Continuous monitoring, early warning and control of key performance indicator variables

Abstract

The method, system, and computer program product involves a computer system collecting sensor data from a manufacturing system, which is measured at intervals shorter than the time interval for a target measurement of the manufacturing system. The sensor data is determined to be related to the target measurement. Based on this relationship, synthetic target measurements are generated at intervals shorter than the time interval. Based on the synthetic target measurements within these intervals, advance warnings for the target measurement are automatically generated.

Description

Technical Field

[0001] The present invention relates generally to computer systems, and more specifically to computer-implemented methods, computer systems, and computer program products configured and arranged to provide continuous monitoring, early warning, and control of process key performance indicator variables with infrequent and time-aggregated measurements. Background Technology

[0002] Manufacturing is the production of products for use or sale using labor, machinery, tools, and chemical or biological processing or formulations. Manufacturing can encompass a range of activities from mechanical to high-tech, but is applied to industrial design, where raw materials, including chemicals, from the primary sector are transformed on a large scale into finished products. Furthermore, the manufacturing process includes steps that transform raw materials into final products. The manufacturing process begins with product design and the material specifications for manufacturing the product. These materials are then modified through the manufacturing process to become the desired components. Modern manufacturing includes all intermediate processes required in the production and integration of product components. While samples or products can be tested in a time-series aggregated manner to check quality, samples or products from the manufacturing process require testing in a more granular manner to enable the early identification of potential problems before normal measurements are taken. Summary of the Invention

[0003] Embodiments of the present invention relate to a computer-implemented method for providing continuous monitoring, early warning, and control of process key performance indicator variables with infrequent and time-aggregated measurements. One non-limiting example of a computer-implemented method includes collecting sensor data from a manufacturing system by a computer system. The sensor data is measured at intervals less than the time interval for a target measurement of the manufacturing system. The sensor data is determined to be related to the target measurement. The computer-implemented method includes generating a synthetic target measurement by the computer system based on the relationship at intervals less than the time interval. Furthermore, the computer-implemented method includes automatically generating an early warning for the target measurement by the computer system based on the synthetic target measurement within the interval less than the time interval.

[0004] In addition to one or more of the features described above or below, the computer-implemented method provides improvements over known methods for monitoring key performance indicator variables with infrequent and time-aggregated measurements by generating synthetic target measurements at intervals less than those for normal target measurements. Therefore, the computer-implemented method described above effectively provides early warning indications about target samples before the normal measurement process, enabling early identification and mitigation of potential problems. Furthermore, the frequency of target measurements is often beyond the control of the system designer, for example, because performing target measurements more rapidly is too expensive or impossible.

[0005] In addition to one or more features described above or below, or as an alternative, further embodiments of the invention may include: wherein one or more setpoints associated with the manufacturing system are automatically modified in response to a prior warning for target measurement based on synthetic target measurement.

[0006] In addition to one or more of the features described above or below, the computer-implemented method provides an improvement over known methods for monitoring key performance indicator variables of infrequent and time-aggregated measurements by adjusting the setpoint based on synthetic target measurements prior to the normal measurement process.

[0007] In addition to one or more features described above or below, or as an alternative, further embodiments of the invention may include: wherein one or more control units associated with the manufacturing system are automatically modified in response to a pre-warning for target measurement based on synthetic target measurement.

[0008] In addition to one or more of the features described above or below, the computer-implemented method provides an improvement over known methods for monitoring key performance indicator variables with infrequent and time-aggregated measurements by adjusting the control components based on synthetic target measurements prior to the normal measurement process.

[0009] In addition to one or more of the features described above or below, or as an alternative, further embodiments of the invention may include: wherein a pre-warning is generated in response to a synthetic target measurement being outside a predetermined range.

[0010] In addition to one or more of the features described above or below, the computer-implemented method provides an improvement over known methods for monitoring key performance indicator variables for measurements that are infrequent and time-aggregated by enabling advance warnings based on synthetic target measurements before the normal measurement process.

[0011] In addition to one or more of the features described above or below, or as an alternative, further embodiments of the invention may include modifications to the manufacturing system to bring the synthetic target measurement within a predetermined range.

[0012] In addition to one or more of the features described above or below, the computer-implemented method provides an improvement over known methods for monitoring key performance indicator variables with infrequent and time-aggregated measurements by making changes to the manufacturing system based on synthetic target measurements prior to the normal measurement process.

[0013] In addition to one or more of the features described above or below, or as an alternative, further embodiments of the invention may include: wherein the target measurement is a quality-related variable corresponding to the physical material process outflow of the manufacturing system. The target measurement has a non-instantaneous, time-accumulated nature due to the mixing of equal sample volumes in the container, wherein the equal sample volumes are collected at multiple moments throughout the time interval and measured at the end of the time interval, thereby obtaining the target measurement.

[0014] In addition to one or more of the features described above or below, the computer-implemented method provides an improvement over known methods for monitoring key performance indicator variables with infrequent and time-aggregated measurements by generating synthetic target measurements that are inherently non-instantaneous and occur prior to the target measurement as instantaneous measurements.

[0015] In addition to one or more features described above or below, or as an alternative, further embodiments of the invention may include: wherein the target measurement is a composite measurement of the aggregated total volume of the sample. The composite target measurement is a generated value based on the sensor data at a time point and is not a measurement of the aggregated total volume of the sample; the composite target measurement represents the state of the individual sampled volume at the time point within the time interval.

[0016] In addition to one or more of the features described above or below, the computer-implemented method provides improvements over known methods for monitoring key performance indicator variables of measurements with infrequent and temporally aggregated data through composite measurements that are independent of the aggregated total volume of the samples. Conversely, synthetic target measurements use sensor data to represent the state of individual sample volumes at any given time point.

[0017] Other embodiments of the present invention implement the features of the above-described method in computer systems and computer program products.

[0018] Additional technical features and advantages are achieved through the technology of this invention. Embodiments and aspects of the invention are described in detail herein and are considered part of the claimed subject matter. For a better understanding, refer to the detailed description and accompanying drawings. Attached Figure Description

[0019] The details of the exclusive rights described herein are specifically pointed out and explicitly claimed in the claims at the end of the specification. The foregoing and other features and advantages of embodiments of the invention will become apparent from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0020] Figure 1A block diagram of an exemplary computer system used in conjunction with one or more embodiments of the present invention is depicted;

[0021] Figure 2 A block diagram is depicted for a system for providing continuous monitoring, early warning, and control of process key performance indicator variables with infrequent and time-aggregated measurements according to one or more embodiments of the present invention;

[0022] Figure 3 The invention describes one or more embodiments of the invention. Figure 2 Further details of the computer system in the document;

[0023] Figure 4 A flowchart is depicted illustrating a computer-implemented process for continuous monitoring, early warning, and control of process key performance indicator variables for infrequent and time-accumulated measurements in a manufacturing system, according to one or more embodiments of the present invention.

[0024] Figure 5 A diagram depicting the instantaneous quality of an unknown potential measurement in a measurement system according to one or more embodiments of the present invention;

[0025] Figure 6 An example representation of a model architecture depicting one or more embodiments of the present invention;

[0026] Figure 7 This is a flowchart of a computer-implemented method for providing continuous monitoring, early warning, and control of process key performance indicator variables with infrequent and time-aggregated measurements, according to one or more embodiments of the present invention.

[0027] Figure 8 A cloud computing environment according to one or more embodiments of the present invention is described; and

[0028] Figure 9 An abstract model layer is described according to one or more embodiments of the present invention. Detailed Implementation

[0029] One or more embodiments of the present invention provide computer-implemented methods, computer systems, and computer program products for determining and / or inferring synthetic / estimated mass measurements of a target sample from available sensor data. One or more embodiments generate synthetic / estimated mass measurements of a target sample from available sensor data prior to the normal measurement of interest. Therefore, one or more embodiments of the present invention provide early indications about the target sample before the normal measurement process, thereby enabling early identification and mitigation of potential problems.

[0030] The measurements of interest, as quality measurements, are collected periodically at certain consistent frequencies. Knowledge exists regarding how the target sample, as the object of the target measurement, is collected and measured. The system of interest (e.g., a manufacturing system) is equipped with multiple sensors that collect data at a higher frequency than the time-accumulated quality measurements of the target sample. One or more embodiments of the present invention improve the timeliness of knowledge about the output quality of the target sample by synthesizing / estimating quality measurements without requiring any further measurements in the manufacturing system.

[0031] Turn now Figure 1 This document generally illustrates a computer system 100 according to one or more embodiments of the present invention. As described herein, the computer system 100 may be an electronic computer framework comprising and / or employing any number and combination of computing devices and networks utilizing different communication technologies. The computer system 100 may be easily expandable, scalable, and modular, with the ability to be changed to different services or reconfigured independently of other features. The computer system 100 may be, for example, a server, desktop computer, laptop computer, tablet computer, or smartphone. In some examples, the computer system 100 may be a cloud computing node. The computer system 100 may be described in the general context of computer system executable instructions such as program modules executed by the computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, etc., that perform specific tasks or implement specific abstract data types. The computer system 100 may be practiced in a distributed cloud computing environment in which tasks are performed by remote processing devices linked via a communication network. In a distributed cloud computing environment, program modules may reside in both local and remote computer system storage media, including memory storage devices.

[0032] like Figure 1As shown, computer system 100 has one or more central processing units (CPUs) 101a, 101b, 101c, etc. (collectively or generally referred to as processors 101). Processor 101 can be a single-core processor, a multi-core processor, a computing cluster, or any number of other configurations. Processor 101 (also referred to as processing circuitry) is coupled to system memory 103 and various other components via system bus 102. System memory 103 may include read-only memory (ROM) 104 and random access memory (RAM) 105. ROM 104 is coupled to system bus 102 and may include a basic input / output system (BIOS) or its successor, such as a unified extensible firmware interface (UEFI), which controls certain basic functions of computer system 100. RAM is read-write memory coupled to system bus 102 for use by processor 101. System memory 103 provides temporary memory space for the operation of the instructions during operation. System memory 103 may include random access memory (RAM), read-only memory, flash memory, or any other suitable memory system.

[0033] Computer system 100 includes an input / output (I / O) adapter 106 and a communication adapter 107 coupled to a system bus 102. I / O adapter 106 may be a Small Computer System Interface (SCSI) adapter that communicates with a hard disk 108 and / or any other similar component. I / O adapter 106 and hard disk 108 are collectively referred to herein as mass storage device 110.

[0034] Software 111 executing on computer system 100 may be stored in mass storage device 110. Mass storage device 110 is an example of a tangible storage medium readable by processor 101, wherein software 111 is stored as instructions for execution by processor 101 to operate computer system 100, as described below with respect to the various figures. Examples of computer program products and the execution of such instructions are discussed in more detail herein. Communication adapter 107 interconnects system bus 102 with network 112, which may be an external network, enabling computer system 100 to communicate with other such systems. In one embodiment, a portion of system memory 103 and mass storage device 110 jointly store an operating system, which may be used for coordination Figure 1 The functions of the various components shown can be applied to any suitable operating system.

[0035] Additional input / output devices are shown connected to the system bus 102 via display adapter 115 and interface adapter 116. In one embodiment, adapters 106, 107, 115, and 116 may be connected to one or more I / O buses connected to the system bus 102 via an intermediate bus bridge (not shown). A display 119 (e.g., a screen or display monitor) is connected to the system bus 102 via display adapter 115, which may include a graphics controller and a video controller to improve performance in graphics-intensive applications. Keyboard 121, mouse 122, speakers 123, etc., may be interconnected to the system bus 102 via interface adapter 116, which may include, for example, a super I / O chip that integrates multiple device adapters into a single integrated circuit. Suitable I / O buses for connecting peripheral devices such as hard disk controllers, network adapters, and graphics adapters typically include common protocols such as Peripheral Component Interconnect (PCI) and High-Speed ​​Peripheral Component Interconnect (PCIe). Therefore, as Figure 1 The computer system 100 configured includes processing power in the form of a processor 101, storage capacity including system memory 103 and mass storage device 110, input devices such as keyboard 121 and mouse 122, and output capacity including speaker 123 and display 119.

[0036] In some implementations, communication adapter 107 may use any suitable interface or protocol (such as Internet Small Computer System Interface) to send data. Network 112 may be a cellular network, radio network, wide area network (WAN), local area network (LAN), or the Internet. External computing devices may be connected to computer system 100 via network 112. In some examples, the external computing device may be an external web server or a cloud computing node.

[0037] It should be understood that Figure 1 The block diagram is not intended to indicate that the computer system 100 includes Figure 1 All components shown. Conversely, computer system 100 may include... Figure 1 Any suitable fewer or additional components not shown herein (e.g., additional memory components, embedded controllers, modules, additional network interfaces, etc.). Furthermore, the embodiments described herein with respect to computer system 100 can be implemented with any suitable logic, wherein, in various embodiments, the logic mentioned herein may include any suitable hardware (e.g., processor, embedded controller, or application-specific integrated circuit, etc.), software (e.g., applications, etc.), firmware, or any suitable combination of hardware, software, and firmware.

[0038] Figure 2This is a block diagram of a system 200 for providing continuous monitoring, early warning, and control of process key performance indicator variables with infrequent and time-aggregated measurements according to one or more embodiments of the present invention. System 200 includes one or more computer systems 202 connected to a manufacturing system 230. The manufacturing system 230 can be operated using control theory, which uses control dynamics systems in engineering processes and machines. The manufacturing system 230 can utilize automated process control in continuous production processes. Automated process control is a combination of control engineering and chemical engineering disciplines that use industrial control systems to achieve consistent, economical, and safe production levels, and is widely implemented in industries such as oil refining, pulp and paper manufacturing, chemical processing, and power plants. The manufacturing system 230 may include various system components 232, wherein system components 232 represent various devices for converting inputs into target outputs, as known to those skilled in the art. System components 232 may include electrical equipment, mechanical equipment, chemical equipment, etc. System components 232 represent machines or machine systems required for use as the manufacturing system 230 and / or the plant.

[0039] Manufacturing system 230 includes a control component 234 (or controller) for controlling the function of system component 232. Control component 234 relates to any component that can be controlled, changed, and / or modified within manufacturing system 230. Exemplary control component 234 may include actuators, control values, relays, switches, etc. A setpoint 236 (also a setting point) is used in manufacturing system 230. A setpoint is an expected or target value for a critical variable or process value of manufacturing system 230. Control component 234 may include one or more setpoints 236 that control the operation of the corresponding control component 234. Setpoints 236 can be modified to change the operation and settings of control component 234. One or more control systems 240 are located in manufacturing system 230. Control systems 240 are used to control the function and operation of control component 234, thereby controlling the function and operation of system component 232. Control systems 240 can be used to modify setpoints 236 of manufacturing system 230. Control systems use control loops to manage, command, direct, and / or regulate the behavior of other devices or systems (such as, for example, control component 234). For continuous modulation control, a feedback controller is used to automatically control the process or operation. The control system compares the value or state of the controlled process variable with the desired value or setpoint and applies the difference as a control signal to make the process variable output of the plant (e.g., manufacturing system 230) reach the same value as the setpoint.

[0040] In manufacturing system 230, a quality measurement 250 of the target product is performed, which can be at any desired stage of manufacturing. Quality measurement 250 is the measurement of interest and is periodically collected from the output of a given one of system components 232 at a consistent frequency. For example, quality measurements may be performed every “T” hours (e.g., T = 12 hours, 24 hours, 1 week, etc.). In the process industry, maintaining the quality of output products can include ensuring compliance with any regulatory constraints, including federal standards set by regulatory bodies. Typically, this is accomplished by periodically performing expensive and time-consuming laboratory measurements based on aggregated samples collected over time. Depending on the results of the laboratory measurements, the target product may have insufficient quality and / or may have violated regulatory constraints. Either case is a costly error. Therefore, quality measurement 250 is performed to meet given requirements. Quality measurement 250 is a true aggregated measurement. For example, quality measurement 250 is based on a combination of small volumes collected over time (e.g., from system component 232) and combined in the same sample container 252. Therefore, a single quality measurement 250 consists of a measurement of the time-gathered sample / target product in sample container 252, where the sample in sample container 252 is a time-gathered sample. Subsequent measurements of the time-gathered amount in sample container 252 also add an additional delay.

[0041] Manufacturing system 230 is equipped with various sensors 238 coupled to system component 232, and sensors 238 measure (i.e., collect) data at a higher frequency than the measurements of mass measurement 250. Control system 240 may be coupled to sensor 238 and receive measurements (i.e., sensor data) from sensor 238. Sensor measurements may be measured at time increments of T' (e.g., minutes), where T' is less than T. For example, sensor measurements performed by sensor 238 may be measured / acquired every five minutes, ten minutes, etc. Sensor measurements performed by sensor 238 may be measured / acquired every hour, two hours, etc. Sensor measurements of sensor 238 do not measure the actual sample collected in sample container 252. Instead, sensor 238 may provide measurements and readings of various equipment, materials, flow rates, etc., at various / different stages of the manufacturing process to produce samples in sample container 252. Sensor measurements and readings may be used for any type of measurable value in and / or related to manufacturing system 230. For example, sensor measurements and readings can be used for ore quality, temperature, density, flow rate, voltage, current, speed, revolutions per minute, vibration, etc.

[0042] Computer system 202 is connected to manufacturing system 230. Computer system 202 may be coupled to control system 240, sensors 238, control components 234, and / or setpoints 236. In one or more embodiments, control system 240 may include one or more control applications 242 configured to interface with software application 204 of computer system 202. Furthermore, control applications 242 may monitor and control system components, control components 234, setpoints 236, and sensors 238, as understood by those skilled in the art. Figure 3 Further details of the computer system 202 are shown below. In one or more embodiments, the computer system 202 may be implemented and / or integrated into the control system 240. Figure 1 As shown, software application 204 can be implemented as software 111 that executes on one or more processors 101. Similarly, control application 242 can be implemented using software 111 configured to execute on one or more processors 101. Components of computer system 100 can be used in and / or integrated into computer system 202 and control system 240.

[0043] Figure 4 A flowchart is depicted of a computer-implemented process 400 that provides continuous monitoring, early warning, and control of process key performance indicator variables with infrequent and time-aggregated measurements during the manufacturing process, according to one or more embodiments of the present invention. Figure 4 The computer-implemented process 400 can be used Figure 2 The system 200 shown and Figure 3 This will be implemented using computer system 202. Therefore, reference will now be made to... Figure 2 System 200 and computer system 202 describe the computer-implemented process 400.

[0044] At block 402, software application 204 of computer system 202 is configured to collect sensor data from sensor 238 in manufacturing system 230. The sensor data from sensor 238 is measured at a higher frequency (e.g., at shorter intervals or time "T") than the measurements of mass measurement 250 (e.g., measured at intervals or time "T"). In one or more embodiments, software application 204 may receive sensor data from sensor 238 from control application 242 of control system 240. Sensor data from sensor 238 may be pushed to and / or pulled by software application 204 of computer system 202. In one or more embodiments, control system 240 may buffer and / or store sensor data from the corresponding sensor 238 before sending the sensor data to computer system 202. In one or more embodiments, software application 204 may receive sensor data from sensor 238 in real-time and / or near real-time.

[0045] At block 404, software application 204 of computer system 202 is configured to slice / divide sensor data of sensor 238 into blocks (e.g., tensor slices), where different blocks correspond to different groups / time periods of measurements performed by sensor 238. At block 406, software application 204 of computer system 202 is configured to use blocks of sensor data (e.g., tensor slices) from different groups / time periods as input to model 306, which is configured to determine synthetic / estimated quality measurements for manufacturing system 230. Synthetic / estimated quality measurements are generated at a higher frequency than measurements of quality measurement 250. In other words, synthetic / estimated quality measurements are generated more frequently than measurements of quality measurement 250. The synthetic / estimated quality measurements also have less hysteresis and / or delay than quality measurement 250 as a physical measurement.

[0046] In box 408, software application 204 of computer system 202 is configured to check whether the synthesized / estimated quality measurement is within the normal range of manufacturing system 230. The normal range is predetermined. The normal range may have a lower limit and an upper limit, and the synthesized / estimated quality measurement is assumed to remain within the lower and upper limits considered normal. If the synthesized / estimated quality measurement is within the normal range (“Yes”), the process proceeds to box 402 for continued monitoring. Continued monitoring includes the continuous generation of synthesized / estimated quality measurements. If the synthesized / estimated quality measurement is outside the normal range (“No”), software application 204 is configured to issue an advance warning (e.g., notification 280) based on one or more synthesized / estimated quality measurements falling outside the normal range. For example, software application 204 may issue a warning to control application 242 of control system 240 and the operator. The advance warning is an early indicator of a problem in manufacturing system 230, including potential problems with the samples in sample container 252, prior to the time period “T” used for performing quality measurement 250 of the samples in sample container 252.

[0047] At block 410, software application 204 of computer system 202 is configured to cause modification of one or more setpoints 236 and / or one or more control elements 234 in manufacturing system 230. In one or more embodiments, software application 204 may cause, instruct, and / or request changes to one or more setpoints 236 and / or one or more control elements 234. In one or more embodiments, software application 204 may transmit a request (and / or notification 280) to control system 240, such that the request causes control application 242 of control system 240 to modify setpoints 236 and / or control elements 234. In one or more embodiments, software application 204 may cause values, operations, and / or functions associated with one or more setpoints 236 and / or one or more control elements 234 to increase and / or decrease due to notification 280.

[0048] Therefore, based on notification 280 as an advance warning, the operation of manufacturing system 230 can be improved and errors can be avoided. Synthetic / estimated mass measurements are generated using model 306 at shorter intervals and / or shorter time periods than the measurements of mass measurement 250. Thus, the synthetic / estimated mass measurements generated using sensor data from sensor 238 provide an early indication of the target sample collected in sample container 252 prior to normal mass measurement 250. As noted herein, mass measurement 250 is based on a delayed time aggregation of the sample collected in sample container 252, while the synthetic / estimated mass measurement is an instantaneous representation of the sample at any time point less than the time “T” used to perform / measure mass measurement 250. Therefore, by using synthetic / estimated mass measurements at different time points, software application 204 is configured to generate / infer finer-grained information about the actual mass measurement 250, thereby enabling early identification of potential problems in manufacturing system 230 and, consequently, providing mitigation.

[0049] Figure 5 This is a block diagram illustrating the modeling of the instantaneous quality of unknown latent measurements in a manufacturing system 230 according to one or more embodiments of the present invention. The unknown latent measurements are used to generate synthetic / estimated quality measurements. Model 306 is a physics-based model associated with the manufacturing system 230. Figure 5 In the diagram, various measurements / readings of different sensors 238 are shown by circles along the timeline. Each measurement / reading of sensor 238 occurs at a time less than the interval between measurements performed for mass measurement 250. The circles on timeline 502 represent mass measurement 250 performed at each time interval or period "T". For illustrative purposes, sensor 238 may include sensor A, sensor B through sensor N, where N represents the last sensor, and each sensor has multiple measurements on the timeline. Each circle on timeline 504 represents an unknown potential result that can provide additional knowledge about manufacturing system 230. The circles on timeline 504 are not measured in manufacturing system 230 but are derived from the relationship between sensor data of sensor 238 and mass measurement 250.

[0050] When training model 306, software application 204 uses historical data 310 (e.g., stored in memory 308) to learn the relationship between sensor data of sensor 238 and quality measurement 250 over the same time period "T", and repeats this process continuously. Historical data 310 includes historical sensor data of sensor 238 and historical quality measurement 250 aligned in time. Historical data 310 stored in a database can represent many databases. A database can contain hundreds, thousands, and / or millions of documents, also known as "big data". According to one or more embodiments, the enormous size of historical data 310 in the database requires a machine (such as computer system 202) to manage, process, and search it, for example, using computer-executable instructions, and historical data 310 in the database cannot be practically managed, stored, analyzed, and / or processed in the human mind as discussed herein.

[0051] When training model 306, software application 204 is configured to fit a regression model that correlates high-frequency covariates (which are individual measurements of sensor data from sensor 238) over measurement time period "T" with delayed low-frequency measurements 250 (e.g., delayed low-frequency (laboratory) results) via unknown intermediate quality results on timeline 504. For illustrative purposes, y is a symbol indicating quality measurement 250, where y t This represents the mass measurement value of 250 at time t. The unknown intermediate mass result at time t is represented as q. t =f(X) t-T’ :t), with an initial unknown relation f() and an unknown output. Function y t =g(q) t-T The model 306 learns a function f() and uses f() later during prediction. In one or more embodiments, the function g() is known because the mass measurement 250 is based on equal volumes sampled uniformly over time, and in this case, the model 306 considers the average mass over the observation window “T”. In this case, the final sample delivered for measurement in the sample container 252 consists of a homogeneous mixture of samples over time, and the target measurement results in an average measurement of the mass.

[0052] During training, sensor data (i.e., measurements / readings) from sensor 238 are aligned to match the time period "T" during which samples are collected in sample container 252, where mass measurement 250 is performed on the samples collected in sample container 252. For each mass measurement 250, the sample data from sensor 238 is used within the same time period "T". Once model 306 is trained, it can be used to generate a synthetic / estimated mass measurement (e.g., synthetic y) for each mass measurement 250 at time intervals shorter than the time period "T". Software application 204 can apply (train) model 306 to a window of length T' with different numbers (potentially overlapping) of covariates to predict unknown intermediate results q. t And then process those intermediate results (q) t This generates an average quality measurement, which is a synthesized / estimated average quality measurement for a specific time interval / window smaller than the original quality measurement 250(T). As noted herein, this is particularly useful in providing more timely alerts to the system operators of manufacturing system 230. Figure 5 In this model, any desired or estimated quality measurement for a given time window can be generated using one or more blocks of sensor data (e.g., tensor slices). Model 306 can be implemented using various regression models.

[0053] Figure 6 This is an example representation of the model architecture 600 of model 306 according to one or more embodiments of the present invention. Figure 6 The use of a neural network is illustrated. For a total of "m" blocks, sensor data blocks, denoted by "X", from sensor 238 are input to model 306. The sensor data blocks are grouped according to matching batch / time groups. Sensor data from multiple sensors 238 are combined in each block. In each of these matrices, a block can be viewed as a matrix (e.g., a second-order tensor) where columns consist of individual sensors and each row has a separate time point. For example, by assigning X... t-T’：t A sensor data block, a sensor data block (e.g., a tensor slice) can be X t-T’-m+1：t-m+1 Another sensor data block could be X t-T’-m+2：t-m+2 As noted in this paper, sensor data measurements are performed in increments of time "T", and "m" is the number of sensor data blocks (e.g., tensor slices) used to generate a single synthetic / estimated quality measurement at time "t", where time "t" is the moment typically used to refer to the current time. (See reference...) Figure 6Each sensor data block is fed into a copy of the neural network representing the function f(). Multiple neural networks are identical copies of the same trained neural network, responsible for implementing and / or capturing local dynamics, as discussed in this paper. After each sensor data block is fed into the function f() of the copy of the neural network, the copy outputs an unknown intermediate quality result q, which is the q of each neural network. t For example, for a neural network that receives corresponding sensor data blocks, a neural network can generate q. t-m+1 Another neural network can generate q t-m+2 q is generated through the last neural network. t The known aggregation function g() is used to aggregate the output from copies of the neural network. For a given time "t", the known output "y" of the aggregation function g() is a single synthetic / estimated quality measurement of manufacturing system 230, where the synthetic / estimated quality measurement represents the condition / state of manufacturing system 230. The output of g() is the quality over time period T (i.e., the aggregated measurement of quality over the time period from tT to t). Software application 204 can utilize model 306 to generate synthetic / estimated quality measurements for different times "t", where time "t" can be a smaller interval than the time period "T" used for quality measurement 250.

[0054] Quality measurement 250 can be considered a measurement of a process variable. Sensor data from sensor 238 can be considered a covariate and / or other process variables different from the process variable. Using model 306, the non-instantaneous nature of the process variable measurement takes the general form of time-aggregated measurement, whose time resolution and value correspond to the aggregation of the instantaneous values ​​of the variable over a time interval using a general aggregation function (e.g., function g()) over that time interval. This is in contrast to the case of instantaneous online and / or offline measurements, where the time resolution and value correspond to a point in time.

[0055] The process variables under discussion (e.g., corresponding to mass measurement 250) are mass-related variables corresponding to the physical material process effluent, the measurements of which are rarely obtained from the laboratory; the non-transient, time-aggregated nature of the mass measurement is specifically due to the mixing / aggregation of equal sampling volumes (e.g., into sample container 252), which are drawn from the corresponding process effluent at multiple moments over a long period and subsequently composite measurements are performed on the aggregated total volume. As learned by model 206, the non-transient, time-aggregated nature of mass measurement 250 is due to any specified general aggregation function applied at each aggregation interval corresponding to the historical process variable measurements (historical data 310), such as the relationship determined in model 306. The non-transient, time-aggregated nature of mass measurement 250 is due to an unspecified general aggregation function applied uniformly at each aggregation interval corresponding to the historical process variable measurements and automatically learned by model 306 from historical data 310.

[0056] By using Model 306, instantaneous, point-in-time values ​​of process variables (e.g., q) are automatically estimated. t The value of the process variable is used for continuous monitoring of 250 process variables for quality measurement. Relationships for generating time-point values ​​(e.g., synthetic / estimated quality measurements) are automatically learned from historical data by reconstructing a set of historical base-fact measurements aggregated over infrequent time intervals: this is achieved by first constructing unobserved potential time-point instantaneous values ​​of the process variable at corresponding time intervals (e.g., q). t The value of q is obtained by transforming these latent estimates (e.g., q) using a general aggregation function (e.g., function g()). t The potential time-point estimate of the process variable at any given time is the result of model 306, which takes the values ​​of all covariates (e.g., sensor data blocks of sensor 238) as input to a window of historical process influence relative to each such time point, where the length of this window (e.g., time “T”) is chosen as a model hyperparameter, independent of the duration (e.g., time period “T”) between consecutive (infrequent) measurements of the process variable.

[0057] like Figure 6As described, model 306 can be a neural network with multiple layers of nonlinear activation functions, weights, and biases. Model 306 uses both the raw values ​​of all covariates (e.g., sensor data) at a window (e.g., time “T”) influenced by the historical process, and various time-series features (including mean, standard deviation, kurtosis, variance, etc.) on the sequence of historical covariate values ​​within this window as input. Automatic learning of a general aggregation function (e.g., function g()) is achieved using a neural network with multiple layers of nonlinear activation functions, weights, and biases, and the general aggregation function (e.g., function g()) uses a latent estimate at time points (e.g., q...). t Values, such as q t-m+1 q t-m+2 to q t () as input.

[0058] According to one or more embodiments, the technical benefits and advantages include systems and methods that are better aligned with the fundamental physics of the manufacturing process. Estimated potential quality results (e.g., q) t The estimated quality result (e.g., q) depends on sensor measurements over a shorter time span. t This effectively outlines the local behavior of a process in a way that can be accurately used to capture longer-term dynamic characteristics present in average quality measurements, and therefore, the estimated quality results (e.g., q) t This is used to generate the synthesized / estimated quality measurement. Knowledge of potential quality results allows for the development of alternative time-based averages that may be more beneficial than long-term averages in ensuring process behavior.

[0059] Figure 7 This is a flowchart of a computer-implemented process 700 for providing continuous monitoring, early warning, and control of process key performance indicator variables with infrequent and time-aggregated measurements in a manufacturing system 230, according to one or more embodiments of the present invention. Figure 7 The computer-implemented process 700 can be used Figure 2 The system 200 shown and Figures 3-6 The discussion in the document will be implemented accordingly.

[0060] At box 702, software application 204 on computer system 202 is configured to collect sensor data from manufacturing system 230 at intervals (e.g., time interval "T") shorter than the time interval (e.g., time period "T") of a target measurement (e.g., quality measurement 250) of manufacturing system 230, wherein the sensor data is determined to be related to the target measurement (e.g., quality measurement 250). For example, software application 204 may collect sensor data from sensor 238 from control system 240 and / or directly from sensor 238.

[0061] In box 704, software application 204 on computer system 202 is configured to generate synthetic target measurements at intervals (e.g., time "T'") smaller than the relationship (e.g., time period "T"). For example, the relationship may include a function f(), a function g(), and a relationship between the mass measurement 250, where a block of sensor data ("X") is used as input. Model 306 can be used to generate synthetic target measurements representing the mass measurement 250 within a given time period.

[0062] At box 706, software application 204 on computer system 202 is configured to automatically generate advance warnings (e.g., notification 280) for target measurements based on synthetic target measurements within an interval (e.g., time “T”) that is less than a time interval.

[0063] In response to a pre-warning (e.g., notification 280) for a target measurement based on a synthetic target measurement, one or more setpoints 236 associated with manufacturing system 230 are automatically modified. In response to a pre-warning (e.g., notification 280) for a target measurement based on a synthetic target measurement, one or more control components 234 associated with manufacturing system 230 are automatically modified. A pre-warning (e.g., notification 280) is generated (by computer system 202) because the synthetic target measurement is outside a predetermined range (e.g., a pre-determined normal range). Modifications are made to manufacturing system 230 (e.g., caused and / or indicated by computer system 202) to bring the synthetic target measurement within the predetermined range.

[0064] The target measurement (e.g., mass measurement 250) is a quality-related variable corresponding to the physical material process outflow of the manufacturing system 230. Due to the mixing of equal sample volumes in a vessel (e.g., sample container 252), the target measurement has a non-transient, time-gathering nature, wherein the equal sample volumes are collected at multiple moments throughout the entire time interval (e.g., time period “T”) and measured at the end of the time interval to obtain the target measurement (e.g., mass measurement 250).

[0065] The target measurement (e.g., mass measurement 250) is a composite measurement of the aggregated total volume of the sample (e.g., collected in sample container 252). The composite target measurement is a generated value based on sensor data at a specific time point and is not a measurement of the aggregated total volume of the sample, but rather a composite target measurement representing the state of an individual sampled volume at a specific time point within a time interval.

[0066] The target measurement (e.g., mass measurement 250) is a composite measurement of the total aggregate volume of the samples (e.g., collected in sample container 252). The aggregation function (e.g., aggregation function g()) that generates the composite target measurement from the equal sampling volumes constituting the total aggregate volume is unknown and is automatically learned as a function of the target measurement across the equal sampling volumes. For a known aggregation function (g()), since the function is known in this case, the training process incorporates this information, as in... Figure 6 As described herein, and as previously discussed. In one or more embodiments, there may be an unknown aggregation function (i.e., the aggregation function (g()) is unknown), and in this case, the system designer / engineer may not know g(). Accordingly, the software application 204 and / or model 306 on computer system 202 are configured to also learn g() in the process. In this case, Figure 6 The aggregation function g() in the model is replaced by another neural network whose parameters need to be learned. The synthetic target measurement (e.g., model 306 via software application 204) is a time-point value generated based on sensor data (e.g., from manufacturing system 230) and is not a measurement of the total aggregate volume of the sample. The synthetic target measurement represents the state of individual sample volumes (e.g., collected in sample container 252) at time points within a time interval (e.g., time period “T”). It should be noted that the synthetic target measurement is synthetic in the sense of its output based on the learning model (f()). This can be a time-point measurement (i.e., instantaneous mass) and / or a new aggregation generated at a finer granularity (i.e., 1-hour average mass). In one case, the time-point measurement can correlate (e.g., coincide or nearly coincide) the time of individual sample volumes, thereby providing insight into what is happening in manufacturing system 230 and / or the state at that time.

[0067] The target measurement (e.g., mass measurement 250) is a composite measurement of the total aggregate volume of the sample (e.g., collected in sample container 252). It is known that the aggregation function (e.g., aggregation function g()) that generates the composite target measurement from the equal sample volumes constituting the total aggregate volume is the average of the target measurements across the equal sample volumes. The synthetic target measurement (e.g., model 306 via software application 204) is a time-point generated value based on sensor data and is not a measurement of the total aggregate volume of the sample; the synthetic target measurement represents the state of individual sample volumes within the equal sample volumes at a time point within a time interval (e.g., time period “T”).

[0068] A target measurement (e.g., mass measurement 250) is a composite measurement of the total aggregate volume of the sample (e.g., collected in sample container 252). An aggregation function (e.g., aggregation function g()) that generates the composite target measurement from the equal sample volumes constituting the total aggregate volume is known from a user-specified function of the target measurement across equal sample volumes. A synthetic target measurement (e.g., model 306 via software application 204) is a time-point generated value based on sensor data and is not a measurement of the total aggregate volume of the sample; the synthetic target measurement represents the state of individual sample volumes within the equal sample volumes at a time point within a time interval (e.g., time period “T”).

[0069] It should be understood that while this disclosure includes a detailed description of cloud computing, the implementation of the teachings cited herein is not limited to cloud computing environments. Rather, embodiments of the invention can be implemented in conjunction with any other type of computing environment now known or developed hereafter.

[0070] Cloud computing is a service delivery model that enables convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services), which can be rapidly provisioned and released with minimal management effort or interaction with the service provider. This cloud model may include at least five features, at least three service models, and at least four deployment models.

[0071] The features are as follows:

[0072] On-demand self-service: Cloud consumers can unilaterally and automatically provide computing power, such as server time and network storage, as needed, without requiring human interaction with the service provider.

[0073] Extensive network access: Capabilities are available through networks and accessed via standard mechanisms that facilitate the use of heterogeneous thin client or thick client platforms (e.g., mobile phones, laptops, and PDAs).

[0074] Resource pooling: A provider's computing resources are pooled to serve multiple consumers using a multi-tenant model, where different physical and virtual resources are dynamically assigned and reassigned as needed. There is a sense of location independence because consumers typically do not have control or knowledge of the exact location of the resources provided, but may be able to specify the location at a higher level of abstraction (e.g., country, state, or data center).

[0075] Rapid flexibility: The ability to provide capacity quickly and flexibly, automatically scaling down and up rapidly in some situations to scale up rapidly. For consumers, the available supply capacity often appears unlimited and can be purchased in any quantity at any time.

[0076] Measuring services: Cloud systems automatically control and optimize resource usage by leveraging metering capabilities at a level of abstraction appropriate to the service type (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency to both service providers and consumers.

[0077] The service model is as follows:

[0078] Software as a Service (SaaS): This provides consumers with the ability to use the provider's applications running on cloud infrastructure. Applications can be accessed from different client devices via thin client interfaces such as web browsers (e.g., web-based email). Consumers do not manage or control the underlying cloud infrastructure, including the network, servers, operating system, storage, or even individual application capabilities, with possible exceptions such as limited user-specific application configuration settings.

[0079] Platform as a Service (PaaS): This provides consumers with the ability to deploy applications created or acquired by the consumer using programming languages ​​and tools supported by the provider onto cloud infrastructure. Consumers do not manage or control the underlying cloud infrastructure, including networks, servers, operating systems, or storage, but they have control over the deployed applications and the configuration of any application hosting environment.

[0080] Infrastructure as a Service (IaaS): The capabilities offered to consumers are processing, storage, networking, and other basic computing resources that enable consumers to deploy and run arbitrary software, which may include operating systems and applications. Consumers do not manage or control the underlying cloud infrastructure, but rather have control over the operating system, storage, deployed applications, and potentially limited control over selected networking components (e.g., host firewalls).

[0081] The deployment model is as follows:

[0082] Private cloud: A cloud infrastructure that operates solely for an organization. It can be managed by the organization or a third party and can exist on-site or off-site.

[0083] Community cloud: A cloud infrastructure shared by several organizations and supporting a specific community with shared concerns (e.g., tasks, security requirements, policies, and compliance considerations). It can be managed by an organization or a third party and can exist on-site or off-site.

[0084] Public cloud: Makes cloud infrastructure available to the public or large industry groups and is owned by an organization that sells cloud services.

[0085] Hybrid cloud: A cloud infrastructure is a combination of two or more clouds (private, community, or public) that remain a single entity but are bound together by standardized or proprietary technologies that enable data and applications to be ported (e.g., cloud bursting for load balancing between clouds).

[0086] Cloud computing environments are service-oriented, focusing on statelessness, loose coupling, modularity, and semantic interoperability. At the heart of cloud computing is the infrastructure comprising a network of interconnected nodes.

[0087] See now Figure 8 This describes an illustrative cloud computing environment 50. As shown, the cloud computing environment 50 includes one or more cloud computing nodes 10 that can communicate with local computing devices used by cloud consumers, such as, for example, personal digital assistants (PDAs) or cellular phones 54A, desktop computers 54B, laptop computers 54C, and / or automotive computer systems 54N. The nodes 10 can communicate with each other. They can be physically or virtually grouped (not shown) in one or more networks, such as private clouds, community clouds, public clouds, or hybrid clouds (as described above), or combinations thereof. This allows the cloud computing environment 50 to provide infrastructure, platforms, and / or software as services that cloud consumers do not need to maintain on their local computing devices. It should be understood that... Figure 8 The types of computing devices 54A-N shown are intended to be illustrative only, and computing node 10 and cloud computing environment 50 can communicate with any type of computerized device via any type of network and / or network-addressable connectivity (e.g., using a web browser).

[0088] See now Figure 9 This demonstrates a cloud computing environment of 50 ( Figure 8 This provides a set of functional abstractions. It should be understood beforehand that... Figure 9 The components, layers, and functions shown are merely illustrative, and embodiments of the invention are not limited thereto. As described, the following layers and corresponding functions are provided:

[0089] The hardware and software layer 60 includes hardware and software components. Examples of hardware components include: a mainframe 61; a RISC (Reduced Instruction Set Computer) based server 62; a server 63; a blade server 64; a storage device 65; and networking and interconnection components 66. In some implementations, software components include network application server software 67 and database software 68.

[0090] The virtualization layer 70 provides an abstraction layer from which the following examples of virtual entities can be provided: virtual server 71; virtual storage 72; virtual network 73, including virtual private network; virtual application and operating system 74; and virtual client 75.

[0091] In one example, management layer 80 may provide the following functionalities: Resource Provisioning 81 provides dynamic procurement of computing resources and other resources used to perform tasks within the cloud computing environment. Metering and Pricing 82 provides cost tracking as resources are utilized within the cloud computing environment and bills or invoices for the consumption of these resources. In one example, these resources may include application software licenses. Security provides authentication for cloud consumers and tasks, as well as protection for data and other resources. User Portal 83 provides access to the cloud computing environment for consumers and system administrators. Service Level Management 84 provides cloud resource allocation and management to ensure that required service levels are met. Service Level Agreement (SLA) Planning and Fulfillment 85 provides pre-scheduling and procurement of cloud resources based on anticipated future needs according to the SLA.

[0092] Workload layer 90 provides examples of functionalities that can leverage a cloud computing environment. Examples of workloads and functionalities that can be provided from this layer include: mapping and navigation 91; software development and lifecycle management 92; virtual classroom education delivery 93; data analytics and processing 94; transaction processing 95; and software applications implemented in workloads and functionalities 96 (e.g., software application 204, control application 242, model 306, etc.).

[0093] Various embodiments of the invention are described herein with reference to the accompanying drawings. Alternative embodiments of the invention may be devised without departing from the scope thereof. In the following description and drawings, various connections and positional relationships (e.g., above, below, adjacent, etc.) are illustrated between elements. Unless otherwise specified, these connections and / or positional relationships may be direct or indirect, and the invention is limited in this respect by not illustrating the figures. Thus, the connection of entities may refer to direct or indirect connections, and the positional relationship between entities may be direct or indirect positional relationships. Furthermore, the various tasks and process steps described herein may be incorporated into a more comprehensive procedure or process with additional steps or functions not described in detail herein.

[0094] One or more methods described herein can be implemented using any of the following techniques or combinations thereof, each of which is well known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having appropriately combined logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0095] For the sake of brevity, conventional techniques relating to the manufacture and use of the present invention may or may not be described in detail herein. Specifically, various aspects of the computing systems and specific computer programs used to implement the different technical features described herein are well known. Consequently, for the sake of brevity, many conventional implementation details are only briefly mentioned or omitted entirely herein, without providing well-known system and / or process details.

[0096] In some implementations, various functions or actions may occur at a given location and / or in conjunction with the operation of one or more devices or systems. In some implementations, a portion of a given function or action may be performed at a first device or location, and the remainder of the function or action may be performed at one or more additional devices or locations.

[0097] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well. It should also be understood that when the terms “comprises” and / or “comprising” are used in this specification, they specify the presence of the stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or combinations thereof.

[0098] All means or steps in the following claims, along with corresponding structures, materials, actions, and equivalents of the functional elements, are intended to include any structure, material, or action for performing the function in conjunction with other claimed elements as specifically claimed. This disclosure has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art without departing from the scope of this disclosure. These embodiments were chosen and described in order to best explain the principles and practical application of this disclosure, and to enable others skilled in the art to understand this disclosure with different embodiments having different modifications suitable for the particular intended use.

[0099] The diagrams depicted herein are illustrative. Many variations may be made to the diagrams or steps (or operations) described herein without departing from the scope of this disclosure. For example, actions may be performed in a different order, or actions may be added, deleted, or modified. Furthermore, the term "coupled" describes a signal path between two elements and does not imply a direct connection between elements without intermediate elements / connections. All such variations are considered part of this disclosure.

[0100] The following definitions and abbreviations will be used to interpret the claims and description. As used herein, the terms “comprising,” “including,” “comprise,” “having,” “possessing,” “containing,” or “encompassing,” or any other variation thereof, are intended to cover non-exclusive inclusion. For example, a composition, mixture, process, method, article, or apparatus that comprises a list of elements is not necessarily limited to those elements, but may include other elements not expressly listed or inherent to such composition, mixture, process, method, article, or apparatus.

[0101] Furthermore, the term "exemplary" is used herein to mean "used as an example, illustration, or illustration." Any implementation or design described herein as "exemplary" is not necessarily to be construed as superior to or better than other implementations or designs. The terms "at least one" and "one or more" should be understood to include any integer greater than or equal to one, i.e., one, two, three, four, etc. The term "multiple" should be understood to include any integer greater than or equal to two, i.e., two, three, four, five, etc. The term "connection" can include both indirect "connection" and direct "connection."

[0102] The terms “about,” “substantially,” “roughly,” and their variations are intended to include the degree of error associated with a measurement of a specific quantity based on the equipment available at the time of application submission. For example, “about” could include a range of ±8%, 5%, or 2% of a given value.

[0103] This invention can be a system, method, and / or computer program product with any possible level of technical detail integration. The computer program product may include a computer-readable storage medium (or media) having computer-readable program instructions thereon for causing a processor to execute aspects of the invention.

[0104] Computer-readable storage media can be tangible means for retaining and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer-readable storage media includes: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital universal disk (DVD), memory sticks, floppy disks, mechanical encoding devices such as punch cards or protrusions in slots having instructions recorded thereon, and any suitable combination of the foregoing. As used herein, computer-readable storage media should not be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0105] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a suitable computing / processing device via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network), or to an external computer or external storage device. The network may include copper cables, optical fibers, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to a computer-readable storage medium within the suitable computing / processing device.

[0106] Computer-readable program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and procedural programming languages ​​(such as the "C" programming language or similar programming languages). The computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as a standalone software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer may be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)) or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs) may be personalized to execute computer-readable program instructions by utilizing state information from the computer-readable program instructions in order to perform aspects of this invention.

[0107] The present invention will now be described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0108] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / actions specified in one or more blocks of a flowchart and / or block diagram. These computer-readable program instructions may also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner, such that the computer-readable storage medium storing the instructions includes an article of manufacture containing instructions that implement aspects of the functions / actions specified in one or more blocks of a flowchart and / or block diagram.

[0109] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus, or other device to produce computer-implemented processing, such that the instructions executed on the computer, other programmable apparatus, or other device perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0110] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to different embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of instructions, including one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than indicated in the figures. For example, depending on the functions involved, two consecutively shown blocks may actually be executed substantially simultaneously, or these blocks may sometimes be executed in reverse order. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action or executes a combination of dedicated hardware and computer instructions.

[0111] Various embodiments of the invention have been described for illustrative purposes, but are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope of the described embodiments. The terminology used herein has been chosen to best explain the principles of the embodiments, their practical application, or technical improvements over those found in the market, or to enable those skilled in the art to understand the embodiments described herein.

Claims

1. A computer-implemented method for monitoring a manufacturing system, comprising: Sensor data from a manufacturing system is collected by a computer system. The sensor data is measured at multiple intervals that are less than the time interval T of the target measurement of the manufacturing system, wherein the sensor data is determined to be related to the target measurement. The computer system generates synthetic target measurements at a first interval among a plurality of intervals less than the time interval T based on the relationship, wherein the computer system employs a machine learning model to generate the synthetic target measurements, the machine learning model comprising a neural network trained on training data including historical sensor data and historical target measurement data, wherein training the machine learning model comprises fitting a regression model, the regression model associating sensor data measured at multiple intervals with historical target measurement data at the time interval T through unknown intermediate quality results, wherein the historical target measurement data is based on equal volumes sampled uniformly over time; The computer system automatically generates a pre-warning for the target measurement based on synthetic target measurements within the interval less than the time interval T; and The computer system automatically modifies one or more control components associated with the manufacturing system in response to the advance warning.

2. The computer-implemented method of claim 1, wherein, In response to a prior warning for the target measurement based on the synthetic target measurement, one or more setpoints associated with the manufacturing system are automatically modified.

3. The computer-implemented method of claim 1, wherein, The advance warning is generated in response to the synthetic target measurement being outside the predetermined range.

4. The computer-implemented method of claim 1, wherein, The manufacturing system is modified to ensure that the synthetic target measurement is within a predetermined range.

5. The computer-implemented method of claim 1, wherein, The target measurement is a quality-related variable corresponding to the physical material process outflow of the manufacturing system. Due to the mixing of equal sample volumes in the container, the target measurement has a non-instantaneous, time-accumulated nature, wherein the equal sample volumes are collected at multiple moments throughout the time interval and measured at the end of the time interval to obtain the target measurement.

6. The computer-implemented method according to claim 5, wherein: The target measurement is a composite measurement of the total aggregate volume of the sample; and The synthetic target measurement is a generated value based on the sensor data at a specific time point, and is not a measurement of the aggregated total volume of the sample. The synthetic target measurement represents the state of the individual sampled volume at the specific time point within the time interval.

7. The computer-implemented method according to claim 5, wherein: The target measurement is a composite measurement of the total aggregate volume of the sample; The aggregation function that generates the composite measurement from the equal sampling volumes that constitute the total aggregation volume is unknown, and is automatically learned as a function of the target measurement across the equal sampling volumes. as well as The synthetic target measurement is a generated value based on the sensor data at a given time point, and is not a measurement of the aggregated total volume of the sample. The synthetic target measurement represents the state of an individual sample volume among equal sample volumes at the given time point within the time interval.

8. The computer-implemented method according to claim 5, wherein: The target measurement is a composite measurement of the total aggregate volume of the sample; It is known that the aggregation function that generates composite measurements from equal sampling volumes constituting the total aggregation volume is the average of the target measurements across the equal sampling volumes; and The synthetic target measurement is a generated value based on the sensor data at a given time point, and is not a measurement of the aggregated total volume of the sample. The synthetic target measurement represents the state of an individual sample volume among equal sample volumes at the given time point within the time interval.

9. The computer-implemented method according to claim 5, wherein: The target measurement is a composite measurement of the total aggregate volume of the sample; It is known that the aggregation function that generates composite measurements from equal sampling volumes constituting the total aggregation volume is a user-specified function derived from the target measurement across equal sampling volumes; and The synthetic target measurement is a generated value based on the sensor data at a given time point, and is not a measurement of the aggregated total volume of the sample. The synthetic target measurement represents the state of an individual sample volume among equal sample volumes at the given time point within the time interval.

10. A system for monitoring a manufacturing system, comprising: Memory, containing computer-readable instructions; as well as One or more processors for executing the computer-readable instructions, the computer-readable instructions controlling the one or more processors to perform operations including: Sensor data of a manufacturing system is collected, the sensor data being measured at multiple intervals less than the time interval of a target measurement of the manufacturing system, wherein the sensor data is determined to be related to the target measurement; Based on the relationship, a synthetic target measurement is generated at a first interval among a plurality of intervals less than the time interval T, wherein the computer system employs a machine learning model to generate the synthetic target measurement, the machine learning model comprising a neural network trained on training data including historical sensor data and historical target measurement data, wherein training the machine learning model comprises fitting a regression model, the regression model associating sensor data measured at multiple intervals with historical target measurement data at the time interval T through unknown intermediate quality results, wherein the historical target measurement data is based on equal volumes sampled uniformly over time; Automatically generate advance warnings for the target measurement based on the synthetic target measurement within the interval less than the time interval T; and In response to the aforementioned advance warning, one or more control components associated with the manufacturing system are automatically modified.

11. The system of claim 10, wherein, In response to a prior warning for the target measurement based on the synthetic target measurement, one or more setpoints associated with the manufacturing system are automatically modified.

12. The system according to claim 10, wherein, The advance warning is generated in response to the synthetic target measurement being outside the predetermined range.

13. The system according to claim 10, wherein, The manufacturing system is modified to ensure that the synthetic target measurement is within a predetermined range.

14. The system according to claim 10, wherein, The target measurement is a quality-related variable corresponding to the physical material process outflow of the manufacturing system. Due to the mixing of equal sample volumes in the container, the target measurement has a non-instantaneous, time-accumulated nature, wherein the equal sample volumes are collected at multiple moments throughout the time interval and measured at the end of the time interval to obtain the target measurement.

15. The system according to claim 14, wherein: The target measurement is a composite measurement of the total aggregate volume of the sample; and The synthetic target measurement is a generated value based on the sensor data at a specific time point, and is not a measurement of the aggregated total volume of the sample. The synthetic target measurement represents the state of the individual sampled volume at the specific time point within the time interval.

16. A computer program product comprising program instructions executable by a processor to cause the processor to perform operations including: Sensor data from the manufacturing system is collected, and the sensor data is measured at multiple intervals less than the time interval T of the target measurement of the manufacturing system, wherein... The sensor data is determined to be related to the target measurement; Based on the relationship, a synthetic target measurement is generated at a first interval among a plurality of intervals less than the time interval T, wherein the computer system employs a machine learning model to generate the synthetic target measurement, the machine learning model comprising a neural network trained on training data including historical sensor data and historical target measurement data, wherein training the machine learning model comprises fitting a regression model, the regression model associating sensor data measured at multiple intervals with historical target measurement data at the time interval T through unknown intermediate quality results, wherein the historical target measurement data is based on equal volumes sampled uniformly over time; Pre-warnings for the target measurement are automatically generated based on the synthetic target measurement within the interval less than the time interval T; as well as In response to the advance warning, one or more control components associated with the manufacturing system are automatically modified.

17. The computer program product according to claim 16, wherein, In response to a prior warning for the target measurement based on the synthetic target measurement, one or more setpoints associated with the manufacturing system are automatically modified.

Images