data pool containing data from an industrial environment, such as sensor data from or related to industrial machines or components. For example, the data pool can ingest data streams from various machines or components in the environment (e.g., turbines, compressors, batteries, reactors, engines, motors, vehicles, pumps, rotors, axles, bearings, valves, and many other components), where the data streams contain analog and / or digital sensor data (of various types), published data about operating conditions, diagnostic and fault data, machine or component identification data, asset tracking data, and many other types of data. Each data stream can have an identifier in the pool, such as an identifier indicating its source and (optionally) type. The data pool can be accessed by external systems, for example, via one or more interfaces or APIs (e.g., RESTful APIs), or by data integration components such as gateways, proxies, bridges, connectors, etc., and the data pool can use similar functionality to access available data streams. The data pool can be managed by a self-organizing machine learning facility that can configure the data pool, for example, by managing the sources used for the pool, managing the available streams, and managing APIs or other connections to and from the data pool. Self-organization can receive feedback, for example, based on success metrics, utilization and benefit metrics (including consideration of the cost of acquiring and / or storing data and the benefits of the pool, as measured by profit or other metrics including user usefulness indicators, etc.). For example, a self-organizing data pool can identify that chemical and radiation data for an energy production environment is regularly accessed and extracted, while vibration and temperature data are not being used. In this case, the data pool can be automatically reorganized, for example, by ceasing to store vibration and / or temperature data, or by obtaining better sources for such data. This automatic reorganization can also be applied to data structures, for example, promoting different data types, different data sources, different data structures, etc. through gradual iteration and feedback.

[0190] In embodiments, a platform is provided that features self-organization of a data pool based on utilization and / or revenue metrics. In embodiments, data pool 4020 can be a self-organizing data pool 4020, such as one organized through cognitive capabilities, as described herein. Data pool 4020 can self-organize in response to learning feedback 4012, such as feedback based on metrics and results, including metrics and results calculated in analysis system 4018. Organization can include determining the data or data packets to be stored in the pool (e.g., representing a particular combination, arrangement, aggregation, etc.), the structure of such data (e.g., flat, hierarchical, linked, or other structure), storage duration, the nature of the storage media (e.g., hard drive, flash memory, SSDs, network-based storage, etc.), the arrangement of storage bits, and other parameters. The content and nature of the storage can vary, allowing data pool 4020 to learn and adapt, such as based on the state of host system 112, one or more data collection systems 102, storage environment parameters (e.g., capacity, cost, and performance factors), data collection environment parameters, market parameters, and many other parameters. In an embodiment, the data pool 4020 can learn and adapt, for example, by changing the above parameters and other parameters in response to benefit metrics such as return on investment, power utilization optimization, revenue optimization.

[0191] Disclosed herein are methods and systems for training AI models based on industry-specific feedback, including training AI models based on industry-specific feedback that reflects metrics of utilization, revenue, or impact, and wherein the AI ​​models operate on sensor data from industrial environments. As described above, these models can include models for industrial environments, machines, workflows, models for predicting states, models for predicting failures and optimizing maintenance, models for self-organizing storage (on devices, in data pools, and / or in the cloud), models for optimizing data transmission (e.g., for optimizing network coding, network condition-sensitive routing, etc.), models for optimizing data markets, and many other models.

[0192] In an embodiment, a platform is provided for training AI models based on industry-specific feedback. In an embodiment, various embodiments of the cognitive systems disclosed herein can obtain input and feedback from industry-specific and domain-specific sources 116, such as those related to the optimization of specific machines, devices, components, processes, etc. Thus, learning and adaptation of storage organization, network usage, combination of sensor and input data, data aggregation, data packaging, data pricing, and other features (e.g., for market 4102 or for other purposes of the host processing system 112) can be configured by learning domain-specific feedback metrics for a given environment or application, such as applications involving IoT devices, such as industrial environments. This can include efficiency optimization (e.g., in electrical, electromechanical, magnetic, physical, thermodynamic, chemical, and other processes and systems), output optimization (e.g., for generating energy, materials, products, services, and other outputs), fault prediction, avoidance, and mitigation (e.g., in the above systems and processes), performance metric optimization (e.g., return on investment, benefit, profit, gross profit, revenue, etc.), cost reduction (including labor costs, bandwidth costs, data costs, material input costs, licensing costs, etc.), benefit optimization (e.g., benefits related to safety, satisfaction, health), workflow optimization (e.g., optimizing time and resource allocation for processes), etc.

[0193] Disclosed herein are methods and systems for self-organizing swarms of industrial data collectors, including a self-organizing swarm of industrial data collectors that organizes the swarms based on the capabilities and status of the swarm members to optimize data collection. Each member in the swarm can be configured to have intelligence and the ability to coordinate with other members. For example, swarm members can track information about the data being processed by other members, so that data collection activities, data storage, data processing, and data publishing can be intelligently distributed across the swarm, taking into account environmental conditions, the capabilities of the swarm members, operating parameters, rules (e.g., from a rules engine that manages swarm operations), and the current status of the members. For example, among four collectors, one with a relatively low current power level (e.g., a low battery) may be temporarily assigned the role of publishing data because it may receive a certain amount of power from a reader or interrogation device (e.g., an RFID reader) when it needs to publish data. A second collector with a good power level and strong processing power may be assigned more complex functions, such as processing data, fusing data, organizing the rest of the swarm (including self-organization under machine learning to optimize the swarm over time, including by adjusting operating parameters, rules, etc. based on feedback). A third collector in the swarm, with significant storage capabilities, might be tasked with collecting and storing a class of data, such as vibration sensor data, which consumes significant bandwidth. A fourth collector in the swarm (e.g., one with lower storage capabilities) might be tasked with collecting data that is typically disposable, such as data about current diagnostic conditions, where only data about faults needs to be maintained and communicated. Swarm members can be connected in a peer-to-peer relationship, with one member acting as a "leader" or "hub," or in a tandem or ring connection, where each member passes data (including commands) to the next and understands the nature of the commands and capabilities appropriate to the previous and / or next member. Swarms can be used to distribute storage across the swarm (e.g., using each memory as an aggregate data store), such as to support a distributed ledger that can store transactional data, such as transactions involving data collected by the swarm, transactions generated in an industrial environment, and other data, such as data used to manage the swarm, the environment, or a machine or its components. The swarm can self-organize by providing machine learning capabilities on one or more members of the swarm or based on instructions from an external machine learning facility, which can optimize storage, data collection, data processing, data presentation, data transfer, and other functions based on managing parameters associated with each member. The machine learning facility can start with an initial configuration and change swarm parameters related to any of the above (including also changing the membership of the swarm), for example, based on providing feedback to the machine learning facility regarding success metrics (e.g., utilization metrics, efficiency metrics, predicted or expected state success metrics, productivity metrics, revenue metrics, profit metrics, etc.).Over time, the swarm can be optimized to an advantageous configuration to achieve a desired measure of success for the owner, operator, or host of the industrial environment or its machines, components, or processes.

[0194] In embodiments, a platform is provided that includes a self-organizing cluster of industrial data collectors. In embodiments, a host processing system 112, including its processing architecture 4024 (and optionally including or integrating with a cognitive data marketplace 4102), can be integrated with, connected to, or utilize data from a self-organizing cluster 4202 of data collectors 102. In embodiments, a self-organizing cluster 4202 can organize two or more data collection systems 102, such as by deploying cognitive features on one or more data collection systems 102, to provide collaboration across the cluster 4202. The cluster 4202 can be organized based on a hierarchical organization (e.g., where a master data collector 102 organizes and directs the activities of one or more slave data collectors 102), a collaborative organization (e.g., where organizational decisions for the cluster 4202 are distributed among the data collectors 102 (e.g., using various decision-making models such as voting systems, point systems, least-cost routing systems, priority systems, etc.), or the like. In embodiments, one or more data collectors 102 may have mobile capabilities, such as where the data collector is located on or in a mobile robot, drone, mobile submersible, or the like, such that the organization may include the location and positioning of the data collector 102. The data collection systems 102 may communicate with each other and with the host processing system 112, including sharing collectively allocated storage space, which may involve storage on or accessible by one or more collectors (in embodiments, the collectively allocated storage space may be considered unified storage space even if physically distributed, such as using virtualization capabilities). The organization may be automated based on one or more rules, models, conditions, processes, and the like, such as embodied or executed by conditional logic, and the organization may be managed by policies, such as handled by a policy engine. Rules may be based on industry, application, and domain specific objects, classes, events, workflows, processes, and systems, such as by setting up a group 4202 to collect selected types of data at specified locations and times, such as in coordination with the foregoing. For example, swarm 4202 can assign data collectors 102 to collect diagnostic, sensor, instrument, and / or telematics data, such as the time and location of inputs and outputs of each of the machines, in a serial manner, from each of a series of machines performing an industrial process (e.g., a robotic manufacturing process). In embodiments, self-organization can be cognitive, such as when the swarm changes one or more collection parameters over time and adapts the selection of parameters, the weights applied to the parameters, etc., such as in response to learning and feedback from learning feedback system 4012, etc., such as based on various feedback metrics that can be determined by applying analysis system 4018 (which, in embodiments, can reside on swarm 4202, host processing system 112, or a combination thereof) to data processed by swarm 4202 or other elements of the embodiments disclosed herein (including market elements, etc.).Thus, the swarm 4202 can exhibit adaptive behavior, such as adapting to the current state 4020 or the expected state of its environment (taking into account market behavior), the behavior of various objects (such as IoT devices, machines, components and systems), processes (including events, states, workflows, etc.), and other factors at a given time. Parameters may change (such as parameters implemented through genetic programming or other artificial intelligence-based techniques) during the process of change (such as neural networks, self-organizing maps, etc.), selection, promotion, etc. Parameters that can be managed, changed, selected and adjusted through cognitive machine learning may include storage parameters (location, type, duration, number, structure, etc. on the swarm 4202), network parameters (such as how the swarm 4202 is organized, such as mesh, peer, ring, serial, hierarchical and other network configurations and bandwidth utilization, data routing, network protocol selection, network coding type and other network parameters), security parameters (such as settings for various security applications and services), location and positioning parameters (such as routing the movement of the mobile data collector 102 to locations, relative to data collection points, relative to each other and relative to network availability). The feedback may include parameters such as the desired state and condition of the swarm 4202, an individual collection system 102, a host processing system 112, or one or more objects in the environment, events, etc. The feedback may be based on any of the types of feedback described herein so that the swarm can adapt its current and expected conditions over time to achieve various desired goals.

[0195] Disclosed herein are methods and systems for an Industrial Internet of Things (IIoT) distributed ledger, including a distributed ledger that supports tracking transactions performed on IIoT data in an automated data marketplace. The distributed ledger can distribute storage across devices using a secure protocol, such as a protocol used for cryptocurrencies, such as the blockchain used to support Bitcoin™. TMProtocol. A ledger or similar transaction record may include a structure in which each successive member of the chain stores data for previous transactions, and a competition may be established to determine the "best" structure (e.g., the most complete structure) among alternative data storage data structures. The ledger or similar transaction record may be stored on a data collector, industrial machine or component, data pool, data marketplace, cloud computing element, server, and / or enterprise IT infrastructure (e.g., an owner, operator, or host of an industrial environment or system disclosed herein). The ledger or transactions may be optimized through machine learning to provide storage efficiency, security, redundancy, etc.

[0196] In embodiments, the cognitive data marketplace 4102 can utilize a secure architecture to track and settle transactions, such as a distributed ledger 4004, in which transactions in data packets are tracked in a chained, distributed data structure (e.g., Blockchain™), thereby enabling forensic analysis and verification, with each device storing a portion of the ledger representing the transactions in the data packets. The distributed ledger 4004 can be distributed to IoT devices, data pools 4020, data collection systems 102, and the like, allowing for verification of transaction information without relying on a single central information repository. The transaction system 4114 can be configured to store data in the distributed ledger 4004 and retrieve data from it (as well as from constituent devices) to settle transactions. Thus, a distributed ledger 4004 is provided for processing data transactions, such as transactions for IoT data packets. In embodiments, a self-organizing storage system 4028 can be used to optimize the storage of distributed ledger data and to organize the storage of data packets (e.g., IoT data) that can be presented in the marketplace 4102.

[0197] Disclosed herein are methods and systems for self-organizing collectors, including self-organizing, multi-sensor data collectors that can optimize data collection, power, and / or revenue based on conditions in their environment. For example, a collector can organize data collection by turning specific sensors on and off, such as based on past utilization patterns or success metrics managed by an iterative configuration and a machine learning facility that tracks success measurements. For example, a multi-sensor collector can learn to turn off certain sensors when power levels are low or during periods of low data utilization from such sensors. Self-organization can also automatically organize how data is collected (which sensors, from what external source), how data is stored (at what granularity or compression level, duration, etc.), how data is presented (such as using a fused or multiplexed structure, a byte-like structure, or using intermediate statistical structures such as addition, subtraction, division, multiplication, square root, normalization, scaling, or other operations), and so on. This can be improved over time from an initial configuration by training the self-organizing facility based on data sets from real working environments, such as based on feedback metrics (including the various feedback types described throughout this invention).

[0198] Disclosed herein are methods and systems for network-sensitive collectors, including network-condition-sensitive, self-organizing, multi-sensor data collectors that can optimize based on bandwidth, quality of service, pricing, and / or other network conditions. Network sensitivity can include understanding the price of data transmission (e.g., allowing the system to receive or push data during off-peak hours or within the available parameters of a paid data plan), network quality (e.g., avoiding periods of potential errors), and environmental condition quality (e.g., delaying transmission until signal quality is good, e.g., when the collector emerges from a shielded environment to avoid wasting power while searching for a signal while being shielded by large metal structures, etc., typical of industrial environments).

[0199] Disclosed herein are methods and systems for remotely organizing a universal data collector that can power on and off sensor interfaces based on needs and / or conditions identified in an industrial data collection environment. For example, an interface can identify available sensors and can turn on an interface and / or processor to receive input from those sensors, including a hardware interface that allows sensors to be plugged into a data collector, a wireless data interface (e.g., the collector can ping the sensor, optionally providing some power via an interrogation signal), and a software interface, such as for processing specific types of data. Thus, a collector capable of processing a variety of data can be configured to suit specific uses in a given environment. In embodiments, configuration can be automatic or under machine learning, which can improve the configuration by optimizing parameters based on feedback metrics that change over time.

[0200] Disclosed herein are methods and systems for self-organizing storage of multi-sensor data collectors, including self-organizing storage of multi-sensor data collectors for industrial sensor data. Self-organizing storage can allocate storage based on the application of machine learning, which can improve storage configuration based on feedback measurements that change over time. Storage can be optimized by configuring the type of data used (e.g., a byte-like structure, a structure representing fused data from multiple sensors, a structure representing statistics or metrics calculated by applying mathematical functions to the data, etc.), by configuring compression, by configuring data storage duration, by configuring write strategies (e.g., by splitting data across multiple storage devices using a protocol in which one device stores instructions for other devices in the chain, etc.), and by configuring a storage hierarchy (e.g., by providing pre-computed intermediate statistics for faster access to frequently accessed data items). Thus, a highly intelligent storage system can be configured and optimized based on feedback over time.

[0201] Disclosed herein are methods and systems for self-organizing network coding for multi-sensor data networks, including self-organizing network coding for data networks that transmit data from multiple sensors in an industrial data collection environment. Network coding, including random linear network coding, enables efficient and reliable transmission of large amounts of data over various networks. Different network coding configurations can be selected based on machine learning to optimize network coding and other network transmission characteristics based on network conditions, environmental conditions, and other factors (e.g., the nature of the data being transmitted, environmental conditions, operating conditions, etc.) (including by training a network coding selection model over time based on feedback from a success metric, such as any of the metrics described herein).

[0202] In an embodiment, a platform with self-organizing network coding for multi-sensor data networking is provided. The cognitive system can change one or more parameters for networking, such as network type selection (e.g., selecting among available local, cellular, satellite, WiFi, Bluetooth, NFC, Zigbee, and other networks), network selection (e.g., selecting a specific network, such as a network known to have desired security characteristics), network coding selection (e.g., selecting a network coding type for efficient transmission), network timing selection (e.g., configuring transmission based on network pricing conditions, traffic volume, etc.), network feature selection (e.g., selecting cognitive features, security features, etc.), network conditions (e.g., network quality based on current environmental or operating conditions), network feature selection (e.g., enabling available authentication, licensing, and similar systems), network protocol selection (e.g., selecting between HTTP, IP, TCP / IP, cellular, satellite, serial, packet, streaming, and many other protocols), etc. Selecting the optimal network configuration can be complex and depends on the specific situation, taking into account bandwidth limitations, price changes, sensitivity to environmental factors, security considerations, etc. The self-organizing network system 4030 can change the combination and arrangement of these parameters while obtaining input from the learning feedback system 4012 (such as using information from the analysis system 4018 regarding various result metrics (such as overall system metrics, analysis success metrics and local performance indicators), information from various sensors and input sources 116, information about the status of the state system 4020 (including events, environmental conditions, working conditions and others), or other information or obtaining other input. By changing and selecting alternative configurations of network parameters in different states, the self-organizing network system can find a configuration that is fully adapted to the environment monitored or controlled by the host system 112, such as an example of the location of one or more data collection systems 102, and is fully adapted to emerging network conditions. Therefore, a self-organizing, network condition adaptive data collection system is provided.

[0203] Referring to Figure 42, the data collection system 102 can have one or more output interfaces and / or ports 4010. These interfaces and / or ports can include network ports and connections, application programming interfaces, etc. Disclosed herein is a method and system for a tactile or multi-sensory user interface, including a wearable tactile or multi-sensory user interface for an industrial sensor data collector having vibration, thermal, electrical, and / or sound outputs. For example, based on a data structure configured to support the interface, the interface can be configured to provide input or feedback to the user, such as based on data from sensors in the environment. For example, if a fault condition based on vibration data is detected (e.g., a fault condition caused by bearing wear, shaft misalignment, or a resonance condition between machines), it can be presented in the tactile interface by vibrating the interface, such as by shaking a wrist-worn device. Similarly, thermal data indicating overheating can be presented by heating or cooling the wearable device, for example, when a worker is working on a machine and is not necessarily able to view the user interface. Similarly, electrical or magnetic data can be presented by a humming sound, etc., to indicate the presence of an open electrical connection or wire, etc. That is, multi-sensory interfaces can intuitively help users, for example, users with wearable devices get a quick indication of what is happening in the environment, and wearable interfaces have various interaction modes that do not require users to pay attention to the graphical UI, which may be very difficult or impossible in many industrial environments where users need to pay attention to the environment.

[0204] In one embodiment, a platform is provided that includes a wearable haptic user interface for industrial sensor data collectors, the interface having vibration, thermal, electrical, and / or acoustic output. In one embodiment, a haptic user interface 4302 is provided as an output of a data collection system 102, such as a system that processes information from the vibration, thermal, electrical, and / or acoustic output and provides it to one or more components of the data collection system 102 or other systems, such as a wearable device, a mobile phone, etc. The data collection system 102 can be provided in a form factor suitable for delivering haptic input, such as vibration, heating or cooling, buzzing, etc., to a user, such as in a headgear, armband, wristband or watch, belt, clothing, uniform, etc. In such cases, the data collection system 102 can be integrated with equipment, uniform, equipment, etc., worn by a user (e.g., an individual responsible for operating or monitoring an industrial environment). In one embodiment, signals from various sensors or input sources (or a selective combination, permutation, blend, etc., managed by one or more of the cognitive input selection systems 4004, 4014) can trigger haptic feedback. For example, if a nearby industrial machine is overheating, the haptic interface can warn the user by increasing its temperature or by sending a signal to another device (e.g., a mobile phone) to cause it to increase its temperature. If the system is experiencing abnormal vibrations, the haptic interface may vibrate. Thus, through various forms of haptic input, the data collection system 102 can notify the user that attention is needed for one or more devices, machines, or other factors, such as those in an industrial environment, without requiring the user to read the message or divert their visual attention from the task at hand. The haptic interface and the output selection that should be provided can be considered in the cognitive input selection systems 4004 and 4014. For example, user behavior (e.g., responses to inputs) can be monitored and analyzed in the analysis system 4018, and feedback can be provided by the learning feedback system 4012 so that signals can be provided at the right time and in the right manner based on the correct collection or packaging of sensors and inputs to optimize the effectiveness of the haptic system 4202. This can include rule-based or model-based feedback (e.g., providing output that corresponds in some logical manner to the source data being transmitted). In embodiments, a cognitive haptic system may be provided in which the selection of inputs or triggers for haptic feedback, the selection of outputs, timing, intensity levels, durations, and other parameters (or weights applied thereto) may be varied (e.g., using genetic programming) over a process of variation, upgrading, and selection based on real-world feedback in response to feedback in real situations or based on feedback from simulations and testing of user behavior. Thus, an adaptive haptic interface for the data collection system 102 is provided that can learn and adjust feedback to meet relevant requirements and optimize the impact on user behavior, e.g., with respect to overall system results, data collection results, analysis results, etc.

[0205] Disclosed herein are methods and systems for an AR / VR industrial eyewear presentation layer, wherein heat map elements are presented based on patterns and / or parameters in the collected data. Disclosed herein are methods and systems for condition-sensitive, self-organizing adjustment of an AR / VR interface based on feedback metrics and / or training in an industrial environment. In embodiments, any data, metrics, etc. described throughout this invention may be presented by visual elements, overlays, etc., to be presented in an AR / VR interface, for example, in industrial eyewear, on an AR / VR interface on a smartphone or tablet, on an AR / VR interface on a data collector (which may be embodied in a smartphone or tablet), on a display on a machine or component, and / or on a display in an industrial environment.

[0206] In embodiments, a platform is provided with a heat map displaying AR / VR collected data. In embodiments, a platform is provided with a heat map 4204 displaying data collected from a data collection system 102 for providing input to an AR / VR interface 4208. In embodiments, a heat map interface 4304 is provided as an output of the data collection system 102, for example, for processing and providing visualizations of various sensor data and other data (e.g., map data, simulated sensor data, and other data) to one or more components of the data collection system 102 or other systems, such as mobile devices, tablets, dashboards, computers, AR / VR devices, etc. The data collection system 102 can be provided in a form factor suitable for delivering visual input to a user, such as presenting a map that includes indicators of levels of analog and digital sensor data, such as data indicating levels of rotation, vibration, heating or cooling, pressure, and many other conditions. In this case, the data collection system 102 can be integrated with devices used by individuals responsible for operating or monitoring industrial environments, etc. In embodiments, signals from various sensors or input sources (or selective combinations, permutations, blends, etc., managed by one or more of cognitive input selection systems 4004 and 4014) can provide input data to a heat map. Coordinates can include real-world location coordinates (e.g., geographic locations or locations on a map of the environment) as well as other coordinates, such as time-based coordinates, frequency-based coordinates, or other coordinates that allow for representation of analog sensor signals, digital signals, input source information, and various combinations thereof in a map-based visualization, such that colors can represent different input levels along relevant dimensions. For example, if a nearby industrial machine is overheating, the heat map interface can alert the user by displaying the machine in bright red. If the system is experiencing abnormal vibration, the heat map interface might display a different color for the machine's visual elements or might cause an icon or display element representing the machine in the interface to vibrate, drawing attention to that element. Clicking, touching, or otherwise interacting with the map can allow the user to gain insight into the underlying sensor or input data used as input to the heat map display. Thus, through various forms of display, the data collection system 102 can notify the user that one or more devices, machines, or other factors, such as those in an industrial environment, require attention, without requiring them to read text-based messages or inputs. The heat map interface and the output selections that should be provided can be considered in the cognitive input selection system 4004, 4014. For example, user behavior (e.g., responses to inputs or displays) can be monitored and analyzed in the analysis system 4018, and feedback can be provided through the learning feedback system 4012 so that based on the correct collection or packaging of sensors and inputs, signals can be provided at the right time and in the right manner to optimize the effectiveness of the heat map UI 4304. This can include rule-based or model-based feedback (e.g., providing output that corresponds in some logical way to the source data being transmitted).In embodiments, a cognitive heatmap system may be provided in which the selection of inputs or triggers, selection of outputs, colors, visual representation elements, timing, intensity levels, durations, and other parameters of a heatmap display (or weights applied thereto) can be varied (e.g., using genetic programming) in a process of variation, upgrading, and selection based on real-world feedback in response to feedback from actual situations or feedback based on simulations and test results of user behavior. Thus, an adaptive heatmap interface for the data collection system 102 or data collected thereby or processed by the host processing system 112 is provided that can learn and adapt feedback to meet relevant requirements and optimize the impact on user behavior and reactions, e.g., with respect to overall system results, data collection results, analysis results, etc.

[0207] In embodiments, a platform is provided having automatically adjustable AR / VR visualizations of data collected by a data collector. In embodiments, a platform is provided having automatically adjustable AR / VR visualization system 4308 for visualizing data collected by a data collection system 102, e.g., where the data collection system 102 has an AR / VR interface 4208 or provides input to an AR / VR interface 4308 (e.g., a mobile phone in virtual reality, or an AR headset, AR glasses, etc.). In embodiments, an AR / VR system 4308 is provided as an output interface for the data collection system 102, e.g., a system for processing and providing visualizations of various sensor data and other data (e.g., map data, simulated sensor data, and other data) to one or more components of the data collection system 102 or other systems, e.g., a mobile device, tablet, dashboard, computer, AR / VR device, etc. The data collection system 102 may be provided in a form factor suitable for delivering AR or VR visual, auditory, or other sensory input to a user, such as by presenting one or more displays, such as 3D reality visualizations, objects, maps, camera overlays or other overlay elements, maps, etc., that include or correspond to indicators of levels of analog and digital sensor data, such as data indicating levels of rotation, vibration, heating or cooling, pressure, or other conditions, relative to an input source 116, etc. In such a case, the data collection system 102 may be integrated with devices used by individuals responsible for operating or monitoring the industrial environment, etc.

[0208] In embodiments, signals from various sensors or input sources (or selective combinations, permutations, blends, etc., managed by one or more of the cognitive input selection systems 4004, 4014) can provide input data to populate, configure, modify, or otherwise determine AR / VR elements. Visual elements can include various icons, map elements, menu elements, sliders, triggers, colors, shapes, sizes, and the like to represent analog sensor signals, digital signals, input source information, and various combinations. For example, the color, shape, and size of a visual overlay element can represent different input levels along the relevant dimension of a sensor or sensor combination. For example, if a nearby industrial machine is overheating, an AR element can alert the user by displaying an icon representing that type of machine in a flashing red in a portion of a pair of AR glasses' display. If a system is experiencing abnormal vibration, a VR interface displaying a visualization of the machine's components (e.g., a camera view of the machine overlaid with a 3D visualization element) can display the vibrating component with highlight color and motion to ensure that the component stands out in the VR environment used to help the user monitor or maintain the machine. Clicking, touching, moving the eyes toward, or otherwise interacting with visual elements in an AR / VR interface can allow the user to gain insight into the underlying sensor or input data used as input to the display. Thus, through various forms of display, the data collection system 102 can notify the user that one or more devices, machines, or other factors require attention, such as in an industrial environment, without requiring the user to read text-based messages or input or divert attention from the applicable environment (whether a real environment with AR features or a virtual environment for simulation, training, etc.).

[0209] The selection and configuration of the AR / VR output interface 4208 and the outputs or displays that should be provided can be handled in the cognitive input selection system 4004, 4014. For example, user behavior (e.g., responses to inputs or displays) can be monitored and analyzed in the analysis system 4018, and feedback can be provided through the learning feedback system 4012 so that AR / VR display signals can be provided at the right time and in the right manner based on the correct collection or packaging of sensors and inputs to optimize the effectiveness of the AR / VR UI 4308. This can include rule-based or model-based feedback (e.g., providing outputs that correspond in some logical manner to the source data being transmitted). In embodiments, a cognitively regulated AR / VR interface control system 4308 can be provided, in which the selection of inputs or triggers for AR / VR display elements, the selection of outputs (e.g., colors, visual representation elements, timing, intensity levels, durations, and other parameters (or weights applied thereto)), and other parameters of the VR / AR environment can be varied, updated, and selected (e.g., using genetic programming) based on real-world feedback in response to feedback from actual situations or based on feedback from simulations and testing of user behavior. Thus, an adaptively regulated AR / VR interface is provided for the data collection system 102 or data collected thereby or data processed by the host processing system 112 that can learn and adjust feedback to meet relevant requirements and optimize the impact on user behavior and reactions, such as with respect to overall system results, data collection results, analysis results, etc.

[0210] As described above, disclosed herein are methods and systems for continuous ultrasonic monitoring, including providing continuous ultrasonic monitoring of rotating elements and bearings in energy production facilities. Embodiments include using continuous ultrasonic monitoring of an industrial environment as a source for a cloud-deployed pattern recognizer. Embodiments include using continuous ultrasonic monitoring to provide updated state information to a state machine, which serves as input to the cloud-deployed pattern recognizer. Embodiments include providing continuous ultrasonic monitoring information to users based on policies declared in a policy engine. Embodiments include storing continuous ultrasonic monitoring data along with other data in a fused data structure on industrial sensor devices. Embodiments include making continuous ultrasonic monitoring data streams from an industrial environment available as services in a data marketplace. Embodiments include feeding continuous ultrasonic data streams into a self-organizing data pool. Embodiments include training a machine learning model to monitor the continuous ultrasonic monitoring data stream, wherein the model is based on a training set created from manual analysis of such data streams and is improved based on performance data collected in the industrial environment. Embodiments include a data collector cluster comprising at least one data collector for performing continuous ultrasonic monitoring of an industrial environment and at least one other type of data collector. Embodiments include using a distributed ledger to store time series data from continuous ultrasonic monitoring across multiple devices. Embodiments include collecting continuous ultrasonic monitoring data streams in an ad hoc data collector. Embodiments include collecting continuous ultrasonic monitoring data streams in a network-aware data collector.

[0211] Embodiments include collecting a continuous ultrasonic monitoring data stream in a remotely organized data collector. Embodiments include collecting a continuous ultrasonic monitoring data stream in a data collector with self-organizing storage. Embodiments include transmitting an ultrasonic data stream collected from an industrial environment using self-organizing network coding. Embodiments include transmitting parameter indicators of the continuously collected ultrasonic data stream via a sensor interface of a wearable device. Embodiments include transmitting parameter indicators of the continuously collected ultrasonic data stream via a heat map visual interface of a wearable device. Embodiments include transmitting parameter indicators of the continuously collected ultrasonic data stream via an interface that operates through self-organizing regulation of an interface layer.

[0212] As described above, disclosed herein are methods and systems for cloud-based machine pattern recognition based on remote simulated industrial sensor fusion. Embodiments include acquiring input from multiple simulated sensors deployed in an industrial environment, multiplexing the sensors into a multiplexed data stream, feeding the data stream into a cloud-deployed machine learning facility, and training a model of the machine learning facility to recognize defined patterns associated with the industrial environment. Embodiments include using a cloud-based pattern recognizer on input states from a state machine representing the state of the industrial environment. Embodiments include deploying policies via a policy engine that governs which users can use which data and for what purposes in the cloud-based machine learning. Embodiments include feeding input from multiple devices with fusion of multiple sensor streams and on-device storage into the cloud-based pattern recognizer. Embodiments include making output from the cloud-based machine pattern recognizer, which analyzes fused data from remote simulated industrial sensors, available as a data service in a data marketplace. Embodiments include using a cloud-based platform to identify data patterns in multiple data pools containing data published by industrial sensors. Embodiments include training a model to identify a preferred set of sensors to diagnose a condition of an industrial environment, wherein a training set is created by a human user, and the model is improved based on collected data feedback regarding the condition of the industrial environment.

[0213] Embodiments include a swarm of data collectors controlled by policies that are automatically propagated through the swarm. Embodiments include storing sensor fusion information on multiple devices using a distributed ledger. Embodiments include feeding input from a set of self-organizing data collectors to a cloud-based pattern recognizer that utilizes data from multiple sensors for an industrial environment. Embodiments include feeding input from a set of network-aware data collectors to a cloud-based pattern recognizer that utilizes data from multiple sensors for an industrial environment. Embodiments include feeding input from a set of remotely organized data collectors to a cloud-based pattern recognizer that utilizes data from multiple sensors for an industrial environment.

[0214] Embodiments include feeding inputs from a set of data collectors with self-organizing storage to a cloud-based pattern recognizer that utilizes data from multiple sensors in an industrial environment. Embodiments include a data collection system in an industrial environment with self-organizing network coding for data transmission of fused data from multiple sensors in the environment. Embodiments include transmitting information formed by fusing inputs from multiple sensors in the industrial data collection system in a multi-sensor interface. Embodiments include transmitting information formed by fusing inputs from multiple sensors in the industrial data collection system in a heat map interface. Embodiments include transmitting information formed by fusing inputs from multiple sensors in the industrial data collection system in an interface that operates with self-organizing adjustments at the interface layer.

[0215] As described above, disclosed herein are methods and systems for performing cloud-based machine pattern analysis on state information from multiple simulated industrial sensors to provide expected state information for an industrial system. Embodiments include providing cloud-based pattern analysis of state information from multiple simulated industrial sensors to provide expected state information for an industrial system. Embodiments include using a policy engine to determine state information that can be used for cloud-based machine analysis. Embodiments include feeding input from multiple devices, fused with multiple sensor streams and stored on-device, into a cloud-based pattern recognizer to determine an expected state of an industrial environment. Embodiments include making the expected state information from the cloud-based machine pattern recognizer available as a data service in a data marketplace, where the cloud-based machine pattern recognizer analyzes fused data from remote simulated industrial sensors. Embodiments include using the cloud-based pattern recognizer to determine the expected state of an industrial environment based on data collected from a data pool containing information streams from machines in the environment. Embodiments include training a model to identify preferred state information for diagnosing conditions in the industrial environment, wherein the training set is created by a human user, and the model is improved based on collected data feedback regarding the state of the industrial environment. Embodiments include a cluster of data collectors that feeds a state machine that maintains current state information for the industrial environment. Embodiments include using a distributed ledger to store historical state information of fused sensor states, with a self-organizing data collector feeding into a state machine that maintains current state information of an industrial environment. Embodiments include a network-aware data collector that feeds into a state machine that maintains current state information of an industrial environment. Embodiments include a remotely organized data collector that feeds into a state machine that maintains current state information of an industrial environment. Embodiments include a data collector with self-organizing storage that feeds into a state machine that maintains current state information of an industrial environment. Embodiments include a data collection system in an industrial environment having self-organizing network coding for data transmission and maintaining expected state information of the environment. Embodiments include transmitting expected state information in an industrial data collection system determined by machine learning in a multi-sensor interface. Embodiments include transmitting expected state information in an industrial data collection system determined by machine learning in a heat map interface. Embodiments include transmitting expected state information in an industrial data collection system determined by machine learning in an interface that operates through self-organizing regulation of an interface layer.

[0216] As described above, this document discloses methods and systems for a cloud-based policy automation engine for the Internet of Things, wherein the creation, deployment, and management of IoT devices (including a cloud-based policy automation engine for the Internet of Things) enables the creation, deployment, and management of policies applicable to IoT devices. Embodiments include deploying data usage policies to an on-device storage system storing fused data from multiple industrial sensors. Embodiments include deploying policies regarding who can provide what data in an ad hoc marketplace for IoT sensor data. Embodiments include deploying policies across a set of ad hoc data pools containing data streamed from industrial sensor devices to govern data usage in the data pools. Embodiments include training a model to determine policies that should be deployed in an industrial data collection system. Embodiments include deploying policies that govern how ad hoc groups should be organized for a specific industrial environment. Embodiments include storing policies on a device that govern the use of the device's storage capabilities for a distributed ledger. Embodiments include deploying policies that govern how ad hoc data collectors should be organized for a specific industrial environment. Embodiments include deploying policies that govern how network-sensitive data collectors should use network bandwidth for a specific industrial environment. Embodiments include deploying policies that govern how a remotely organized data collector should collect data about a specified industrial environment and make the data available. Embodiments include deploying policies that govern how a data collector should self-organize storage of a particular industrial environment. Embodiments include a data collection system in an industrial environment having a policy engine for deploying policies within the system and self-organizing network coding for data transmission. Embodiments include a data collection system in an industrial environment having a policy engine for deploying policies within the system, wherein a policy is applied to how data will be presented in a multi-sensor interface. Embodiments include a data collection system in an industrial environment having a policy engine for deploying policies within the system, wherein a policy is applied to how data will be presented in a heat map visual interface. Embodiments include a data collection system in an industrial environment having a policy engine for deploying policies within the system, wherein the policy is applied to how data will be presented in an interface that operates with self-organization regulation of the interface layer.

[0217] As described above, disclosed herein are methods and systems for on-device sensor fusion and data storage for Industrial IoT devices, including on-device sensor fusion and data storage for Industrial IoT devices, wherein data from multiple sensors is multiplexed and processed on a device for storing fused data streams. Embodiments include presenting an ad hoc marketplace for fused sensor data extracted from on-device storage of IoT devices. Embodiments include streaming fused sensor information from multiple industrial sensors and from an on-device data storage facility to a data pool. Embodiments include training a model to determine data that should be stored on devices in a data collection environment. Embodiments include a self-organizing swarm of industrial data collectors that are organized to optimize data collection, wherein at least some data collectors have on-device storage of fused data from multiple sensors. Embodiments include storing distributed ledger information along with the fused sensor information on Industrial IoT devices. Embodiments include on-device sensor fusion and data storage for self-organizing industrial data collectors. Embodiments include on-device sensor fusion and data storage for network-aware industrial data collectors. Embodiments include on-device sensor fusion and data storage for remotely organized industrial data collectors. Embodiments include on-device sensor fusion and ad hoc data storage for industrial data collectors. Embodiments include a data collection system in an industrial environment that includes on-device sensor fusion and ad hoc network coding for data transmission. Embodiments include a data collection system with on-device sensor fusion for industrial sensor data, wherein data structures are stored to support an alternative multi-sensor presentation mode. Embodiments include a data collection system with on-device sensor fusion for industrial sensor data, wherein data structures are stored to support a visual heat map presentation mode. Embodiments include a data collection system with on-device sensor fusion for industrial sensor data, wherein data structures are stored to support an interface that self-organizes and regulates operations through an interface layer.

[0218] As described above, disclosed herein are methods and systems for a self-organizing data marketplace for industrial IoT data, including a self-organizing data marketplace for industrial IoT data, wherein available data elements are organized in the marketplace for consumption by consumers based on training a self-organizing facility using a training set and feedback from marketplace success metrics. Embodiments include organizing a set of data pools in the self-organizing data marketplace based on utilization metrics of the data pools. Embodiments include training a model to determine pricing for data in the data marketplace. Embodiments include feeding the data marketplace with data streams from a self-organizing group of industrial data collectors. Embodiments include using a distributed ledger to store transaction data for the self-organizing market for industrial IoT data. Embodiments include feeding the data marketplace with data streams from self-organizing industrial data collectors. Embodiments include feeding the data marketplace with data streams from a group of network-aware industrial data collectors. Embodiments include feeding the data marketplace with data streams from a group of remotely organized industrial data collectors. Embodiments include feeding the data marketplace with data streams from a group of industrial data collectors with self-organizing storage. Embodiments include using self-organizing network coding to transmit sensor data collected in an industrial environment to the marketplace. Embodiments include providing a data structure library suitable for presenting data in an alternative multi-sensor interface mode in the data marketplace. Embodiments include providing a data structure library in a data marketplace adapted to present data in a heat map visualization. Embodiments include providing a data structure library in a data marketplace adapted to present data in an interface that operates through self-organizing regulation of an interface layer.

[0219] As described above, disclosed herein are methods and systems for self-organizing data pools, including self-organizing data pools based on utilization and / or revenue metrics, including utilization and / or revenue metrics tracked for multiple data pools. Embodiments include training a model to present the most valuable data in a data marketplace, wherein the training is based on industry-specific success metrics. Embodiments include populating a set of self-organizing data pools with data from a self-organizing group of data collectors. Embodiments include using a distributed ledger to store transaction information for data deployed in the data pools, wherein the distributed ledger is distributed across the data pools. Embodiments include self-organizing data pools based on utilization and / or revenue metrics tracked for multiple data pools, wherein the data pools contain data from self-organizing data collectors. Embodiments include populating a set of self-organizing data pools with data from a set of network-aware data collectors. Embodiments include populating a set of self-organizing data pools with data from a set of remotely organized data collectors. Embodiments include populating a set of self-organizing data pools with data from a set of data collectors with self-organizing storage. Embodiments include a data collection system in an industrial environment, the system having self-organizing pools for data storage and self-organizing network coding for data transmission. Embodiments include a data collection system in an industrial environment having self-organizing pools for data storage, the self-organizing pools including source data structures for supporting data presentation in a multi-sensor interface. Embodiments include a data collection system in an industrial environment having self-organizing pools for data storage, the self-organizing pools including source data structures for supporting data presentation in a heat map interface. Embodiments include a data collection system in an industrial environment having self-organizing pools for data storage, the self-organizing pools including source data structures for supporting data presentation in an interface that operates with self-organizing adjustments of an interface layer.

[0220] As described above, disclosed herein are methods and systems for training AI models based on industry-specific feedback, including training AI models based on industry-specific feedback reflecting metrics of utilization, benefit, or impact, wherein the AI ​​models operate on sensor data from an industrial environment. Embodiments include training a swarm of data collectors based on industry-specific feedback metrics. Embodiments include training an AI model to identify and utilize available storage locations in an industrial environment for storing distributed ledger information. Embodiments include training a swarm of self-organizing data collectors based on industry-specific feedback metrics. Embodiments include training a network-aware data collector based on network and industrial conditions in an industrial environment. Embodiments include training a remote manager for a remotely organized data collector based on industry-specific feedback metrics. Embodiments include training a self-organizing data collector to configure storage based on industry-specific feedback. Embodiments include a data collection system in an industrial environment having cloud-based training for a network coding model used to organize data transmission. Embodiments include a data collection system in an industrial environment that performs cloud-based training for facilities that manage data presentation in a multi-sensor interface. Embodiments include a data collection system in an industrial environment that performs cloud-based training for facilities that manage data presentation in a heat map interface. Embodiments include a data collection system in an industrial environment that provides cloud-based training for facilities managing the presentation of data in an interface that operates through self-organizing regulation of an interface layer.

[0221] As described above, disclosed herein are methods and systems for self-organizing swarms of industrial data collectors, including self-organizing swarms of industrial data collectors that organize among the industrial data collectors based on the capabilities and status of the swarm members to optimize data collection. Embodiments include deploying a distributed ledger data structure in the data swarm. Embodiments include self-organizing swarms of self-organizing data collectors for data collection in an industrial environment. Embodiments include self-organizing swarms of network-sensitive data collectors for data collection in an industrial environment. Embodiments include self-organizing swarms of network-sensitive data collectors for data collection in an industrial environment, wherein the self-organizing pool is also used for remote organization. Embodiments include self-organizing swarms of data collectors with self-organizing storage for data collection in an industrial environment. Embodiments include a data collection system in an industrial environment having a self-organizing swarm of data collectors and self-organizing network coding for data transmission. Embodiments include a data collection system in an industrial environment having a self-organizing swarm of data collectors that forwards information for use in a multi-sensor interface. Embodiments include a data collection system in an industrial environment having a self-organizing swarm of data collectors that forwards information for use in a heat map interface. Embodiments include a data collection system in an industrial environment having an ad hoc group of data collectors that forward information for use in an interface that regulates operation through self-organization of an interface layer.

[0222] As described above, this document discloses methods and systems for an Industrial Internet of Things distributed ledger, including a distributed ledger that supports tracking transactions performed on Industrial IoT data in an automated data marketplace. Embodiments include a self-organizing data collector that is configured to distribute collected information to a distributed ledger. Embodiments include a network-sensitive data collector that is configured to distribute collected information to a distributed ledger based on network conditions. Embodiments include a remotely organized data collector that is configured to distribute collected information to a distributed ledger based on intelligent remote management of distribution. Embodiments include a data collector with self-organizing local storage that is configured to distribute collected information to a distributed ledger. Embodiments include a data collection system in an industrial environment that uses a distributed ledger for data storage and self-organizing network coding for data transmission. Embodiments include a data collection system in an industrial environment that uses a distributed ledger to store data for a data structure that supports a tactile interface for data presentation. Embodiments include a data collection system in an industrial environment that uses a distributed ledger to store data for a data structure that supports a heat map interface for data presentation. Embodiments include a data collection system for an industrial environment that uses a distributed ledger to store data in a data structure that supports an interface that regulates operations through self-organization of an interface layer.

[0223] As described above, disclosed herein are methods and systems for self-organizing collectors, including self-organizing, multi-sensor data collectors that can optimize data collection, power, and / or revenue based on conditions in their environment. Embodiments include self-organizing data collectors that organize based at least in part on network conditions. Embodiments include self-organizing data collectors that are also responsive to remote organization. Embodiments include self-organizing data collectors with self-organizing storage for collected data in an industrial data collection environment. Embodiments include data collection systems in industrial environments with self-organizing data collection and self-organizing network coding for data transmission. Embodiments include a data collection system in an industrial environment with a self-organizing data collector that feeds a data structure supporting a tactile or multi-sensory wearable interface for data presentation. Embodiments include a data collection system in an industrial environment with a self-organizing data collector that feeds a data structure supporting a heatmap interface for data presentation. Embodiments include a data collection system in an industrial environment with a self-organizing data collector that feeds a data structure supporting an interface for data presentation that operates through self-organizing regulation of an interface layer.

[0224] As described above, disclosed herein are methods and systems for network-sensitive collectors, including network condition-sensitive, self-organizing multi-sensor data collectors that can optimize based on bandwidth, quality of service, price, and / or other network conditions. Embodiments include a remotely organized, network condition-sensitive universal data collector that can power on and off sensor interfaces based on needs and / or conditions (including network conditions) identified in an industrial data collection environment. Embodiments include a network condition-sensitive data collector having self-organizing storage for data collected in an industrial data collection environment. Embodiments include a network condition-sensitive data collector having self-organizing network coding for data transmission in an industrial data collection environment. Embodiments include a data collection system in an industrial environment having a network-sensitive data collector that forwards data structures supporting a tactile wearable interface for data presentation. Embodiments include a data collection system in an industrial environment having a network-sensitive data collector that forwards data structures supporting a heat map interface for data presentation. Embodiments include a data collection system in an industrial environment having a network-sensitive data collector that transmits data structures supporting an interface for self-organizing regulation of operations through an interface layer.

[0225] As described above, disclosed herein are methods and systems for remotely organizing a universal data collector that can power on and off sensor interfaces based on needs and / or conditions identified in an industrial data collection environment. Embodiments include a remotely organizing universal data collector having self-organizing storage for data collected in an industrial data collection environment. Embodiments include a data collection system in an industrial environment having data collection remote control and self-organizing network coding for data transmission. Embodiments include a remotely organizing data collector for storing sensor data and transmitting instructions for using the data in a tactile or multi-sensory wearable interface. Embodiments include a remotely organizing data collector for storing sensor data and transmitting instructions for using the data in a heat map visual interface. Embodiments include a remotely organizing data collector for storing sensor data and transmitting instructions for using the data in an interface that operates with self-organizing regulation of the interface layer.

[0226] As described above, disclosed herein are methods and systems for self-organizing storage of multi-sensor data collectors, including self-organizing storage of multi-sensor data collectors for industrial sensor data. Embodiments include a data collection system in an industrial environment having self-organizing data storage and self-organizing network coding for data transmission. Embodiments include a data collector with self-organizing storage for storing sensor data and instructions for converting the data for use in a tactile wearable interface. Embodiments include a data collector with self-organizing storage for storing sensor data and instructions for converting the data for use in a heat map presentation interface. Embodiments include a data collector with self-organizing storage for storing sensor data and instructions for converting the data for use in an interface that operates in or with self-organizing regulation of an interface layer.

[0227] As described above, disclosed herein are methods and systems for self-organizing network coding for multi-sensor data networks, including self-organizing network coding for data networks for transmitting multiple sensor data in an industrial data collection environment. Embodiments include a data collection system in an industrial environment having self-organizing network coding for data transmission and a data structure supporting a tactile wearable interface for data presentation. Embodiments include a data collection system in an industrial environment having self-organizing network coding for data transmission and a data structure supporting a heat map interface for data presentation. Embodiments include a data collection system in an industrial environment having self-organizing network coding for data transmission and self-organizing regulation of an interface layer for data presentation.

[0228] As described above, disclosed herein are methods and systems for tactile or multi-sensory user interfaces, including wearable tactile or multi-sensory user interfaces for industrial sensor data collectors having vibration, thermal, electrical, and / or acoustic outputs. Embodiments include wearable tactile user interfaces for transmitting industrial status information from a data collector having vibration, thermal, electrical, and / or acoustic outputs. Embodiments include wearable tactile user interfaces for transmitting industrial status information from a data collector having vibration, thermal, electrical, and / or acoustic outputs, wherein the wearable interface further has a visual presentation layer for presenting a heat map indicating data parameters. Embodiments include condition-sensitive self-organizing regulation of AR / VR interfaces and multi-sensory interfaces based on feedback metrics and / or training in industrial environments.

[0229] As described above, disclosed herein are methods and systems for an AR / VR industrial eyewear presentation layer, wherein heatmap elements are presented based on patterns and / or parameters in collected data. Embodiments include condition-sensitive self-organizing adjustments to heatmap AR / VR interfaces based on feedback metrics and / or training in industrial environments. As described above, disclosed herein are methods and systems for condition-sensitive self-organizing adjustments to AR / VR interfaces based on feedback metrics and / or training in industrial environments.

[0230] The following illustrative clauses describe certain embodiments of the present invention. The data collection system referred to in the following invention can be a local data collection system 102, a host processing system 112 (e.g., using a cloud platform), or a combination of a local system and a host system. In an embodiment, a data collection system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has IP front-end signal conditioning on the multiplexer to improve the signal-to-noise ratio. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has a multiplexer that continuously monitors alarm features. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and uses a distributed CPLD chip with a dedicated bus to provide logic control for multiple MUXs and data acquisition components. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has high current input capability using solid-state relays and a design topology. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has power-off capability for at least one analog sensor channel and component boards. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has unique electrostatic protection for trigger and vibration inputs. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has a precision voltage reference for an A / D zero reference.

[0231] In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes a phase-locked loop bandpass tracking filter for acquiring slow RPMs and phase information. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and uses an onboard timer to digitally derive phase relative to input and trigger channels. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes a peak detector for automatic scaling, which is routed to a separate analog-to-digital converter for peak detection. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and routes raw or buffered trigger channels to other analog channels. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and uses higher input oversampling for the oversampled A / D converter to achieve a lower sample rate output, thereby minimizing AA filter requirements. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and uses a CPLD as a clock divider for an oversampling analog-to-digital converter to achieve lower sampling rates without the need for digital resampling.

[0232] In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has long blocks of data acquired at a high sampling rate, rather than multiple sets of data collected at different sampling rates. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and stores calibration data with an onboard card group maintenance history. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has fast route creation capabilities using hierarchical templates. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has intelligent management of data collection bands. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has a neural network expert system that utilizes intelligent management of data collection bands.

[0233] In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and utilizes a database hierarchy for sensor data analysis. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes an expert system GUI graphical method for defining expert system intelligent data collection bands and diagnostics. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes a graphical method for back-calculation definition. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes a bearing analysis method. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and utilizes transient signal analysis for torsional vibration detection / analysis. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and utilizes analog and digital methods for improved integration.

[0234] In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has an adaptive scheduling technique for continuously monitoring analog data in a local environment. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has a data collection resident feature. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has a self-contained data acquisition box. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has SD card storage. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has extended onboard statistics capabilities for continuous monitoring. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and uses ambient noise, local noise, and vibration noise for prediction. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and makes intelligent routing changes based on incoming data or alarms to synchronize dynamic data for analysis or correlation. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has an intelligent ODS and transfer function. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has a layered multiplexer. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has a sensor overload indicator. In an embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and has an RF identification and inclinometer.

[0235] In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes continuous ultrasonic monitoring. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes cloud-based machine pattern recognition based on remote analog industrial sensor fusion. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and performs cloud-based machine pattern analysis on state information from multiple analog industrial sensors to provide expected state information for the industrial system. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes a cloud-based policy automation engine for the Internet of Things, as well as the creation, deployment, and management of IoT devices. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes on-device sensor fusion and data storage for industrial IoT devices. In one embodiment, a data collection and processing system is provided that uses an analog crosspoint switch to collect data with a variable set of analog sensor inputs and includes a self-organizing data marketplace for industrial IoT data. In one embodime...

Claims

1. A system for data collection, processing and utilization of signals from at least a first element in a machine in an industrial environment, the system comprising: a platform comprising a computing environment connected to a local data collection system having at least a first sensor signal and a second sensor signal obtained from the machine in the industrial environment; a first sensor located in the local data collection system and configured to be connected to the machine for obtaining the first sensor signal from the machine; as well as a second sensor, the second sensor being located in the local data collection system and configured to obtain the second sensor signal from the machine; wherein the local data collection system comprises a plurality of inputs and a plurality of outputs, the plurality of inputs and the plurality of outputs having a first input connected to the first sensor and a second input connected to the second sensor, wherein the plurality of outputs comprises a first output and a second output, the first output and the second output being configured to allow simultaneous transmission of the first sensor signal from the first output and the second sensor signal from the second output, wherein the local data collection system is configured to simultaneously record gapless digital waveform data from at least the first input and the second input, and wherein the local data collection system is configured to manage a data collection band based on the recorded gapless digital waveform data.

2. The system according to claim 1, wherein: The data collection band defines a specific frequency band and at least one of a set of spectral peaks, true peak levels, a crest factor derived from a time waveform, or a complete waveform derived from a vibration envelope.

3. The system according to claim 1, wherein: The local data collection system includes a neural network expert system that utilizes the data collection band intelligent management.

4. The system according to claim 1, wherein: The local data collection system is configured to create data collection paths based on hierarchical templates, each hierarchical template including the data collection bands associated with the machines associated with the data collection paths.

5. The system according to claim 4, wherein: At least one of the layered templates is associated with a plurality of interconnected elements of the machine.

6. The system according to claim 4, wherein: At least one of the layer templates is associated with similar elements associated with the machine and another second machine.

7. The system according to claim 4, wherein: At least one of the layer templates is associated with the machine that is located adjacent to another second machine.

8. The system according to claim 1, wherein: The local data collection system includes a graphical user interface system configured to manage the data collection bands.

9. The system according to claim 8, wherein: The graphical user interface system includes an expert system diagnostic tool.

10. A system for data collection, processing and utilization of signals from at least a first element in a machine in an industrial environment, the system comprising: a platform comprising a computing environment connected to a local data collection system having at least a first sensor signal and a second sensor signal obtained from the machine in the industrial environment; a first sensor located in the local data collection system and configured to be connected to the machine for obtaining the first sensor signal from the machine; as well as a second sensor, the second sensor being located in the local data collection system and configured to obtain the second sensor signal from the machine; wherein the local data collection system comprises a plurality of inputs and a plurality of outputs, the plurality of inputs and the plurality of outputs having a first input connected to the first sensor and a second input connected to the second sensor, wherein the plurality of outputs comprises a first output and a second output, the first output and the second output being configured to allow simultaneous transmission of the first sensor signal from the first output and the second sensor signal from the second output, wherein the local data collection system is configured to simultaneously record gapless digital waveform data from at least the first input and the second input, and wherein the platform comprises a cloud-based machine pattern analysis of status information from the plurality of sensors to provide expected status information of the industrial environment based on the recorded gapless digital waveform data.

11. The system according to claim 10, wherein: The platform is configured to provide self-organization of the data pool based on at least one of a utilization metric and a revenue metric.

12. The system according to claim 10, wherein: The platform includes a self-organizing group of industrial data collectors.

13. The system according to claim 10, wherein: The local data collection system includes a wearable tactile user interface for an industrial sensor data collector, the wearable tactile user interface having at least one of a vibration, thermal, electrical, or acoustic output.

14. The system according to claim 10, wherein: The plurality of inputs to the local data collection system include a third input connected to the second sensor and a fourth input connected to the second sensor, and wherein the first sensor signal is from a single axis sensor located at a constant position associated with the machine.

15. The system of claim 14, wherein: The second sensor is a three-axis sensor.

16. The system of claim 10, wherein: The platform is configured to determine a change in relative phase based on the simultaneously recorded gapless digital waveform data.

17. The system according to claim 10, wherein: The second sensor is configured to be movable to a plurality of locations associated with the machine while acquiring the simultaneously recorded gapless digital waveform data.

18. The system of claim 10, wherein: The plurality of outputs of the local data collection system include a third output and a fourth output, and wherein the second output, the third output, and the fourth output are distributed together to a series of three-axis sensors each located at a different location associated with the machine.

19. The system of claim 16, wherein: The platform is configured to determine an operating deformation mode based on the change in relative phase and the simultaneously recorded gapless digital waveform data.

20. A computer-implemented method for data collection, processing, and utilization of signals from at least a first element in a machine in an industrial environment using a local data collection system, the local data collection system comprising a plurality of inputs and a plurality of outputs, the plurality of inputs and the plurality of outputs having: connected to a first input of the first sensor and to a second input of the second sensor, the method comprising: obtaining at least one of a first sensor signal and a second sensor signal from the machine; allowing simultaneous transmission of the first sensor signal from the first output and the second sensor signal from the second output; simultaneously recording gapless digital waveform data from at least the first input and the second input; Data collection bands are managed based on the recorded gapless digital waveform data.

21. The method according to claim 20, further comprising: A plurality of multiplexing units and a plurality of data acquisition units receive a plurality of data streams from a plurality of machines in the industrial environment.

22. The method according to claim 21, wherein: The local data collection system includes distributed complex programmable hardware device (CPLD) chips, and also includes a data bus that dedicates each distributed complex programmable hardware device (CPLD) chip to logically control the multiple multiplexing units and the multiple data acquisition units that receive the multiple data streams from the multiple machines in the industrial environment.

Images