WO2023053511A1 - Système de commande, procédé de traitement d'informations et dispositif de traitement d'informations - Google Patents

Système de commande, procédé de traitement d'informations et dispositif de traitement d'informations Download PDF

Info

Publication number
WO2023053511A1
WO2023053511A1 PCT/JP2022/012284 JP2022012284W WO2023053511A1 WO 2023053511 A1 WO2023053511 A1 WO 2023053511A1 JP 2022012284 W JP2022012284 W JP 2022012284W WO 2023053511 A1 WO2023053511 A1 WO 2023053511A1
Authority
WO
WIPO (PCT)
Prior art keywords
feature amount
controlled object
control system
feature
calculation code
Prior art date
Application number
PCT/JP2022/012284
Other languages
English (en)
Japanese (ja)
Inventor
真輔 川ノ上
健斗 土川
Original Assignee
オムロン株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by オムロン株式会社 filed Critical オムロン株式会社
Publication of WO2023053511A1 publication Critical patent/WO2023053511A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
    • G05B19/05—Programmable logic controllers, e.g. simulating logic interconnections of signals according to ladder diagrams or function charts
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring

Definitions

  • the present invention relates to a control system including a control device, an information processing method, and an information processing device.
  • Predictive maintenance is a type of maintenance that detects an abnormality that occurs in a machine or device and performs maintenance work such as maintenance and replacement before the equipment must be stopped. In order to realize such anomaly detection, it is required to prepare an appropriate model according to the object to be monitored.
  • Patent Document 1 discloses a technique for providing an environment in which anomaly detection can be performed by reliable machine learning on the control device side.
  • the model described above is input with feature values based on measured values obtained from the monitored object.
  • Information to be extracted as a feature amount and information to be excluded from the feature amount may be mixed and measured from the monitored object.
  • one object of the present invention is to provide a mechanism that can flexibly customize the calculation processing of the feature amount used for anomaly detection according to the monitoring target.
  • a control system includes a PLC engine that executes control calculations according to a user program, and an abnormality that occurs in a controlled object based on a feature amount calculated from information collected from the controlled object by referring to a learned model.
  • a control device having an anomaly detection engine that detects the and a program development unit for generating a user program from the project including the feature amount calculation code.
  • the feature amount calculation code may include an instruction for determining at least one of the start position and end position of the frame section, which is the section for which the feature amount is to be calculated, based on the frame variables. According to this configuration, it is possible to customize so that the feature amount used for anomaly detection can be calculated for an appropriate section.
  • the control system calculates an autocorrelation coefficient for each of a plurality of frame variables based on time-series data of information collected from the controlled object, and selects the frame variable with the maximum calculated autocorrelation coefficient. It may further include a selection unit to select. According to this configuration, it is possible to more appropriately determine the frame variable for determining the frame interval.
  • the feature amount calculation code may include instructions for smoothing the feature amount. According to this configuration, it is possible to reduce noise and the like included in the feature amount used for abnormality detection, thereby improving the detection accuracy.
  • the control system may further include a unit determination unit that determines a smoothing unit for smoothing the feature amount based on the crest factor of the measured value used to calculate the target feature amount. According to this configuration, a smoothing unit suitable for smoothing can be determined more easily.
  • the feature quantity calculation code may include instructions for determining the interval in which anomaly detection is enabled. According to this configuration, it is possible to improve detection accuracy by appropriately setting an interval in which abnormality detection is enabled.
  • the control system determines the removal set so that the degree of separation between sets obtained by removing the removal set from each of the set indicating the normal state and the set indicating the abnormal state is maximized, and determines the removal set thus determined.
  • An interval determination unit may further be included that determines an interval to be excluded from abnormality detection based on. According to this configuration, it is possible to more appropriately determine the section in which abnormality detection is enabled.
  • the control system may further include an editing unit that accepts user editing of the feature quantity calculation code. According to this configuration, the user can arbitrarily edit the feature amount calculation code.
  • an information processing method in a control system includes the step of the control device executing control calculations according to the user program, and the control device referring to the learned model and generating in the controlled object based on the feature amount calculated from the information collected from the controlled object.
  • a step of detecting an abnormality a step of generating a trained model from time-series data of information collected from the controlled object, and a step of calculating a feature value input to the trained model from the information collected from the controlled object. It includes steps of generating feature quantity calculation code including instructions, and generating a user program from the project including the feature quantity calculation code.
  • an information processing device connectable to a control device for controlling a controlled object.
  • the control device includes a PLC engine that executes control calculations according to a user program, and an anomaly detection that detects anomalies occurring in the controlled object based on feature values calculated from information collected from the controlled object by referring to the learned model. including the engine.
  • the information processing device includes a model generating unit that generates a trained model from time-series data of information collected from the controlled object, and a model generating unit that calculates feature values to be input to the trained model from the information collected from the controlled object. It includes a code generation unit that generates feature amount calculation code including instructions, and a program development unit that generates a user program from a project including the feature amount calculation code.
  • FIG. 1 is a schematic diagram showing an example of the overall configuration of a control system according to an embodiment
  • FIG. 2 is a block diagram showing an example hardware configuration of a control device of the control system according to the present embodiment
  • FIG. 2 is a block diagram showing an example hardware configuration of a support device of the control system according to the present embodiment
  • FIG. 3 is a schematic diagram showing a configuration example for realizing abnormality detection in the control system according to the present embodiment
  • 1 is a schematic diagram showing an example of a model used in a control system according to an embodiment
  • FIG. FIG. 2 is a schematic diagram showing an example of a functional configuration for realizing abnormality detection in the control system according to the present embodiment
  • FIG. 7 is a diagram for explaining a setting example of a frame interval among customization items of the control system according to the present embodiment
  • FIG. 4 is a diagram for explaining smoothing processing among customization items of the control system according to the present embodiment
  • FIG. 5 is a diagram for explaining monitoring condition setting among customization items of the control system according to the present embodiment
  • 4 is a flow chart showing an example of an overall processing procedure for generating a model and feature amount calculation code in the control system according to the present embodiment
  • FIG. 11 is a flowchart showing a more detailed processing procedure of frame variable generation processing (step S2) in FIG. 10
  • FIG. FIG. 11 is a diagram for explaining processing in the frame variable generation processing (step S2) of FIG. 10;
  • FIG. 11 is a diagram showing an example of a user interface screen provided in the frame variable generation process (step S2) of FIG. 10;
  • FIG. 11 is a flowchart showing a more detailed processing procedure of the feature quantity calculation process (step S3) in FIG. 10;
  • FIG. FIG. 11 is a diagram showing an example of a user interface screen provided in the feature quantity calculation process (step S3) of FIG. 10;
  • FIG. 11 is a flow chart showing a more detailed processing procedure of a monitoring condition setting process (step S6) in FIG. 10;
  • FIG. FIG. 11 is a diagram showing an example of a user interface screen provided in the monitoring condition setting process (step S6) of FIG. 10;
  • FIG. 4 is a schematic diagram showing an example of a code template for generating feature amount calculation code in the control system according to the present embodiment;
  • controlled object any object that is directly or indirectly related to the control operation executed by the control device is referred to as a "controlled object".
  • controlled object is not limited to the machines and devices to which the control device gives commands, and the devices from which the control device collects information, but also any equipment and units that include those machines and devices. can include
  • monitoring target is a generic term for targets for detecting whether any abnormality has occurred.
  • the “monitored object” may include a specific machine or device, or may include a production process or the like.
  • FIG. 1 is a schematic diagram showing an example of the overall configuration of a control system 1 according to this embodiment.
  • a control system 1 according to the present embodiment includes, as main components, a control device 100 for controlling a controlled object, and a support device 200 which is an example of a computer connectable to the control device 100. including.
  • the control device 100 may be embodied as a kind of computer such as a PLC (Programmable Logic Controller).
  • the control device 100 is connected to the field device group 10 via the fieldbus 2 .
  • Fieldbus 2 preferably employs an industrial communication protocol.
  • EtherCAT registered trademark
  • EtherNet/IP registered trademark
  • DeviceNet registered trademark
  • CompoNet registered trademark
  • the field device group 10 includes devices that collect input data from control targets or control-related manufacturing devices, production lines, and the like (hereinafter also collectively referred to as "fields"). An input relay, various sensors, and the like are assumed as a device for collecting such input data. Field device group 10 further includes devices that give some effect to the field based on commands (hereinafter also referred to as “output data”) generated by control device 100 . Output relays, contactors, servo drivers and servo motors, and any other actuators are contemplated as devices that exert some action on such fields. These field device groups 10 exchange data including input data and output data with the control device 100 via the fieldbus 2 .
  • the field device group 10 includes a remote I/O (Input/Output) device 12, a relay group 14, a servo driver 18 and a servo motor 20.
  • a remote I/O (Input/Output) device 12 includes a remote I/O (Input/Output) device 12, a relay group 14, a servo driver 18 and a servo motor 20.
  • the remote I/O device 12 includes a communication unit that communicates via the fieldbus 2 and an input/output unit (hereinafter also referred to as "I/O unit") for collecting input data and outputting output data. including. Input data and output data are exchanged between the control device 100 and the field via such an I/O unit.
  • FIG. 1 shows an example in which digital signals are exchanged as input data and output data via the relay group 14 .
  • the I/O unit may be directly connected to the fieldbus.
  • FIG. 1 shows an example in which an I/O unit 16 is directly connected to the fieldbus 2 .
  • the servo driver 18 drives the servo motor 20 in accordance with output data (eg, position command, etc.) from the control device 100 .
  • I/O refresh processing As described above, input data and output data are exchanged between the control device 100 and the field device group 10 via the fieldbus 2. These exchanged data are on the order of hundreds of microseconds. It will be updated in a very short cycle on the order of several tens of milliseconds. Note that such processing for updating exchanged data is sometimes referred to as "I/O refresh processing".
  • the control device 100 has a PLC engine that executes control calculations for controlling controlled objects such as equipment and machines.
  • the PLC engine corresponds to a control calculation unit, and determines output data by executing control calculation based on input data.
  • the control device 100 has a time series database (hereinafter referred to as “TSDB ( (also referred to as Time Series Data Base)") 140.
  • TSDB time series database
  • the data stored in the TSDB 140 will also be referred to as "measurement value time series data”.
  • the control device 100 is configured to apply information collected from the controlled object to a pre-prepared learned model (hereinafter also simply referred to as "model"). More specifically, the control device 100 has an anomaly detection engine 150 that refers to the model and detects an anomaly that occurs in the controlled object. A feature value calculated from information collected from a controlled object is input to the model, and a score is output. The score output by the anomaly detection engine 150 is an index indicating the possibility that some kind of anomaly has occurred in the controlled object.
  • the anomaly detection engine 150 may be provided with a decision threshold as a parameter for determining the score output from the model.
  • Information collected from controlled objects includes arbitrary information collected from the field and arbitrary information managed inside the control device 100 .
  • the control device 100 may be connected to the server 400 via the host network 6 or may be connected to one or more HMI (Human Machine Interface) 500 via the fieldbus 4 .
  • HMI Human Machine Interface
  • the server 400 is in charge of processing such as providing arbitrary information to the control device 100 or collecting data from the control device 100 .
  • the HMI 500 receives an operation from the user, transmits a command according to the user's operation to the control device 100, and graphically displays the calculation result of the control device 100.
  • the support device 200 which is an example of an information processing device, includes a development environment (program creation and editing tool, parser, compiler, etc.) for a user program executed by the control device 100, the control device 100, and various devices connected to the control device 100. It provides a function of setting parameters (configuration), a function of transmitting a generated user program to the control device 100, a function of online modifying and changing a user program executed on the control device 100, and the like.
  • the support device 200 also has a function for generating a model to be referenced by the anomaly detection engine 150 implemented in the control device 100 .
  • the support device 200 acquires measurement value time-series data from the control device 100 ((1) measurement value time-series data). Subsequently, the support device 200 generates a model based on the measured value time-series data acquired from the control device 100 ((2) model generation). The support device 200 generates a feature quantity calculation code ((3) feature quantity calculation code) in conjunction with the generation of the model.
  • the support device 200 generates a user program from the project containing the generated feature amount calculation code ((4) User program generation). Finally, the user program and model are transferred from the support device 200 to the control device 100 ((5) User program and model).
  • the feature quantity calculation code includes instructions for calculating the feature quantity to be input to the model from the information collected from the controlled object.
  • FIG. 2 is a block diagram showing a hardware configuration example of the control device 100 of the control system 1 according to this embodiment.
  • control device 100 includes a processor 102 such as a CPU (Central Processing Unit) or MPU (Micro-Processing Unit), a chipset 104, a main storage device 106, a secondary storage device 108, Host network controller 110, USB (Universal Serial Bus) controller 112, memory card interface 114, internal bus controller 122, field bus controllers 118, 120, I/O units 124-1, 124-2, . including.
  • the processor 102 implements the PLC engine and the anomaly detection engine 150 by reading out various programs stored in the secondary storage device 108, deploying them in the main storage device 106, and executing them.
  • the chipset 104 controls data transmission between the processor 102 and each component.
  • the secondary storage device 108 stores a system program 131 for realizing the PLC engine and anomaly detection engine 150, as well as a user program 132 that is executed using the PLC engine.
  • a partial area of the secondary storage device 108 may be used as the TSDB 140 .
  • the host network controller 110 controls data exchange with other devices via the host network 6 .
  • the USB controller 112 controls data exchange with the support device 200 via a USB connection.
  • the memory card interface 114 is configured such that a memory card 116 can be attached/detached, and data can be written to the memory card 116, and various data (user program, trace data, etc.) can be read from the memory card 116. ing.
  • the internal bus controller 122 is an interface that exchanges data with the I/O units 124-1, 124-2, .
  • the fieldbus controller 118 controls data exchange with other devices via the fieldbus 2 .
  • the fieldbus controller 120 controls data exchange with other devices via the fieldbus 4 .
  • FIG. 2 shows a configuration example in which necessary functions are provided by the processor 102 executing a program.
  • the main part of the control device 100 may be implemented using hardware conforming to a general-purpose architecture (for example, an industrial personal computer based on a general-purpose personal computer).
  • virtualization technology may be used to run multiple OSs (Operating Systems) for different purposes in parallel, and to run necessary applications on each OS.
  • Support device 200 is implemented, for example, by executing a program using hardware conforming to a general-purpose architecture (for example, a general-purpose personal computer).
  • a general-purpose architecture for example, a general-purpose personal computer.
  • FIG. 3 is a block diagram showing a hardware configuration example of the support device 200 of the control system 1 according to this embodiment.
  • support device 200 includes a processor 202 such as a CPU or MPU, an optical drive 204, a main storage device 206, a secondary storage device 208, a USB controller 212, a network controller 214, an input A portion 216 and a display portion 218 are included. These components are connected via bus 220 .
  • the processor 202 reads various programs stored in the secondary storage device 208, develops them in the main storage device 206, and executes them, thereby realizing various processes to be described later.
  • the secondary storage device 208 is composed of, for example, an HDD (Hard Disk Drive) or an SSD (Flash Solid State Drive).
  • the secondary storage device 208 typically contains an OS 222 and a PLC interface program 224 for exchanging data relating to anomaly detection functions between the control device 100 and a user program to be executed in the support device 200.
  • a development program 226 for debugging created user programs, defining system configurations, setting various parameters, etc.
  • a model generation program 228 for generating models.
  • the secondary storage device 208 may store necessary programs other than the programs shown in FIG.
  • the support device 200 has an optical drive 204, and from a recording medium 205 (for example, an optical recording medium such as a DVD (Digital Versatile Disc)) that stores a computer-readable program non-transitory,
  • a recording medium 205 for example, an optical recording medium such as a DVD (Digital Versatile Disc)
  • the stored program is read and installed in the secondary storage device 208 or the like.
  • Various programs executed by the support device 200 may be installed via the computer-readable recording medium 205, or may be installed by downloading them from a server device on the network. Also, the functions provided by the support apparatus 200 according to the present embodiment may be realized by using some of the modules provided by the OS 222. FIG.
  • the USB controller 212 controls data exchange with the control device 100 via a USB connection.
  • Network controller 214 controls the exchange of data to and from other devices over any network.
  • the input unit 216 is composed of a keyboard, mouse, etc., and receives user operations.
  • a display unit 218 includes a display, various indicators, a printer, and the like, and outputs processing results from the processor 202 and the like.
  • FIG. 3 shows a configuration example in which necessary functions are provided by the processor 202 executing a program. Alternatively, it may be implemented using an FPGA, etc.).
  • FIG. 4 is a schematic diagram showing a configuration example for realizing abnormality detection in the control system 1 according to the present embodiment.
  • FIG. 4A shows a configuration example using a feature amount calculation unit
  • FIG. 4B shows a configuration example using an AI library.
  • control device 100 has a PLC engine 130, a TSDB 140, an anomaly detection engine 150, and a feature amount calculator 156 as main functional configurations.
  • the PLC engine 130 executes control calculations according to a user program 132 that is arbitrarily created according to the object to be controlled.
  • the PLC engine 130 also has a data management unit 136 for holding input data, output data, and internal data in a form that can be referenced from the user program 132 .
  • the data values managed by the data management unit 136 are updated every predetermined cycle (I/O refresh cycle) by the I/O refresh process.
  • the data management unit 136 stores predetermined data among the data to be managed in the TSDB 140 at predetermined intervals. As a result, the TSDB 140 can output the value change of the designated data every predetermined period, that is, the measured value time-series data 142 .
  • the feature amount calculation unit 156 refers to data managed by the data management unit 136 and calculates a plurality of feature amounts 152 specified in advance at predetermined intervals.
  • the feature amount calculated by the feature amount calculation unit 156 is a value calculated according to an arbitrary function from the data managed by the data management unit 136. For example, the maximum value, minimum value, median value, average value, standard deviation, strain degree, kurtosis, etc. can be employed. Note that the data managed by the data management unit 136 (measurement value time-series data 142) can be used as the feature amount 152 as it is.
  • the anomaly detection engine 150 refers to the model 160 and detects an anomaly occurring in the controlled object based on the feature amount calculated from the information collected from the controlled object. More specifically, the anomaly detection engine 150 determines the occurrence of an anomaly or the possibility of an anomaly based on a plurality of feature quantities 152 from the feature quantity calculator 156 .
  • FIG. 5 is a schematic diagram showing an example of the model 160 used in the control system 1 according to this embodiment.
  • a superspace having a plurality of feature quantities 152 as dimensions is defined, and learning data 164 representing the normal state of the monitored object is arranged.
  • the distance from the learning data 164 of the multiple input feature values 152 corresponds to the score. That is, the score is a value indicating the degree of detachment from the normal state. Based on whether the calculated score is within a predetermined determination threshold value 162, it is determined whether or not the state is normal.
  • the anomaly detection engine 150 refers to the model 160 to calculate scores for the plurality of feature quantities 152, and determines whether the calculated scores are within the determination threshold value 162. to judge whether The anomaly detection engine 150 outputs a determination result 154, which is a comparison result between the calculated score and the determination threshold value 162, to the PLC engine 130 (user program 132).
  • the user program 132 executed by the PLC engine 130 executes control calculations according to the determination result 154 .
  • target values and functions to be used for calculating the plurality of feature quantities 152 are determined.
  • feature quantities 152 are selected to maximize the accuracy of anomaly detection.
  • Parameters and settings for calculating feature quantity 152 are provided from support device 200 to control device 100 .
  • FIG. 4(B) shows an example in which the AI library 134 is used in place of the feature quantity calculation unit 156 .
  • the AI library 134 is a library in which feature quantities and the like unique to specific monitoring targets are defined in advance.
  • the AI library 134 is available for the user program 132 .
  • the AI library 134 may be incorporated as part of the user program 132 .
  • the AI library 134 refers to data managed by the data management unit 136 and calculates a plurality of predefined feature amounts 152 at predetermined intervals. A plurality of feature quantities 152 calculated by the AI library 134 are provided to the anomaly detection engine 150 .
  • the anomaly detection engine 150 refers to the model 160 to calculate scores for the plurality of feature quantities 152 and determines whether the calculated scores are within the determination threshold value 162 .
  • the anomaly detection engine 150 outputs a determination result 154, which is a comparison result between the calculated score and the determination threshold value 162, to the PLC engine 130 (user program 132).
  • Such an AI library 134 is created by analyzing the behavior of monitored objects. As a more specific procedure, first, a large number of measurable measured value time-series data are collected from the monitored object. In order to improve the generalization performance, it is necessary to measure under various conditions. Subsequently, the collected measured value time-series data is analyzed to search for appropriate feature amounts.
  • the control device 100 when the support device 200 generates the model 160, the control device 100 also outputs a code for calculating the feature amount.
  • FIG. 6 is a schematic diagram showing an example of a functional configuration for realizing abnormality detection in the control system 1 according to this embodiment.
  • support device 200 includes a development tool 240 in charge of development of user program 132, and a data analysis/code generation tool 250 in charge of generation of model 160 and feature quantity calculation code 264. .
  • the development tool 240 is implemented by the processor 202 of the support device 200 executing the development program 226.
  • the data analysis/code generation tool 250 is implemented by the processor 202 of the support device 200 executing the model generation program 228 .
  • development tool 240 and the data analysis/code generation tool 250 are mounted on the same support device 200 is shown, but the development tool 240 and the data analysis/code generation tool 250 are different. Each may be mounted in an information processing device.
  • the development tool 240 provides a development environment for the user program 132. More specifically, the development tool 240 creates the project 242 and builds the project 242 according to the user's operation, thereby generating the user program 132 in the object format. The generated user program 132 is transferred from the support device 200 to the control device 100 .
  • the feature quantity calculation code 244 generated by the data analysis/code generation tool 250 it is possible to incorporate the feature quantity calculation code 244 generated by the data analysis/code generation tool 250 into the project 242 .
  • the development tool 240 generates the user program 132 from the project 242 including the feature quantity calculation code 244 . More specifically, the development tool 240 generates the user program 132 including the feature quantity calculation module 138 by building the project 242 incorporating the feature quantity calculation code 244 .
  • the feature quantity calculation module 138 is an object type execution module generated by building the feature quantity calculation code 244 .
  • the user can also change the generated feature amount calculation code 244 as appropriate.
  • the development tool 240 has an editing function for receiving user's editing of the feature amount calculation code 244 . Therefore, the user program 132 can be generated after appropriately customizing the feature amount calculation code 244 generated by the data analysis/code generation tool 250 according to the monitoring target.
  • the data analysis/code generation tool 250 includes an analysis unit 252 , a user interface 254 , a model generation unit 256 and a code generation unit 258 .
  • the analysis unit 252 typically performs feature amount calculation processing, feature amount selection acceptance processing, abnormality detection performance evaluation processing, parameter determination processing, and the like.
  • the user interface 254 presents the obtained measurement value time-series data 142 and the analysis results of the analysis unit 252 to the user, and accepts user operations.
  • the model generation unit 256 generates a model from the time-series data (measurement value time-series data 142) of information collected from the controlled object. More specifically, the model generation unit 256 generates the model 160 including the learning data 164 based on the analysis results of the analysis unit 252 and the like.
  • the model 160 typically includes a function defining a hyperspace whose dimensions are each of a plurality of selected feature quantities, and learning data 164 for calculating a score.
  • Model generator 256 also generates decision threshold 162 corresponding to model 160 .
  • the code generation unit 258 generates feature amount calculation code 264 including instructions for calculating feature amounts to be input to the model 160 from information collected from the controlled object.
  • the feature amount calculation code 264 includes commands such as the frame section for which the feature amount is to be calculated, monitoring conditions, smoothing processing, etc., in addition to specification of target values for calculating the feature amount and functions to be used. Details of the process of generating the feature amount calculation code 264 will be described later.
  • the feature amount calculation code 264 is not limited to these items, and any item that requires customization of the feature amount calculation process used for abnormality detection according to the monitored object can be described in the feature amount calculation code 264 .
  • the feature amount is calculated for each predetermined fixed or variable frame section.
  • Such frame intervals may be determined based on control signals such as work detect and cycle start.
  • FIG. 7 is a diagram for explaining a setting example of the frame section among the customization items of the control system 1 according to the present embodiment.
  • waveforms having similar characteristics appear repeatedly at predetermined intervals. In such a case, it is preferable to determine a cycle in which waveforms having similar characteristics appear as one frame interval.
  • the customization item includes processing for determining the frame interval.
  • the frame interval is started when an arbitrary start condition is satisfied, and ended when an arbitrary end condition is satisfied or when the start condition is satisfied again.
  • an air-driven actuator is connected to a branch pipe branched from a supply air pipe and operated asynchronously with other actuators.
  • the pressure supplied to the monitored actuator may be influenced by other actuators.
  • FIG. 8 is a diagram for explaining the smoothing process among the customization items of the control system 1 according to the present embodiment.
  • FIG. 8(A) shows an example of time-series data of the pressure in the supply source air pipe
  • FIG. 8(B) shows an example of time-series data of the time required for the actuator to operate (operation time). show.
  • FIG. 9 is a diagram for explaining monitoring condition settings among customization items of the control system 1 according to the present embodiment.
  • FIG. 9 shows an example of time-series data of scores calculated for each frame.
  • a score is calculated from a plurality of feature values calculated for each frame, but in an unstable state such as immediately after the start of motion, a score exceeding the determination threshold may be calculated. It is preferable to set such an outlier in score to be excluded from monitoring.
  • arbitrary monitoring conditions can be set.
  • FIG. 10 is a flow chart showing an example of an overall processing procedure for generating the model 160 and the feature amount calculation code 244 in the control system 1 according to this embodiment. Each step shown in FIG. 10 is implemented by executing the model generation program 228 by the processor 202 of the support device 200 .
  • support device 200 reads measurement value time-series data 142 acquired from control device 100 (step S1). The support device 200 then executes frame variable generation processing (step S2). In the frame variable generation process, a frame variable (control signal) used for determining the frame interval is determined.
  • the support device 200 calculates multiple types of feature amounts from the measured value time-series data 142 for each frame interval determined by the determined frame variables (step S3).
  • the support device 200 accepts selection of a feature amount by the user from among the calculated plural types of feature amounts (step S4). Further, the support device 200 receives input of a label for the selected feature amount (step S5).
  • the support device 200 evaluates the degree of separation of the score calculated by the selected feature amount, and accepts setting of monitoring conditions (step S6).
  • the support device 200 generates the model 160 (including the learning data 164), the determination threshold value 162, and the feature amount calculation code 244 in response to user's operation (step S7). That is, the support device 200 generates the feature quantity calculation code 244 including instructions for calculating the feature quantity to be input to the model 160 from the information collected from the controlled object.
  • the user incorporates the generated feature quantity calculation code 244 into the project 242, builds the project 242, and generates the user program 132. That is, the support device 200 generates the user program 132 from the project 242 including the feature quantity calculation code 244 . The generated user program 132 is transferred to the control device 100 .
  • FIG. 11 is a flow chart showing a more detailed processing procedure of the frame variable generation processing (step S2) of FIG.
  • the support device 200 determines whether or not the frame variables have been set (step S21). If the frame variable has already been set (YES in step S21), the following processing is skipped.
  • the support device 200 selects one frame variable from among the frame variable candidates (step S22), and calculates the autocorrelation coefficient (step S23). The support device 200 determines whether autocorrelation coefficients have been calculated for all of the frame variable candidates (step S24). If frame variable candidates for which autocorrelation coefficients have not been calculated remain among the frame variable candidates (NO in step S24), the processing from step S22 onward is repeated.
  • step S24 If autocorrelation coefficients have been calculated for all frame variable candidates (YES in step S24), support device 200 selects the frame variable for which the maximum autocorrelation coefficient among the calculated autocorrelation coefficients is calculated. is selected (step S25), and the frame period is determined based on the autocorrelation coefficient of the selected frame variable (step S26). It should be noted that the determined frame period can be changed according to user operation. Then, the process proceeds to step S3 in FIG.
  • FIG. 12 is a diagram for explaining the processing in the frame variable generation processing (step S2) of FIG.
  • FIG. 12A shows an example of measured value time-series data 142 of the selected frame variable.
  • FIG. 12B shows an example of waveform data of autocorrelation coefficients of selected frame variables.
  • Waveform data as shown in FIG. 12(B) is calculated as the autocorrelation coefficient of the measured value time-series data 142 as shown in FIG. 12(A).
  • the time between peaks of the waveform data (correlogram) of the autocorrelation coefficient corresponds to the frame period.
  • the support device 200 calculates the autocorrelation coefficient for each of the plurality of frame variables based on the time-series data of the information collected from the controlled object, and the calculated autocorrelation coefficient is the maximum. It has a selection function that selects a frame variable that is
  • FIG. 13 is a diagram showing an example of a user interface screen provided in the frame variable generation process (step S2) of FIG.
  • user interface screen 280A displayed on display unit 218 of support device 200 includes a variable list area 282 that can be selected as a frame variable and a variable selected in variable list area 282 for the measured value. It includes a time series data display area 284 that displays the series data 142 and a histogramming display area 286 that displays histogramming of the measured value time series data displayed in the time series data display area 284 .
  • step S2 The frame variables determined in the frame variable generation process (step S2) are highlighted in the variable list area 282 (reference numeral 288). Also, the determined frame period is displayed in dialog 290 . The user can change the determined frame period by operating the dialog 290 .
  • FIG. 14 is a flowchart showing a more detailed processing procedure of the feature amount calculation process (step S3) in FIG.
  • support device 200 selects one measurement value from the measurement values (variables) for which feature amounts are calculated (step S31), and calculates a plurality of types of feature amounts for each of the selected measurement values. (Step S32).
  • the plurality of types of feature values include, for example, maximum value, minimum value, median value, average value, standard deviation, skewness, and kurtosis.
  • the support device 200 determines whether or not the crest factor of the selected measured value exceeds a predetermined reference value (step S33). The crest factor is calculated as (peak value of selected measurement)/(rms value of selected measurement).
  • step S34 determines a smoothing unit corresponding to the crest factor of the selected measurement value (step S34 ), the selected measurement value is smoothed by the determined smoothing unit, and then a plurality of types of feature amounts are calculated (step S35).
  • the smoothing unit means, for example, the number of input values to be subjected to moving average when the moving average method or the like is adopted.
  • the correspondence between the crest factor and the smoothing unit may be defined in advance or determined according to a function of the crest factor.
  • the support device 200 has a unit determination function that determines a smoothing unit for smoothing the feature amount based on the crest factor of the measured value used to calculate the target feature amount.
  • step S35 After step S35 is executed, or if the crest factor of the selected measured value is equal to or less than the predetermined reference value (NO in step S33), the support device 200 calculates the feature quantity is calculated (step S36). If the measured values for which the feature amount is not calculated remain among the measured values for which the feature amount is calculated (NO in step S36), the processing from step S31 onward is repeated.
  • step S36 If the feature amount has been calculated for all the measurement values for which the feature amount is to be calculated (YES in step S36), the process proceeds to step S4 in FIG.
  • FIG. 15 is a diagram showing an example of a user interface screen provided in the feature quantity calculation process (step S3) of FIG.
  • user interface screen 280B displayed on display unit 218 of support device 200 includes a variable list area 282 that can be selected as a measurement value, and a variable list area 282 for the variable selected in variable list area 282. It includes a time series data display area 284 that displays the series data 142 and a histogramming display area 286 that displays histogramming of the measured value time series data displayed in the time series data display area 284 .
  • the arbitrarily selected measurement value is highlighted in the variable list area 282 (reference numeral 292). Also, the determined smoothing unit is displayed in dialog 294 . The user can change the determined smoothing unit by operating the dialog 294 .
  • FIG. 16 is a flowchart showing a more detailed processing procedure of the monitoring condition setting process (step S6) of FIG.
  • the support device 200 calculates the degree of separation for the selected feature amount (step S61). More specifically, the support device 200 determines the removal set so that the degree of separation is maximized from the histogram corresponding to the normal state and the histogram corresponding to the abnormal state for the selected feature amount. do.
  • the elements included in the normal state histogram be the normal set N ⁇
  • the elements included in the abnormal state histogram be the abnormal set A ⁇ .
  • the removal set R ⁇ ⁇ is determined so that the degree of separation between (between) is maximized.
  • the support device 200 removes the removed sets from the normal set and the abnormal set and determines them as learning data (step S62), and sets the sections corresponding to the removed sets to be removed from the monitored sections (step S63). At this time, the support device 200 may determine conditions for defining the set monitoring exclusion section. Then, the process proceeds to step S7 in FIG.
  • the support device 200 removes the removed set R ⁇ from each of the set indicating the normal state (normal set N ⁇ ) and the set indicating the abnormal state (abnormal set A ⁇ ).
  • a removal set R ⁇ ⁇ is determined so that the degree of separation is maximized, and a section excluded from abnormality detection (monitoring exclusion section) is determined based on the determined removal set R ⁇ ⁇ .
  • FIG. 17 is a diagram showing an example of a user interface screen provided in the monitoring condition setting process (step S6) of FIG.
  • user interface screen 280C displayed on display unit 218 of support device 200 includes a variable list area 282 selectable as a feature amount and a feature amount for the feature amount selected in variable list area 282. It includes a time-series data display area 284 that displays time-series data and a histogramming display area 286 that displays histogramming of the feature amount time-series data displayed in the time-series data display area 284 .
  • the arbitrarily selected feature quantity is highlighted in the variable list area 282 (reference numeral 296).
  • the determined monitoring exclusion section is displayed in a dialog 298 . The user can change the determined monitoring exclusion section by operating the dialog 298 .
  • FIG. 18 is a schematic diagram showing an example of the code template 230 for generating the feature amount calculation code 244 in the control system 1 according to this embodiment.
  • feature amount calculation code 264 is generated by reflecting analysis results and user settings in code template 230 .
  • the code template 230 is configured based on code blocks and includes command descriptions corresponding to predetermined customizable content.
  • the feature amount calculation code 264 is generated by combining the instruction description with the specific parameters based on the analysis results and user settings.
  • the code template 230 includes a code block 231 corresponding to initialization processing, a code block 232 corresponding to monitoring conditions, a code block 233 corresponding to characterization, and a code block 234 corresponding to framing. and a code block 235 corresponding to smoothing.
  • the code block 231 corresponding to initialization processing includes processing such as initialization of variables used in the feature amount calculation code 244, for example.
  • the code block 232 corresponding to the monitoring condition is, for example, when the variable (eg, V a1 ) used for the monitoring condition exceeds a predetermined threshold value (eg, T a1 ), the feature amount calculation process is executed. It includes processing that defines conditions such as starting. At this time, variables used for monitoring conditions and corresponding threshold values are specified as parameters. Thus, the feature quantity calculation code 244 includes instructions for determining intervals for enabling detection of anomalies.
  • the code block 233 corresponding to characterization includes processing for calculating a feature amount selected from a plurality of predetermined types of feature amounts for the selected measurement value (variable). At this time, the variables used for calculating the feature quantity are specified as parameters.
  • the code block 234 corresponding to framing includes processing for determining the start position or end position of the frame section based on the values of the frame variables. At this time, a variable that determines the start position or end position of the frame section is specified as a parameter.
  • the feature amount calculation code 244 includes instructions for determining at least one of the start position and the end position of the frame section, which is the section for which the feature amount is to be calculated, based on the frame variables.
  • a code block 235 corresponding to smoothing includes processing for smoothing the calculated feature amount.
  • a smoothing unit or the like is specified as a parameter.
  • feature calculation code 244 includes instructions for smoothing features.
  • the feature quantity calculation code 244 is incorporated into the project 242 and built by the development tool 240 to generate the user program 132 in object format.
  • a control system (1) A PLC engine (130) that executes control calculations according to a user program (132), and a learned model (160), which is calculated based on a feature amount (156) calculated from information collected from the controlled object.
  • a control device having an anomaly detection engine (150) for detecting an anomaly occurring in a target; a model generation unit (256) that generates the learned model from time-series data (142) of information collected from the controlled object; a code generation unit (258) that generates a feature amount calculation code (264) including instructions for calculating the feature amount to be input to the trained model from information collected from the controlled object;
  • a control system comprising a program development section (240) that generates the user program from a project (242) that includes the feature amount calculation code.
  • [Configuration 2] The configuration according to configuration 1, wherein the feature quantity calculation code includes an instruction (234) for determining at least one of a start position and an end position of a frame section, which is a section for which the feature quantity is calculated, based on a frame variable. control system.
  • the feature quantity calculation code includes an instruction (234) for determining at least one of a start position and an end position of a frame section, which is a section for which the feature quantity is calculated, based on a frame variable. control system.
  • Control system further comprising a unit determination unit that determines a smoothing unit for smoothing the feature amount based on the crest factor of the measured value used to calculate the target feature amount.
  • the removal set is determined so that the degree of separation between the sets obtained by removing the removal set from the set indicating the normal state and the set indicating the abnormal state is maximized, and the abnormality is detected based on the determined removal set.
  • FIG. 9 An information processing method in a control system (1), the step of the controller (100) performing control operations according to the user program (132); a step in which the control device refers to a learned model (160) and detects an abnormality occurring in the controlled object based on a feature amount calculated from information collected from the controlled object; a step of generating the learned model from time-series data of information collected from the controlled object (S7); a step (S7) of generating a feature quantity calculation code including instructions for calculating the feature quantity to be input to the trained model from information collected from the controlled object; and generating the user program from a project including the feature amount calculation code.
  • An information processing device (200) connectable to a control device (100) for controlling a controlled object, the control device comprising a PLC engine (130) executing control calculations according to a user program (132), a learning an anomaly detection engine (150) for detecting an anomaly occurring in the controlled object based on a feature quantity (156) calculated from information collected from the controlled object by referring to the finished model (160);
  • the information processing device a model generation unit (256) that generates the learned model from time-series data (142) of information collected from the controlled object; a code generation unit (258) that generates a feature amount calculation code (264) including instructions for calculating the feature amount to be input to the trained model from information collected from the controlled object; and a program development unit (240) that generates the user program from a project (242) that includes the feature amount calculation code.
  • the feature amount calculation code 264 including the command for calculating the feature amount to be input to the model 160 is generated together with the model 160, the feature amount calculation code 264 is generated as follows. By editing it arbitrarily, it is possible to flexibly customize the calculation processing of the feature value used for anomaly detection according to the monitoring target. Such customization can improve the accuracy of anomaly detection.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • Programmable Controllers (AREA)

Abstract

La présente invention concerne un mécanisme avec lequel il est possible de personnaliser de manière flexible un processus de calcul pour une quantité de caractéristiques pour une utilisation dans une détection d'anomalie en fonction d'une cible de surveillance. Ce système de commande comprend : un dispositif de commande comportant un moteur PLC pour exécuter un calcul de commande conformément à un programme utilisateur et un moteur de détection d'anomalie pour se référer à un modèle entraîné et détecter, sur la base d'une quantité de caractéristiques calculée à partir d'informations collectées à partir d'une cible de commande, une anomalie générée dans la cible de commande ; une unité de génération de modèle pour générer le modèle entraîné à partir de données chronologiques d'informations collectées à partir de la cible de commande ; une unité de génération de code pour générer, à partir des informations collectées depuis la cible de commande, un code de calcul de quantité de caractéristiques comprenant une instruction pour calculer une quantité de caractéristiques à entrer dans le modèle entraîné ; et une unité de développement de programme pour générer le programme utilisateur à partir d'un projet qui comprend le code de calcul de quantité de caractéristiques.
PCT/JP2022/012284 2021-09-30 2022-03-17 Système de commande, procédé de traitement d'informations et dispositif de traitement d'informations WO2023053511A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2021162036A JP2023051402A (ja) 2021-09-30 2021-09-30 制御システム、情報処理方法および情報処理装置
JP2021-162036 2021-09-30

Publications (1)

Publication Number Publication Date
WO2023053511A1 true WO2023053511A1 (fr) 2023-04-06

Family

ID=85782144

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/012284 WO2023053511A1 (fr) 2021-09-30 2022-03-17 Système de commande, procédé de traitement d'informations et dispositif de traitement d'informations

Country Status (3)

Country Link
JP (1) JP2023051402A (fr)
TW (1) TWI843084B (fr)
WO (1) WO2023053511A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018097662A (ja) * 2016-12-14 2018-06-21 オムロン株式会社 制御装置、制御プログラムおよび制御方法
JP2019159697A (ja) * 2018-03-12 2019-09-19 オムロン株式会社 制御システムおよび制御方法
JP2020047016A (ja) * 2018-09-20 2020-03-26 オムロン株式会社 制御装置および制御システム
JP2020047041A (ja) * 2018-09-20 2020-03-26 オムロン株式会社 制御装置および制御システム
JP2020067786A (ja) * 2018-10-24 2020-04-30 オムロン株式会社 制御装置および制御プログラム

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018097662A (ja) * 2016-12-14 2018-06-21 オムロン株式会社 制御装置、制御プログラムおよび制御方法
JP2019159697A (ja) * 2018-03-12 2019-09-19 オムロン株式会社 制御システムおよび制御方法
JP2020047016A (ja) * 2018-09-20 2020-03-26 オムロン株式会社 制御装置および制御システム
JP2020047041A (ja) * 2018-09-20 2020-03-26 オムロン株式会社 制御装置および制御システム
JP2020067786A (ja) * 2018-10-24 2020-04-30 オムロン株式会社 制御装置および制御プログラム

Also Published As

Publication number Publication date
TW202316220A (zh) 2023-04-16
TWI843084B (zh) 2024-05-21
JP2023051402A (ja) 2023-04-11

Similar Documents

Publication Publication Date Title
US10591886B2 (en) Control system, control program, and control method for device switching responsive to abnormality detection
JP6965798B2 (ja) 制御システムおよび制御方法
CN108628263B (zh) 控制系统、控制装置、计算机可读存储介质以及控制方法
JP6724847B2 (ja) 制御装置、制御プログラム、制御システム、および制御方法
CN112673327B (zh) 控制装置及计算机可读取的存储介质
WO2022030041A1 (fr) Système de prédiction, et dispositif et programme de traitement d'informations
CN112673326B (zh) 控制装置及存储介质
CN112639844B (zh) 控制系统以及控制方法
WO2021111936A1 (fr) Système de prédiction, dispositif de traitement d'informations et programme de traitement d'informations
WO2023053511A1 (fr) Système de commande, procédé de traitement d'informations et dispositif de traitement d'informations
JP2023006304A (ja) 制御システム、モデル生成方法およびモデル生成プログラム
JP7571613B2 (ja) 情報処理装置、情報処理プログラムおよび情報処理方法
WO2022181007A1 (fr) Dispositif de traitement d'informations, programme de traitement d'informations et procédé de traitement d'informations
WO2020240945A1 (fr) Système de commande, dispositif de commande et programme de commande
JP7556306B2 (ja) 予測システム、情報処理装置および情報処理プログラム
JP7524784B2 (ja) 情報処理装置、制御システムおよびレポート出力方法
JP2023002962A (ja) 情報処理装置、モデル生成プログラムおよびモデル生成方法
TWI830193B (zh) 預測系統、資訊處理裝置以及資訊處理程式
JP2023151888A (ja) 予測システム、制御装置および制御プログラム
JP2023151755A (ja) 情報処理装置および情報処理プログラム
JP2023151886A (ja) 情報処理装置および情報処理プログラム
CN116894544A (zh) 数据收集装置、记录介质以及方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22875394

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22875394

Country of ref document: EP

Kind code of ref document: A1