Device, system and method for monitoring a steam trap and detecting a failure of the steam trap

Abstract

An example of a steam trap monitoring device includes a housing, a sensor subsystem housed in the housing that measures characteristics of the steam trap, a memory housed in the housing, a communication interface housed in the housing and configured to communicate with a server, and a processor housed in the housing and interconnected to the sensor subsystem, the memory, and the communication interface, wherein the processor is configured to obtain data representative of the characteristics of the steam trap from the sensor subsystem, extract a set of significant features from the data, and transmit the set of significant features via the communication interface to the server for further processing to detect faults.

Description

Technical Field

[0001] This specification generally relates to steam-powered systems, and more specifically to an apparatus, system, and method for monitoring a target device and detecting a failure of the target device.

Background Art

[0002] Systems including pipe network systems, steam-powered systems, etc. may include various components, such as pumps, motors, and traps, and these components sometimes fail and may have an adverse effect on the systems in which they are deployed. For example, a steam trap is used in the steam line of a steam-powered process to remove condensate that may block the steam line and inhibit the steam-powered process. A steam trap may fail when the valve cannot open and close as intended. When a steam trap fails, steam is lost, and since additional steam has to be generated to replace the lost steam, it is costly, and when the condensate blocks the steam line, it becomes harmful to the steam-powered process.

Summary of the Invention

[0003] According to one aspect of this specification, a monitoring device for a steam trap is provided. The monitoring device includes a housing, a sensor subsystem housed in the housing for measuring the characteristics of the steam trap, a memory housed in the housing, a communication interface housed in the housing and configured to communicate with a server, and a processor housed in the housing and interconnected to the sensor subsystem, the memory, and the communication interface. The processor is configured to obtain data representing the characteristics of the steam trap from the sensor subsystem, extract a set of important features from the data, and transmit the set of important features to the server via the communication interface for further processing.

[0004] According to another aspect of the present specification, a method for detecting a failure of a steam trap is provided. The method includes obtaining, by a server, a set of important features representing steam trap data captured by a monitoring device of the steam trap; determining, based on the set of important features, whether a failure of the steam trap has been detected; if a failure is detected, sending an alert to a client device; and outputting dashboard data to the client device.

[0005] According to yet another aspect of the present specification, a system for detecting a failure of a steam trap is provided. The system includes a server and a monitoring device coupled to the steam trap. The monitoring device includes a sensor subsystem configured to measure characteristics of the steam trap and a processor interconnected with the sensor subsystem. The processor is configured to obtain steam trap data representing the characteristics of the steam trap from the sensor subsystem, extract a set of important features from the steam trap data, and send the set of important features to the server. The server is configured to determine whether a failure of the steam trap has been detected based on the set of important features received from the monitoring device.

Brief Description of the Drawings

[0006] Embodiments will be described with reference to the following drawings.

[0007]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7A

Figure 7B

Figure 7C

Figure 8

[0008] Failures of components of a system, such as a steam power system or other systems, are time-consuming and costly for the company operating the system. Accordingly, it is desirable to monitor specific components or target devices that are prone to failure and can have a significant impact on operation. Some solutions may include providing a monitoring device that monitors a target device by capturing data such as audio data, image data, temperature data, etc., and analyzing the data to detect failures of the target device. Some monitoring devices may analyze the captured data using an onboard circuit, but such solutions are expensive, especially when the facility or system requires hundreds or thousands of monitoring devices to monitor each target device. Further, since each target device is monitored individually, such a system does not provide an overview of the overall functional state of the entire facility. Accordingly, other monitoring devices may capture data and transmit the data to a central computing device for further processing. However, in order to transmit the captured data to the central computing device, the monitoring device may face the challenge of transmitting data over long distances because it utilizes a high-bandwidth communication protocol. Therefore, the system is limited to localized computing devices and analysis.

[0009] The present disclosure describes a system including a monitoring device for monitoring a target device. The monitoring device includes a sensor subsystem that captures data regarding the target device (e.g., audio data, vibration data, temperature data), and a processor that applies digital signal processing techniques to the captured data to extract a set of significant features. The set of significant features is representative of the captured data and is concise enough to enable stable transmission of the data via a low-power wide-area network. That is, by applying signal processing to the monitoring device itself, a large data set can be reduced to a small set of significant features, so that the device can transmit the set of significant features to a cloud-based server, for example, using long-range (LoRa) communication. Even more advantageously, since the set of significant features is representative of the large data set, the cloud-based server can analyze the set of significant features to determine the state of the target device. Thus, more memory-intensive operations may be performed on the cloud rather than on individual devices.

[0010] FIG. 1 shows a system 100 for monitoring and detecting faults in a target device. The system 100 includes a monitoring device 104 that communicates with a server 108. In this embodiment, the monitoring device 104 is configured to monitor a steam trap 110 in a steam line (not shown).

[0011] The steam trap 110 includes an inlet line 112, a body 114, a condensate line 116, and a valve 118. During operation, steam and steam condensate flow from the steam line through the inlet line 112 into the body 114 of the steam trap 110. Condensate and other non-condensable fluids are collected in the body 114, and when the body 114 has accommodated a predetermined amount of condensate, the valve 118 is opened to discharge the condensate into the condensate line 116. In particular, due to the configuration and structure of the valve 118 and the inlet line 112 on the body 114, the valve 118 can be opened to discharge the condensate while discharging a small amount of steam from the steam line. For example, the valve 118 may be a floating ball valve where a floating ball floats on the condensate to open the valve 118 when sufficient condensate is accommodated in the body 114. As the condensate is discharged through the condensate line 116, the floating ball drops to close the valve 118.

[0012] As will be seen later, during operation, the valve 118 opens and closes periodically and sometimes experiences mechanical failures. For example, the valve 118 may fail in the open state, in which case, even though less than a predetermined amount of condensate is accommodated in the body 114, the valve 118 remains open. If the valve 118 fails in the open state, steam may escape from the open valve 118, causing steam to be lost in the steam line. Accordingly, a system that generates steam for the steam line needs to generate more steam to maintain the required amount of steam supplied to the steam power process by the steam line. In other embodiments, the valve 118 may fail in the closed state, in which case, even though more than a predetermined amount of condensate is accommodated in the body 114, the valve 118 remains closed. If the valve 118 fails in the closed state, the condensate may block the steam line, causing a process failure in the process supplied by the steam line.

[0013] Accordingly, the monitoring device 104 is arranged close to the steam trap 110 to monitor one or more characteristics of the steam trap 110 and report the characteristics to the server 108 to detect potential failures of the steam trap. For example, the monitoring device 104 may be attached to a fluid line, for example, the inflow line 112, preferably, attached to the aggregation line 116 as shown in this embodiment. Thus, the monitoring device 104 generally includes a plurality of sensors (for example, a sensor subsystem) configured to measure the characteristics of the steam trap. For example, the sensors may measure temperature data, audio data, vibration data, etc. The monitoring device 104 is further configured to communicate with the server 108 and transmit data to the server 108 for further analysis. Preferably, the monitoring device 104 may communicate with the server 108 via a wide area network using a communication protocol such as the Long Range (LoRa) protocol. In particular, since the LoRa protocol is a low-power wide area network communication protocol, the bandwidth of the data transmitted from the monitoring device 104 to the server 108 may be limited. Accordingly, instead of transmitting all the data obtained from the sensors, the monitoring device 104 is further configured to extract a set of important features from the data obtained from the sensors and transmit the important features to the server 108 for further analysis. Hereinafter, the structure, internal components and functions of the monitoring device 104 will be described in more detail.

[0014] The communication link 106 between the monitoring device 104 and the server 108 is preferably substantially wireless and may include a combination of wired and wireless connections including a direct link or a link across one or more networks. For example, the communication link 106 may utilize a network including any one or any combination of a suitable wide area network (WAN) such as a cellular network, the Internet, etc., and any suitable local area network (LAN) defined by one or more routers, switches, wireless access points, etc. For example, the communication link 106 may include a first link reaching a gateway via a long range (LoRa) network and a second link reaching the server 108 via a long-term evolution network (LTE (registered trademark)).

[0015] The server 108 is generally configured to obtain and analyze important features of the data acquired by the sensors of the monitoring device 104 to determine the functional state of the steam trap 110. That is, the server 108 may determine whether the steam trap 110 is functional, whether the valve 118 has failed in the open state, or whether the valve 118 has failed in the closed state based on the important features. If a failure is detected in the steam trap 110, the server 108 may send a notification or alert to, for example, a client device 120 controlled by an operator of the facility where the steam trap 110 is located. The server 108 may further aggregate and store the important features of the steam trap 110 and present the aggregated data to the client device 120. Hereinafter, the internal components and functions of the server 108 will be described in more detail. As will be understood later, in some embodiments, the functions of the server 108 may be executed in any suitable server environment including a plurality of cooperating servers, a cloud-based server environment, etc.

[0016] The client device 120 may be a computing device such as a personal computer or a desktop computer, a laptop, a tablet, a mobile device, or another server. In this embodiment, a single client device 120 is shown, but in other embodiments, the server 108 may communicate with a plurality of client devices 120. In particular, the client device 120 may be operated by a worker such as a facility manager or an operator of a facility where the steam trap 110 is deployed. The client device 120 communicates with the server 108 and in particular receives alerts or notifications and includes appropriate hardware (e.g., speakers, displays) for generating visual or audio signals indicative of the alerts and notifications. The client device 120 is further configured to receive dashboard data from the server 108 that includes past data representing the measured characteristics of the steam trap 110 and to display the dashboard data for viewing by a user of the client device 120. For example, an operator may use a personal computer as the client device 120 to access a web application and view the dashboard data. Further, as will be seen later, the alerts and notifications and the viewing of the dashboard data may occur on different client devices 120.

[0017] Referring to FIG. 2, a cross-sectional view of the monitoring device 104 is shown. The monitoring device 104 includes a housing 200 that houses a circuit board 204, an accelerometer 208, two microphones 212-1 and 212-2 (generally referred to as microphone 212, collectively referred to as microphone 212, and this term is also used elsewhere in this specification), and two temperature sensors 216-1 and 216-2. The monitoring device 104 may further include a mounting bracket 220 coupled to the housing 200 and configured to attach the monitoring device 104 to a fluid line (e.g., a steam line, an inflow line 112, or a condensate line 116).

[0018] The housing 200 is generally configured to house the internal components of the monitoring device 104 and protect the internal components from damage. The housing 200 may include plastics such as polyphenylene sulfide (PPS), polymers, metals, or combinations thereof. For example, the housing 200 may be formed by injection molding a plastic material. Preferably, the housing 200 is formed from a heat-resistant material to reduce heat transfer from the fluid line to which the monitoring device 104 is attached to the monitoring device 104 itself, particularly its internal components.

[0019] The circuit board 204 may be a printed circuit board (PCB) that supports electronic components (described in more detail below) and one or more sensors. For example, in this embodiment, the circuit board 204 supports an accelerometer 208 and a secondary temperature sensor 216-2 as one component of the accelerometer 208.

[0020] The accelerometer 208 may be any suitable motion detection sensor configured to measure the motion of the steam trap 110, particularly vibrations. More specifically, the accelerometer 208 supported and attached to the housing 200 of the monitoring device 104 measures the vibrations of the monitoring device 104. Also, since the monitoring device 104 may be attached to the condensate line 116 of the steam trap 110, the vibrations of the steam trap 110 can be detected by the accelerometer 208 as they are propagated to the monitoring device 104.

[0021] The microphone 212 may be any suitable sensor configured to capture audio data representative of the sound generated by the steam trap 110 and the sound from the environment of the steam trap 110 (e.g., background noise). In this embodiment, the monitoring device 104 includes a microphone 212-1 and a secondary microphone 212-2. During operation, the monitoring device 104 may be oriented such that the microphone 212-1 faces in the direction towards the steam trap 110, while the secondary microphone 212-2 faces in the direction away from the steam trap 110. Accordingly, the microphone 212-1 may primarily capture the sound generated by the steam trap 110, while the secondary microphone 212-2 oriented away from the steam trap 110 is configured to capture secondary audio data representative of the sound from the environment of the steam trap (e.g., background noise).

[0022] In some embodiments, to further limit the sound received by the microphone 212, the housing 200 may include cylinders 214-1 and 214-2 that extend between the microphone 212 housed within the housing 200 and the outer surface of the housing 200. The cylinders 214 may be formed as a separate component configured to connect to the housing 200, or may be formed integrally with the housing 200. The cylinders 214 may have a light conical shape that tapers at the end proximal to each microphone 212 and widens at the opposite end (i.e., the end away from each microphone 212). In other embodiments, the cylinders 214 may have different shapes including separate walls, curved walls, etc. to condition the audio data captured by each microphone 212. The microphones 212 are disposed at each inner end of the cylinders 214 in the housing 200.

[0023] The cylinder 214 may further serve to substantially limit the audio data captured by the microphone 212 to the sound generated within the sector defined by each microphone 212 and corresponding cylinder 214. For example, since the microphone 212-1 is oriented towards the steam trap 110, the cylinder 214-1 is oriented between the microphone 212-1 and the steam trap 110. Thus, the cylinder 214-1 is configured to substantially limit the audio data captured by the microphone 212-1 to the sound originating from a direction corresponding substantially to the direction of the steam trap 110. That is, based on the orientation of the monitoring device 104 and thus the microphone 212-1 and the cylinder 214-1, the microphone 212-1 mainly captures the sound generated by the steam trap 110, and the cylinder 214-1 blocks or limits the sound originating from other directions from reaching the microphone 212-1. That is, the cylinder 214-1 makes the microphone 212-2 substantially unidirectional. Similarly, the secondary microphone 212-2 and the cylinder 214-2 cooperate to limit the audio data captured by the secondary microphone 212-2 to the sound generated from a direction away from the steam trap 110 defined with respect to the monitoring device 104.

[0024] The microphone 212 may additionally include a narrowband filter configured to attenuate frequencies outside a predetermined range. For example, when the steam trap 110 fails in the open state of the valve 118, steam is lost through the valve 118, and the steam trap 110 may emit ultrasonic waves having a frequency of about 40 kHz. Accordingly, a microphone 212-1 configured to capture audio data representing the sound generated by the steam trap 110 may attenuate frequencies outside the above range using a narrowband filter that passes frequencies within the range of about 35 kHz to about 45 kHz. In other embodiments, the band filter may be a wideband filter or a narrowband filter, and may have a different center based on the expected frequency of the noise emitted by the steam trap 110 when the valve 118 fails in the open state. As will be seen later, the microphone 212-1 and the secondary microphone 212-2 may use band filters in different frequency ranges based on the target sound captured by each microphone 212.

[0025] The monitoring device 104 further includes a temperature sensor 216-1 and a secondary temperature sensor 216-2. The temperature sensor 216 may be a thermometer or other suitable sensor configured to capture temperature data. In particular, the temperature sensor 216-1 is configured to capture temperature data representing the approximate temperature of the condensate line 116 of the steam trap 110, and the secondary temperature sensor 216-2 is configured to capture secondary temperature data representing the internal temperature of the housing 200. That is, the secondary temperature sensor 216-2 may be supported by the circuit board 204 housed in the housing 200 to measure the temperature inside the housing 200.

[0026] To capture the temperature of the condensate line 116, the temperature sensor 216-1 may be disposed proximate to the condensate line 116. In particular, the monitoring device 104 may be attached to the condensate line 116 via a mounting bracket 220.

[0027] For example, referring to FIG. 3, a perspective view of the mounting bracket 220 is shown. The mounting bracket 220 includes a mounting arm 300, a passage 304 extending from the mounting arm 300, and a plate 308 coupled to the passage 304 and spaced apart from the mounting arm 300.

[0028] The mounting arm 300 is generally configured to connect to a fluid line (e.g., the agglomeration line 116). Preferably, the mounting arm 300 may have a V-shape to reduce heat transfer from the fluid line to the monitoring device 104 via the mounting bracket 220. That is, the fluid line may be configured to be located inside the V-shape of the mounting arm 300. Based on the generally cylindrical shape of the fluid line (i.e., the pipe) and the V-shape of the mounting arm 300, the fluid line contacts the mounting arm 300 only at points along two lines (e.g., not over the entire surface). In the V-shape, by providing air permeability at the apex of the V-shape, heat transfer is further reduced. Additionally, due to the V-shape, the mounting bracket 220 can fit with fluid lines having various diameters while maintaining heat transfer reduction.

[0029] It will be apparent to those skilled in the art that the mounting arm 300 may be fixed to the fluid line via a clamp, a cord, a chain, or other suitable fasteners. In some embodiments, the mounting arm 300 may include a stopper 302 extending from an end to hold the fastener to the mounting arm 300 (i.e., to prevent the fastener from slipping off the end of the mounting arm 300).

[0030] The mounting bracket 220 further includes a passage 304 extending from the mounting arm 300. The passage 304 generally provides a space 306 that is enclosed and houses the sensor of the monitoring device 104. For example, the temperature sensor 216-1 may be supported within the housing 200 and extend into the space 306 of the passage 304 so as to position the temperature sensor 216-1 close to the fluid line where the temperature is measured. The accommodation of the temperature sensor 216-1 in the passage 304 is shown in FIG. 2. Thus, when the monitoring device 104 is attached to the agglomeration line 116, the accommodation of the temperature sensor 216-1 in the channel 304 allows the temperature sensor 216-1 to be placed close to the agglomeration line 116, and temperature data representing the temperature of the agglomeration line 116 can be captured more accurately.

[0031] Returning to FIG. 3, the passage 304 further serves to space the plate 308 from the mounting arm 300. That is, the plate 308 may be coupled to the passage 304 at an end opposite the mounting arm 300. The plate 308 is configured to connect to the housing 200 to support the monitoring device 104 on the mounting bracket 220. For example, the plate 308 may be received within the housing 200 such that it is within the housing 200 (i.e., inside the housing 200), and the plate 308 may include slots and tabs or pins configured to connect to corresponding slots and tabs or pins of the housing 200 to couple the housing 200 to the plate 308. In other embodiments, the plate 308 may be coupled to the housing 200 outside the housing 200 (e.g., the housing 200 may be fixed to the upper or open surface of the plate 308). Further, in other embodiments, screws, clamps, clips or other suitable fasteners that may be contemplated by those skilled in the art may be used alternatively or additionally to fix the housing 200 to the plate 308 so as to attach the monitoring device 104 to the fluid line via the mounting bracket 220.

[0032] Also, other modifications are possible. For example, in the presently shown embodiment, the mounting bracket 220 is a separate component from the housing 200. In other embodiments, the mounting bracket 220 and the housing 200 may be integrally formed. That is, the housing 200 may be formed to have a passage extending from the housing 200 and a mounting arm at the end of the passage.

[0033] Referring now to FIG. 4, it is a block diagram of certain electronic components of the monitoring device 104. The monitoring device 104 includes a processor 400, a memory 404, a communication interface 416, and a sensor subsystem 420, each housed or supported within the housing 200. For example, the processor 400, the memory 404, and the communication interface 416 may be supported on a circuit board 204.

[0034] The processor 400 may be a central processing unit (CPU), a microcontroller, a processing core, etc. The processor 400 may include a plurality of cooperating processors. In some embodiments, the functions executed by the processor 400 may be executed by one or more specially designed hardware and firmware components such as a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), a digital signal processing (DSP) processor, etc. In some embodiments, the processor 400 may be a dedicated processor that can be executed via a dedicated logic circuit such as an ASIC, an FPGA, a DSP processor, etc. to improve the processing speed of the monitoring operations described herein.

[0035] Processor 400 is interconnected with a non-transitory computer-readable storage medium such as memory 404. Memory 404 may include a combination of volatile memory (e.g., random access memory or RAM) and non-volatile memory (e.g., read-only memory or ROM, electrically erasable programmable read-only memory or EEPROM, flash memory). Processor 400 and memory 404 may include one or more integrated circuits. Some or all of memory 404 may be integrated with processor 400. Memory 404 stores computer-readable instructions to be executed by processor 400. In particular, memory 404 stores control application 408, and when control application 408 is executed by processor 400, it configures processor 400 to perform various functions related to the monitoring operation of the steam trap by monitoring device 104, which will be described in more detail below. Application 408 may also be executed as a series of separate applications. Memory 404 may also store repository 412 including rules, thresholds, and other data used in the monitoring operation of the steam trap by monitoring device 104.

[0036] Monitoring device 104 also includes a communication interface 416 interconnected with processor 400. Communication interface 416 includes appropriate hardware (e.g., transmitter, receiver, network interface controller, etc.) that enables monitoring device 104 to communicate with other computing devices, particularly server 108. Specific components of communication interface 416 are selected based on the type of network or other link including communication link 106, and monitoring device 104 communicates via communication link 106.

[0037] The monitoring device 104 also includes a sensor subsystem 420. The sensor subsystem 420 in this embodiment is shown to include an accelerometer 208, a microphone 212-1, and a temperature sensor 216-1. In other embodiments, the sensor subsystem 420 may include additional sensors such as a secondary microphone 212-2 and a secondary temperature sensor 216-2, but is not limited thereto, and may also include other suitable sensors. In yet another embodiment, the sensor subsystem 420 may include some or alternative sensors of the sensors shown and described herein.

[0038] In some embodiments, the monitoring device 104 may also include one or more input devices and / or output devices (not shown) interconnected with a processor 400. The input devices may include one or more buttons, keypads, touch display screens, etc., that receive input from an operator or the like. The output devices may include one or more display screens, audio generators, vibrators, etc., that provide output or feedback.

[0039] Returning to FIG. 5, a server 108 including specific internal components is shown in more detail. The server 108 includes a processor 500 such as a central processing unit (CPU), a microcontroller, a processing core, etc. The processor 500 may include a plurality of cooperating processors. In some embodiments, the functions executed by the processor 500 may be executed by one or more specially designed hardware and firmware components such as an FPGA, an ASIC, etc. In some embodiments, the processor 500 may be a dedicated processor that can be executed via a dedicated logic circuit such as an ASIC, an FPGA, a DSP processor, etc., to improve the processing speed of the fault determination operation described herein.

[0040] Processor 500 is interconnected with a non-transitory computer-readable storage medium such as memory 504. Memory 504 may include a combination of volatile memory (e.g., random access memory or RAM) and non-volatile memory (e.g., read-only memory or ROM, electrically erasable programmable read-only memory or EEPROM, flash memory). Processor 500 and memory 504 may include one or more integrated circuits. Part or all of memory 504 may be integrated with processor 500. Memory 504 stores computer-readable instructions to be executed by processor 500. In particular, memory 504 stores control application 508, and when executed by processor 500, control application 508 configures processor 500 to perform various functions related to the steam trap fault determination operation by server 108, which will be described in more detail below. Application 508 may also be executed as a series of separate applications. Memory 504 may also store a repository 512 that includes rules (e.g., related to thresholds, ranges, or other conditions that define a steam trap fault) used in the steam trap fault determination operation and previously received steam trap data. In other embodiments, memory 504 and / or repository 512 may also store other rules and data related to the steam trap fault determination operation of system 100.

[0041] Server 108 also includes a communication interface 516 interconnected with processor 500. Communication interface 516 includes appropriate hardware (e.g., a transmitter, a receiver, a network interface controller, etc.) that enables server 108 to communicate with other computing devices, particularly monitoring device 104 and client device 120. Specific components of communication interface 516 are selected based on the type of network or other link that includes communication link 106, and server 108 communicates via communication link 106.

[0042] In some embodiments, server 108 may also include one or more input devices and / or output devices (not shown) interconnected with processor 500. The input devices may include one or more buttons, keypads, touch display screens, etc. that receive input from an operator. The output devices may include one or more display screens, audio generators, vibrators, etc. that provide output or feedback.

[0043] The operation of system 100 executed by application 408 and 508 by processors 400 and 500 respectively will be described in more detail below. FIG. 6 shows a method 600 for monitoring a steam trap and detecting a failure of the steam trap. With reference to the components shown in FIGS. 1-5, method 600 will be described in conjunction with its execution in system 100. In other embodiments, method 600 may be executed by other suitable devices and / or systems.

[0044] Method 600 begins at block 605. At block 605, monitoring device 104 captures steam trap data representing one or more characteristics of steam trap 110. For example, monitoring device 104 may use microphone 212-1 to capture audio data representing the sound generated by steam trap 110, use temperature sensor 216-1 to capture temperature data representing the approximate temperature of condensate line 116, and capture vibration data representing the vibration caused by steam trap 110 and received by monitoring device 104. In some embodiments, the steam trap data captured at block 605 may further include secondary audio data representing the sound from the environment of steam trap 110 captured by secondary microphone 212-2 and secondary temperature data representing the internal temperature of housing 200 of monitoring device 104. In yet another embodiment, additional steam trap data may be captured using other sensors of sensor subsystem 420.

[0045] In block 610, the monitoring device 104 extracts a set of important features from the steam trap data captured in block 605. In particular, by forming a representative sample of the steam trap data captured in block 605 with the set of important features, the server 108 can accurately determine the functional state of the steam trap 110, and by reducing the dataset to a sufficiently small size, the set of important features can be packetized and transmitted to the server 108 via a low-power wide-area network, for example using the LoRa communication protocol.

[0046] For example, referring to FIG. 7A, an exemplary method 700 is shown for processing the audio data captured by the microphone 212-1 to extract audio data points included in a set of important features. In other embodiments, the method 700 may be applied to secondary audio data captured by the microphone 212-2.

[0047] In block 705, the monitoring device 104, particularly the processor 400, samples the audio data captured by the microphone 212-1 at a predetermined number of points within a predetermined time interval or at a predetermined time interval. That is, the monitoring device 104 determines the magnitude of the audio data detected at discrete time points within a predetermined time interval or at a predetermined time interval. For example, the monitoring device 104 may sample the audio data 300 times at intervals of about 1 millisecond. In other embodiments, the monitoring device 104 may sample the audio data about 100 times, about 2000 times, or at other appropriate sampling rates. Further, in other embodiments, the sampling may be performed at intervals of 3 milliseconds, 10 milliseconds, or other appropriate time intervals.

[0048] In block 710, the monitoring device 104 determines the average value of the magnitudes of the samples obtained in block 705 (i.e., the average magnitude of the sampled audio data). This average value is defined as an audio data point included in a set of important features.

[0049] In block 715, the monitoring device 104 determines whether it has acquired a threshold number of audio data points. If it has acquired a threshold number of audio data points, the monitoring device 104 ends method 700 and returns to block 615 of method 600. If it has not acquired a threshold number of audio data points, the monitoring device 104 returns to block 705 to acquire additional audio data points. For example, the threshold number of data points may be about 20 data points. The threshold number of audio data points may be defined, for example, within repository 412 and selected based on the bandwidth capacity of communication interface 416. That is, the threshold number of audio data points is selected to provide sufficient information to server 108 to analyze the audio data and to maintain stable transmission of data from monitoring device 104 to server 108, particularly via a low-power wide-area network.

[0050] Now, referring to FIG. 7B, an exemplary method 720 is shown for processing the vibration data captured by accelerometer 208 to extract vibration data points included in a set of important features.

[0051] In block 725, the monitoring device 104, particularly processor 400, determines the frequency band power of the vibration data within a predetermined time interval. That is, the monitoring device 104 determines the vibration frequency at which the power is strongest within the predetermined time interval. The frequency band power within the predetermined interval is defined as a vibration data point included in a set of important features.

[0052] In block 730, the monitoring device determines whether it has acquired a threshold number of vibration data points. If it has acquired the threshold number of vibration data points, the monitoring device 104 ends method 720 and returns to block 615 of method 600. If it has not acquired the threshold number of vibration data points, the monitoring device returns to block 725 to acquire additional vibration data points. For example, the threshold number of vibration data points may be about 20 data points. The threshold number of vibration data points may be defined, for example, within repository 412 and selected based on the bandwidth capacity of communication interface 416. That is, the threshold number of vibration data points is selected to provide sufficient information to server 108 to analyze the vibration data and to maintain stable transmission of data from monitoring device 104 to server 108, particularly via a low power wide area network.

[0053] Now referring to FIG. 7C, an exemplary method 740 is shown for processing temperature data captured by temperature sensor 216-1 to extract temperature data points included in a set of important features. In other embodiments, method 740 may be applied.

[0054] In block 745, the monitoring device 104 defines the temperature recorded by temperature sensor 216-1 as a temperature data point included in a set of important features. In particular, since temperature sensor 216-1 may be configured to measure temperature at a single discrete point in time, no further processing to obtain discrete data points for the set of important features is required. Next, the monitoring device 104 may proceed to block 615 of method 600.

[0055] In other embodiments, it should be understood that other methods of sampling audio data, vibration data, and temperature data may be used to extract audio data points, vibration data points, and temperature data points included in a set of important features. For example, instead of obtaining the temperature recorded at a single discrete point in time, the monitoring device 104 may determine the average temperature recorded within a predetermined time interval. Other methods of sampling a continuous signal to obtain a predetermined number of discrete data points representing the continuous signal are also conceivable.

[0056] Returning to FIG. 6, after extracting the important features of the steam trap data, the monitoring device 104 proceeds to block 615 of method 600. At block 615, the monitoring device 104 transmits the set of important features to the server 108 using the communication interface 416. In particular, since the monitoring device 104 and the server 108 are separated from each other and the communication link 106 can cross a wide area network, the communication interface 416 may use the LoRa communication protocol. In addition to the set of important features, the monitoring device 104 may additionally transmit identification data related to, for example, the monitoring device 104 itself, the steam trap 110, the facility in which the steam trap 110 is deployed, and the like.

[0057] In block 620, the monitoring device 104 determines whether a predetermined time has elapsed since it sent the important features to the server. If the predetermined time has elapsed, the monitoring device 104 returns to block 605, captures new data, and provides the periodically updated steam trap data to the server 108. If the predetermined time has not elapsed, the monitoring device 104 continues to wait until the predetermined time has elapsed. In some embodiments, the monitoring device 104 may be configured to return to a low-power or sleep state until the predetermined time has elapsed in order to conserve power and energy. The predetermined time may be, for example, 5 minutes, 10 minutes, 30 minutes, 1 hour, or other suitable period. Further, in some embodiments, each data type (e.g., audio data, vibration data, temperature data) may correspond to a different predetermined time for obtaining updated data. For example, audio data and vibration data may be captured every 30 minutes, and temperature data may be captured every 5 minutes.

[0058] In block 625, the server 108 obtains a set of important features from the monitoring device 104 and proceeds to block 630 for further processing. In some embodiments, before proceeding to block 630, the server 108 may first extract secondary temperature data from the set of important features to evaluate the operating state of the monitoring device. In particular, if the temperature data point representing the internal temperature of the housing 200 exceeds the threshold temperature, the server 108 determines that the temperature condition of the monitoring device 104 exceeds the acceptable operating threshold, and thus the data captured by the sensor may be inaccurate. Accordingly, the server 108 may generate an alert and send the alert to the client device 120 to warn the operator of the inoperable state of the monitoring device 104.

[0059] In block 630, the server 108 determines whether a steam trap failure has been detected based on the set of important features.

[0060] For example, referring to FIG. 8, an exemplary method 800 for identifying a steam trap failure is shown. The blocks of method 800 are referred to as blocks rather than steps because they may be executed simultaneously and / or in an order different from that shown. For example, server 108 may analyze temperature data simultaneously with, rather than sequentially with, audio data and vibration data.

[0061] In block 805, server 108 determines whether an audio data point exceeds a threshold magnitude. In some embodiments, server 108 may determine that an audio data point from a set of significant features exceeds the threshold magnitude if a majority (or a threshold percentage) of the audio data points exceed the threshold magnitude. In other embodiments, server 108 may require that all audio data points from a set of significant features exceed the threshold magnitude, or that at least one audio data point from a set of significant features exceeds the threshold magnitude. Since the narrowband filter attenuates frequencies outside a predetermined range, an audio data point having a magnitude exceeding the threshold magnitude indicates that the captured audio data is in a predetermined range corresponding to the valve 118 failing in the open state.

[0062] Furthermore, in some embodiments, server 108 may determine whether the audio data points exceed a threshold magnitude over at least a threshold time period (e.g., 2 hours, 6 hours, 1 day, or another suitable period). Thus, in particular, in some embodiments, server 108 may consider past data of the audio data points. That is, server 108 may search for audio data points from a previously received set of important features (e.g., those stored in repository 512). For example, server 108 may first determine whether a threshold percentage of the audio data points exceed the threshold magnitude. If the determination is affirmative, server 108 may search for past data of the audio data points to determine whether the audio data points exceed the threshold magnitude for at least 2 hours (i.e., whether the previous 4 sets of audio data points also exceed the threshold magnitude).

[0063] The magnitude of the threshold and the specific conditions satisfied by the determination of block 805 may be defined within repository 512. In some embodiments, the magnitude of the threshold may be determined dynamically, for example, with respect to a baseline environmental noise. In other embodiments, the magnitude of the threshold may be determined dynamically based on previously recorded audio data (i.e., to detect a change over time in the pattern of the audio data).

[0064] If it is determined that the audio data point exceeds the threshold magnitude, server 108 may determine that the audio data indicates that steam trap 110 with valve 118 in the open state has failed. Further, a determination that the audio data point has exceeded the threshold magnitude for at least the threshold time (i.e., across multiple sets of audio data points) indicates that the failure is ongoing and not temporary. In some embodiments, after a positive determination is made at block 805, server 108 may proceed directly to block 820 (shown by the dashed line). In other embodiments, after a positive determination is made at block 805, server 108 may proceed to block 810 to confirm the determination of a trap open failure. If the determination at block 805 is negative, server 108 may determine that valve 118 is not failed in the open state and may proceed to block 830 for further analysis.

[0065] At block 810, server 108 confirms the determination of a trap open failure using secondary audio data. In particular, server 108 determines whether secondary audio data points from the secondary audio data also exceed the threshold magnitude. The threshold magnitude and specific conditions in the case of a positive determination may be the same as the threshold magnitude and specific conditions of the audio data points and may be defined within repository 512.

[0066] The secondary audio data points are also not indicative of a steam trap 110 failure, as when determined to exceed a threshold magnitude, such data indicates that the audio detected by microphone 212-1 is not generated by the steam trap 110, but is present within the environment (e.g., facility) of the steam trap 110. Specifically, each of microphones 212 substantially and directionally captures audio, with microphone 212-1 capturing audio from the direction of the steam trap and secondary microphone 212-2 capturing audio originating from a direction away from the direction of the steam trap, such that the captured audio data within the same frequency range indicates that the captured audio data is not generated from the steam trap 110 itself, but rather is generated from an omnidirectional or multi-directional sound source or from multiple sound sources.

[0067] Accordingly, if the determination at block 810 is affirmative, server 108 may determine that valve 118 is open and not failed and proceed to block 830 for further analysis. If the determination at block 810 is negative, server 108 may proceed to block 820 (shown in dashed lines) in some embodiments or to block 815 in other embodiments to further confirm the determination of a trap open failure.

[0068] At block 815, server 108 determines whether the vibration data points exceed a threshold magnitude. That is, server 108 determines whether the vibration data indicates that monitoring device 104 is vibrating beyond a specific frequency. For example, the determination may be made regarding whether a threshold percentage of vibration data points exceed the threshold magnitude. The magnitude of the threshold and specific conditions satisfied by the determination at block 815 may be defined within repository 512. For example, the magnitude of the threshold may be dynamically determined based on a baseline vibration frequency for previously recorded vibration data in order to detect, for example, a change over time in the pattern of vibration data. In particular, if it is determined that the vibration data points exceed the threshold magnitude, the vibration received by monitoring device 104 indicates that steam trap 110 has failed with valve 118 in the open state. That is, steam escaping from the open valve 118 causes vibrations within condensate line 116, and these vibrations are propagated to monitoring device 104.

[0069] Accordingly, if the determination at block 815 is affirmative, server 108 proceeds to block 820. If the determination at block 815 is negative, server 108 proceeds to block 825.

[0070] At block 820, server 108 determines that steam trap 110 has failed with valve 118 in the open state based on the set of important features obtained at block 625. Then, server 108 proceeds to block 635.

[0071] At block 825, server 108 determines that there is a possibility that steam trap 110 has failed, but the data is unclear, based on the set of important features obtained at block 625. For example, the audio data indicates that steam trap 110 has failed with valve 118 in the open state, but this conclusion is not supported by the vibration data, which has not demonstrated a vibration frequency high enough to support the conclusion of a trap open failure. Accordingly, for example, the data needs to be further analyzed, such as by review by an operator of the facility. Then, server 108 proceeds to block 635.

[0072] In block 830, server 108 determines whether the temperature data point obtained from temperature sensor 216-1 is less than a threshold temperature. In some embodiments, server 108 may further determine whether the temperature has been less than the threshold temperature for at least a threshold time period (e.g., 30 minutes, 2 hours, 6 hours, or another suitable period). Thus, in particular, server 108 may consider past data of the temperature data points. That is, server 108 may search for the temperature data point from a previously received set of important features (e.g., those stored in repository 512). Thus, to determine whether the temperature has been less than the threshold temperature for at least 30 minutes, server 108 may also determine whether the previous six sets of temperature data points were also less than the threshold temperature.

[0073] If it is determined that the temperature data point is less than the threshold temperature, server 108 may determine that the temperature data indicates that steam trap 110 has failed with valve 118 in the closed state. That is, the condensate contained within body 114 is generally warm and is continuously slightly warmed by the nearby steam. Thus, when the condensate is discharged into collection line 116, it also warms collection line 116. If collection line 116 remains at a low temperature over a long period of time, this indicates that valve 118 is not opening periodically to release the condensate contained within steam trap 110. That is, if collection line 116 remains below the threshold temperature, these conditions indicate that valve 118 has failed in the closed state.

[0074] Accordingly, if the determination in block 830 is affirmative, server 108 proceeds to block 835. If the determination in block 830 is negative, server 108 proceeds to block 840.

[0075] In block 835, server 108 determines that steam trap 110 has failed with valve 118 in the closed state based on the set of important features obtained in block 625. Then, server 108 proceeds to block 635.

[0076] In block 840, the server 108 determines that the steam trap 110 is functional based on the set of important features obtained in block 625. Then, the server 108 proceeds to block 640 to present the dashboard data to the client device.

[0077] In some embodiments, some of the blocks described above may be skipped or selected based on the configuration of the monitoring device 104, the set of important features received by the server 108, or other factors. For example, if the monitoring device 104 includes a single microphone, the method 800 may proceed directly from block 805 to block 815 to confirm the trap opening failure using vibration data if the audio data indicates a trap opening failure. Also, other combinations are possible.

[0078] Returning to FIG. 6, in block 635, when a failure is detected, the server 108 generates an alert and sends the alert to the client device 120. The alert may be an email notification, a text message, a push notification from a related application, a visual (e.g., pop-up) indicator, an audio indicator, or other suitable alert. The alert may include details of the detected failure, such as the identification information of the steam trap 110 (e.g., the identification name or number including the location of the steam trap 110 within the facility where the steam trap 110 is deployed), the indication information of the type of detected failure (e.g., trap opening failure, trap closing failure, error state / uncertainty failure), etc.

[0079] In block 640, server 108 aggregates a set of important features into dashboard data that is displayed on the visual dashboard of client device 120 along with the previously received set of important features. Then, the dashboard data is output to client device 120. The dashboard data may aggregate display data as a chart, graph, or other visual aid that presents the performance of steam trap 110 to the operator of client device 120. Further, in some embodiments, the dashboard data may aggregate, for example, a set of important features and performance data of a plurality of steam traps all deployed in a particular facility.

[0080] As described above, the monitoring device may be configured to monitor the target device and capture data regarding the target device. The monitoring device applies on-board digital signal processing to reduce the captured data to a set of important features that represent the captured data. The set of important features is selected to be sufficiently concise such that the monitoring device can use LoRa or other low-power wide-area network communication and sufficiently detailed such that the server can perform meaningful analysis.

[0081] In this embodiment, the target device is a steam trap. In other embodiments, the monitoring device may be used to monitor other target devices, which may include, but are not limited to, pumps, motors, or other components that may fail periodically. As will be appreciated later, in such embodiments, the monitoring device may capture data using appropriate sensors to determine the failure of the target device. For example, the sensor subsystem may include an image sensor, an infrared sensor, a microphone, a temperature sensor, and the like. Further, the conditions under which a failure is detected may be selected according to specific failure conditions of the target device (e.g., capturing audio data in different frequency ranges, identifying visual indicators of failure such as a change in the color of a component, etc.).

[0082] The scope of the claims should not be limited by the embodiments described in the above embodiments, but should be given the broadest interpretation consistent with the entire specification.

Claims

1. A monitoring device for a steam trap, comprising: a sensor subsystem housed in a housing and configured to measure characteristics of the steam trap; a memory housed in the housing; a communication interface housed in the housing and configured to communicate with a server; a processor housed in the housing and interconnected to the sensor subsystem, the memory, and the communication interface, wherein the processor is configured to: acquire data representing characteristics of the steam trap from the sensor subsystem; extract a set of significant features from the data; transmit the set of significant features to the server via the communication interface for further processing; wherein the processor is configured to extract the significant features from the data by: sampling audio data at a plurality of points; determining an average magnitude of the sampled audio data; and the set of significant features includes the average magnitude of the audio data. A monitoring device, wherein the set of significant features includes the average magnitude of the audio data.

2. The monitoring device according to claim 1, wherein the sensor subsystem includes a microphone configured to capture audio data representing sound generated by the steam trap.

3. The monitoring device according to claim 2, wherein the housing includes a cylinder oriented between the microphone and the steam trap, and the cylinder is configured to limit audio data captured by the microphone to sound originating from a direction substantially corresponding to the direction of the steam trap.

4. The monitoring device according to claim 2 or claim 3, wherein the sensor subsystem further includes a secondary microphone oriented away from the steam trap, and the secondary microphone is configured to capture secondary audio data representing sound from the environment of the steam trap.

5. The monitoring device according to any one of claims 2 to 4, wherein the microphone is further configured to apply a narrowband filter to the audio data to attenuate frequencies outside a predetermined range.

6. The monitoring device according to any one of claims 1 to 5, wherein the sensor subsystem includes a temperature sensor configured to capture temperature data representing the temperature of the condensate line of the steam trap.

7. The monitoring device according to claim 6, wherein the processor is configured to define the temperature of the agglomeration line as a temperature data point included in the set of the important features.

8. The monitoring device according to any one of claims 1 to 7, wherein the sensor subsystem further includes a secondary temperature sensor configured to capture secondary temperature data representing the internal temperature of the housing.

9. The monitoring device according to claim 8, wherein the processor is configured to define the internal temperature of the housing as a temperature data point included in the set of the important features.

10. The monitoring device according to any one of claims 1 to 9, wherein the sensor subsystem includes an accelerometer configured to capture vibration data representing the vibration received by the monitoring device.

11. The processor is configured to determine the frequency band power of the vibration data within a predetermined time interval, and define the frequency band power as a vibration data point included in the set of the important features. The monitoring device according to claim 10.

12. The monitoring device according to any one of claims 1 to 11, further including a mounting bracket coupled to the housing, wherein the mounting bracket mounts the monitoring device to a fluid line proximate to the steam trap.

13. The mounting bracket includes a mounting arm configured to connect to the fluid line, a passage extending from the mounting arm and configured to accommodate a sensor of the sensor subsystem, and a plate coupled to the passage, spaced apart from the mounting arm, and connected to the housing and configured to support the monitoring device on the mounting bracket. The monitoring device according to claim 12.

14. The monitoring device according to claim 13, wherein the mounting arm has a V shape to reduce heat transfer from the fluid line to the monitoring device.

15. The monitoring device according to any one of claims 1 to 14, wherein the communication interface is configured to use a low power wide area network communication protocol.

16. A method for detecting a failure of a steam trap, comprising: acquiring, by a server, a set of important features representing steam trap data captured by the monitoring device of the steam trap; determining, based on the set of the important features, whether a failure of the steam trap has been detected. When a failure is detected, sending an alert to the client device; outputting dashboard data to the client device, and the method includes: The set of important features includes the average magnitude of audio data points acquired by the sensor system.

17. The step of determining whether the steam trap has failed includes: determining whether an audio data point from the set of important features exceeds a threshold magnitude; when the audio data point exceeds the threshold magnitude, determining that the steam trap has failed in the open state. The method according to claim 16 includes:

18. The step of determining whether an audio data point from the set of important features exceeds a threshold magnitude includes determining whether a threshold ratio of the audio data point exceeds the threshold magnitude. The method according to claim 17 includes:

19. The step of determining whether the steam trap has failed includes: determining whether a secondary audio data point from the set of important features exceeds the threshold magnitude; when the secondary audio data point does not exceed the threshold magnitude, further including the step of confirming that the steam trap has failed in the open state. The method according to claim 17 or claim 18 includes:

20. The step of determining whether the steam trap has failed includes: determining whether a vibration data point from the set of important features exceeds a threshold vibration magnitude; when the vibration data point exceeds the threshold vibration magnitude, further including the step of confirming that the steam trap has failed in the open state. The method according to any one of claims 17 to 19 includes:

21. The step of determining whether the steam trap has failed includes: determining whether a temperature data point from the set of important features is lower than a threshold temperature; when the temperature data point is lower than the threshold temperature, determining that the steam trap has failed in the closed state. The method according to any one of claims 16 to 20 includes:

22. The method according to any one of claims 16 to 21, wherein the alert includes one or more of a mail notification, a text message, a push notification, a visual indicator, and an audio indicator.

23. The method according to any one of claims 16 to 22, wherein the dashboard data includes a set of the important features, which is a set of one or more of a pre-received set of important features and a set of important features of a further steam trap.

24. A system for detecting a failure of a steam trap, comprising: a server; and a monitoring device coupled to the steam trap, the monitoring device including: a sensor subsystem configured to measure characteristics of the steam trap; and a processor interconnected with the sensor subsystem, the processor being configured to: acquire audio steam trap data representing the characteristics of the steam trap from the sensor subsystem; extract a set of important features from the audio steam trap data; be configured to transmit the set of important features to the server; the server being configured to determine whether the steam trap has failed based on the set of important features received from the monitoring device; The fact that the processor is configured to extract the important features from the data means that the processor is configured to: sample the audio steam trap data at a plurality of points; determine an average magnitude of the sampled audio steam trap data; including being configured to perform; The system, wherein the set of important features includes the average magnitude of the audio steam trap data.

Images