SYSTEM FOR DETERMINING A STABLE RANGE, METHOD FOR DETERMINING A STABLE RANGE AND PROGRAM FOR DETERMINING A STABLE RANGE

Abstract

System for determining a stable range (500) which determines a stable range of a multi-level signal in operating data containing the multi-level signal, wherein the system for determining a stable range comprises: a conversion unit (122) to set at least one threshold for the multi-level signal contained in the operating data and to convert the multi-level signal into at least one binary signal using the threshold; a prediction unit (123) for inputting the binary signal converted by the conversion unit into a prediction model (133) that predicts a stable signal value of the operational data, and for calculating a prediction value of the binary signal converted by the conversion unit as a converted binary signal prediction value; and a range determination unit (126) to calculate, based on the converted binary signal prediction value and the threshold, a probability that a signal value of the multi-level signal contained in the operating data exists in a range determined on the basis of the threshold, and to determine the stable range of the multi-level signal contained in the operating data on the basis of the probability.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a stable region determination system, a stable region determination method, and a stable region determination program. More specifically, the present disclosure relates to a stable region determination system, a stable region determination method, and a stable region determination program that determine a stable region of a multi-level signal in operational data. BACKGROUND TO THE STATE OF THE ART

[0002] When a malfunction occurs in a conventional factory, such as a production line stoppage, the factory maintenance personnel determine the cause of the malfunction based on their knowledge or experience and take appropriate measures. However, it is often difficult to identify the cause and quickly resolve the problem from the enormous amount of operational data and complicated programs. Furthermore, it is difficult to establish or create a comprehensive malfunction investigation program with a realistic number of man-hours.

[0003] Patent Literature 1 discloses a system that allows maintenance personnel to obtain clues for determining a sensor or program that is causing a malfunction, without specifying exhaustive conditions. Patent Literature 1 discloses a system that automatically detects unstable or non-stationary temporal changes in a binary signal expressing two values ​​such as the ON and OFF of a sensor, and in a multi-level signal such as a current value and a pressure value that takes values ​​other than 0 and 1.Patent Literature 2 discloses an anomaly detection system comprising a computing device that performs processing of learning a prediction model that predicts a behavior of a monitoring target device based on operation data on the device, processing of adjusting an anomaly score so that the anomaly score for operation data during normal operation falls within a predetermined range, the anomaly score being based on a deviation of the operation data acquired from the monitoring target device from a prediction result obtained by the prediction model, processing of detecting an anomaly or an indication of an anomaly based on the adjusted anomaly score, and processing of displaying information on at least one of the anomaly scores and a result of the detection on an output device.Patent Literature 3 discloses an instability detection device that can detect the operating state of plants using binary digital signals, the instability detection device including a unit for generating a normal model for determining operating states of a plurality of plants based on operating data that are binary digital signals obtained from the plants in their stable operating states, an expected value calculation unit for calculating an expected value of operating data by applying the normal model to past operating data of the plants, and an instability detection unit for detecting whether an operating state of one of the plants is unstable or not by comparing the expected value of the operating data and a measured value of the operating data.Patent Literature 4 discloses an acquisition unit that acquires an operating value by measuring the operating value of components of an industrial facility. A real-time database server stores the operating value and acquisition time information. A signal retrieval unit retrieves the operating values ​​from the real-time database server at a unit sampling time. A signal processing unit generates measurement data by calculating a minimum value, an average value, a maximum value, and a first and third quartile of the specified operating value. A normal range output unit designates the first quartile as a normal low value and the third quartile as a normal high value by reading measurement data for a fixed time.Patent Literature 5 discloses a shock absorber made of a blow-molded product made of thermoplastic resin, which can reduce the impact of an impact on the human body over a wide impact response range and whose shock absorption performance does not deteriorate even under oblique impacts. REFERENCE LISTPATENT LITERATURE Patent literature 1: JP 6 790 311 B1 Patent literature 2: EP 3 379 360 A2 Patent literature 3: DE 11 2017 007 606 T5 Patent literature 4: KR 10 1 170 329 B1 Patent Literature 5: JP 2006-67 465 A DEMOLITION PROBLEM

[0004] The method of Patent Literature 1 converts a multi-level signal into a binary signal, predicts a normal value of the binary signal, and detects an unstable or unsteady change in the signal. When an unstable change in the multi-level signal is detected, an unstable part of the converted binary signal is specified, and a value that the multi-level signal should take in a stable or steady state is obtained as a prediction value. However, it is not immediately obvious how the multi-level signal takes an unstable value before conversion to the binary signal. When a disturbance such as a production line stoppage occurs, it is necessary to examine how the value of the multi-level signal differs from that in the normal state to determine the cause of the disturbance.

[0005] In the present disclosure, a stable or stationary range of a multi-level signal is determined based on the probability that a signal value of the multi-level signal lies within a range determined based on a threshold value. The objective of the present disclosure is to thus provide an operator with an easily understandable indication of the signal value the multi-level signal assumes in comparison to the stable or stationary range. SOLUTION TO THE PROBLEM

[0006] A stable region determination system according to the present disclosure that determines a stable region of a multi-level signal in operating data including the multi-level signal, comprising: a conversion unit for setting at least one threshold value for the multi-level signal contained in the operating data and converting the multi-level signal into at least one binary signal using the threshold value; a prediction unit for inputting the binary signal converted by the conversion unit into a prediction model that predicts a stable signal value of the operating data, and for calculating a prediction value of the binary signal converted by the conversion unit as a converted binary signal prediction value; and a range determining unit for calculating, based on the converted binary signal prediction value and the threshold value, a probability that a signal value of the multi-level signal included in the operation data exists in a range determined based on the threshold value, and for determining the stable range of the multi-level signal included in the operation data based on the probability. ADVANTAGEOUS EFFECTS OF THE INVENTION

[0007] A stable range determination system according to the present disclosure determines a stable range of a multi-level signal based on a probability that a signal value of the multi-level signal lies within a range determined based on a threshold. Therefore, with the stable range determination system according to the present disclosure, the stable range of the multi-level signal can be appropriately determined, and the signal value that the multi-level signal takes compared to that in the stable range can be displayed to an operator in an easily understandable manner. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a diagram illustrating a configuration example of a stable area determination system according to Embodiment 1. Fig. 2 is a diagram showing a configuration example of a stable region determination apparatus according to Embodiment 1. Fig. 3 is a diagram showing a functional configuration example of a model generation unit according to Embodiment 1. Fig. 4 is a diagram showing a functional configuration example of a determination unit according to Embodiment 1. Fig. 5 is a general flowchart of a stable area determination process by the stable area determination apparatus according to Embodiment 1. Fig. 6 is a diagram showing a specific example of a conversion process according to Embodiment 1. Fig. 7 is a diagram showing an example of input / output of a prediction model according to Embodiment 1. Fig. 8 is a diagram illustrating an example in which prediction values ​​of a signal are output in a time series in a prediction process according to Embodiment 1. Fig. 9 is a diagram illustrating an example in which prediction values ​​in three signals in a time series are output in the prediction process according to Embodiment 1. Fig. 10 is a diagram illustrating an example of calculating the probability that a signal value of a multi-level signal is within a range according to Embodiment 1. Fig. 11 is a detailed flowchart of a method for calculating an in-range probability of the signal value of the multi-level signal according to Embodiment 1. Fig. 12 is a diagram showing a specific example of a first determination method of the stable region determination process according to Embodiment 1. Fig. 13 is a flowchart showing an example of a second determination method of the stable region determination process according to Embodiment 1. Fig. 14 is a flowchart showing another example of the second determination method of the stable region determination process according to Embodiment 1. Fig. 15 is a diagram showing a specific example of a third determination method of the stable region determination process according to Embodiment 1. Fig. 16 is a diagram showing a specific example of a fifth determination method of the stable region determination process according to Embodiment 1. Fig. 17 is a diagram illustrating a configuration example of a stable region determining apparatus according to a modification of Embodiment 1. DESCRIPTION OF THE EMBODIMENTS

[0008] The present embodiment will be described below with reference to the drawings. In the drawings, like or equivalent parts are designated by like reference numerals. In describing the embodiment, explanations of like or equivalent parts will be omitted or simplified as appropriate. Furthermore, in the following drawings, the dimensional relationships between constituent parts may appear different from the actual ones. In describing the embodiment, reference may be made to orientations or positions such as top, bottom, left, right, front, back, front and back. These terms are used for descriptive purposes and do not limit the arrangement, direction, and orientation of a device, tool, component, and the like.

[0009] Embodiment 1. *** Description of configurations ***

[0010] Fig. 1 is a diagram illustrating a configuration example of a stable region determination system 500 according to the present embodiment.

[0011] The stable region determination system 500 includes a stable region determination device 100, a data collection server 200, and a target system 300.

[0012] The stable range determination device 100 monitors the target system 300, e.g., a production line. In the target system 300, there are equipment 301 to equipment 305. In Fig. 1, five units of equipment are shown. However, there is no limit to the number of units of equipment. Each equipment consists of a plurality of devices, such as a sensor and a robot. Each device is connected to a network 401, and the equipment's operating data 31 is collected in the data collection server 200. The operating data 31 includes a binary signal and a multi-level signal. The binary signal is a signal expressing, for example, the ON and OFF of a sensor. The multi-level signal is a signal expressing, for example, a torque value of a robot hand.

[0013] The data collection server 200 is connected to the stable area determination device 100 via a network 402.

[0014] The stable region determination device 100 determines a stable or stationary region of the multi-level signal in the equipment's operating data 31. The stable region determination device 100 also detects instabilities or irregularities in the operating data 31. The stable region determination device 100 also indicates the stability or instability of the operating data 31. The stable region determination device 100 is also referred to as an instability detection device or an instability indication device.

[0015] Fig. 2 is a diagram showing a configuration example of the stable region determining apparatus 100 according to the present embodiment.

[0016] The stable region determination device 100 is a computer. The stable region determination device 100 is equipped with a processor 910 and other hardware devices such as a memory 921, an auxiliary storage device 922, an input interface 930, an output interface 940, and a communication device 950. The processor 910 is connected to the other hardware devices via a signal line and controls the other hardware devices.

[0017] The stable region determination device 100 is equipped with a model generation unit 110, a determination unit 120, and a storage unit 130 as functional elements. An operation database 131, a threshold group database 132, and a prediction model 133 are stored in the storage unit 130.

[0018] The functions of the model generation unit 110 and the determination unit 120 are implemented by software. The storage unit 130 is provided on the main memory 921. The storage unit 130 can be supplied to the auxiliary storage device 922, or it can be supplied to the memory 921 and the auxiliary storage device 922 by distribution.

[0019] Processor 910 is a device running a stable region determination program. The stable region determination program is a program that implements the functions of model generation unit 110 and determination unit 120.

[0020] Processor 910 is an integrated circuit (IC) that performs computational processing. Specific examples of processor 910 include a central processing unit (CPU), a digital signal processor (DSP), and a graphics processing unit (GPU).

[0021] The memory 921 is a storage device in which data is temporarily stored. A specific example of the memory 921 is a static random-access memory (SRAM) or a dynamic random-access memory (DRAM).

[0022] The auxiliary storage device 922 is a storage device in which data is stored. A concrete example of the auxiliary storage device 922 is an HDD. The auxiliary storage device 922 may be a portable storage medium, such as an SD (registered trademark), a memory card, a CF, a NAND flash, a flexible disk, an optical disk, a compact disk, a Blu-ray disk (registered trademark), and a DVD. Note that HDD stands for Hard Disk Drive, SD (registered trademark) for Secure Digital, CF for CompactFlash (registered trademark), and DVD for Digital Versatile Disc.

[0023] The input interface 930 is a connector for connecting to an input device, such as a mouse, a keyboard, and a touch panel. Specifically, the input interface 930 is a Universal Serial Bus (USB) connector. The input interface 930 may be a connector for connecting to a local area network (LAN).

[0024] The output interface 940 is a connector to which a cable of a display device, such as a display, can be connected. Specifically, the output interface 940 is a USB port or a High-Definition Multimedia Interface (HDMI; registered trademark) port. The display is specifically a liquid crystal display (LCD). The output interface 940 is also referred to as a display interface.

[0025] The communication device 950 includes a receiver and a transmitter. The communication device 950 is connected to a communication network such as a LAN, the Internet, or a telephone line. Specifically, the communication device 950 is a communication chip or a network interface card (NIC).

[0026] The stable region determination program is executed in the stable region determination device 100. The stable region determination program is read by the processor 910 and executed by the processor 910. The memory 921 stores not only the stable region determination program but also an operating system (OS). The processor 910 executes the stable region determination program while the operating system is running. The stable region determination program and the operating system may be stored in the auxiliary storage device 922. The stable region determination program and the operating system stored in the auxiliary storage device 922 are loaded into the memory 921 and executed by the processor 910. The stable region determination may be partially or completely integrated into the operating system.

[0027] The stable region determination device 100 may be equipped with a plurality of processors that replace the processor 910. The plurality of processors jointly execute the stable region determination program. Each processor is a device that executes the stable region determination program in the same way as the processor 910.

[0028] Data, information, signal values, and variable values ​​used, processed, or output by the program to determine a stable region are stored in memory 921, auxiliary storage device 922, or a register or cache memory in processor 910.

[0029] The term "unit" in the model generation unit 110 and the determination unit 120 may be replaced with "circuit," "stage," "method," "process," or "circuit." The stable region determination program causes the computer to execute a model generation process and a determination process. The term "process" in the model generation and determination processes may be replaced with "program," "program product," "program-stored computer-readable medium," or "program-recorded computer-readable medium." A stable region determination method is a process performed by the stable region determination device 100 that executes the stable region determination program.

[0030] The program for determining a stable region may be provided in the form of a computer-readable recording medium. The program for determining a stable region may be provided as a program product.

[0031] Fig. 3 is a diagram showing a functional configuration example of the model generation unit 110 according to the present embodiment.

[0032] A solid arrow in Fig. 3 expresses a call relationship between function elements, and dashed arrows in Fig. 3 express data flows between functional elements and the databases.

[0033] The model generation unit 110 generates the prediction model 133 for predicting the next signal value of the operating data during normal operation of the equipment. In other words, the model generation unit 110 generates the prediction model 133 for predicting the signal value of a stable or steady-state operating data.

[0034] The model generation unit 110 is equipped with an acquisition unit 111, a threshold group calculation unit 112, a conversion unit 113 and a learning unit 114.

[0035] The acquisition unit 111 receives the operation data from the data collection server 200 via the communication device 950 and stores the operation data in the operation database 131. The operation data is, for example, data such as a binary signal expressing the switching on and off of a sensor or a multi-level signal expressing a torque value of the robot hand. A process of receiving and storing the operation data is performed with the necessary data as the target each time the data in the data collection server 200 increases, as much as possible in real time.

[0036] The threshold group calculation unit 112 acquires the operation data from the operation database 131, calculates a threshold for converting a multi-level signal in the operation data into a binary signal, and stores the threshold in the threshold group database 132.

[0037] The conversion unit 113 obtains the threshold value from the threshold group database 132 and converts the multi-level signal into the binary signal based on the threshold value.

[0038] The learning unit 114 acquires the operating data from the operating database 131 and calls the conversion unit 113 to convert the multi-level signal in the acquired operating data into a binary signal using the conversion unit 113. The learning unit 114 learns a normal signal pattern of the signal contained in the operating data from the binary signal contained in the operating data and the binary signal converted by the conversion unit 113 from the multi-level signal contained in the operating data. The learning unit 114 then stores a learned model that predicts the learned normal signal pattern as a prediction model 133.

[0039] The threshold group calculation unit 112 sets the threshold value so that, for example, the signal value of the multi-level signal is converted into a binary signal that switches when a value trend changes, such as rising, falling, or remaining constant. The threshold value for converting the multi-level signal into the binary signal can be set to any value, and any number of threshold values ​​can be set. Note that there are no restrictions on the calculation of the threshold value.

[0040] Fig. 4 is a diagram showing a functional configuration example of the determination unit 120 according to the present embodiment.

[0041] Solid arrows in Fig. 4 express the calling relationships between the function elements, and dashed arrows express the data flow between the function elements and the databases.

[0042] The determination unit 120 predicts a next signal value of a signal in normal operation from the operation data, judges whether the next signal value is unstable or not, identifies an unstable part and determines a stable range, and displays the stable range together with the operation data.

[0043] The determination unit 120 is equipped with an acquisition unit 121, a conversion unit 122, a prediction unit 123, a judgment unit 124, an identification unit 125, an area determination unit 126 and a display unit 127.

[0044] Just like the acquisition unit 111 in the model generation unit 110, the acquisition unit 121 receives the operation data from the data collection server 200 via the communication device 950 and stores the operation data in the operation database 131.

[0045] Just like the conversion unit 113 in the model generation unit 110, the conversion unit 122 acquires the threshold value from the threshold group database 132 and converts the multi-level signal into a binary signal based on the threshold value.

[0046] Since the operating data is a binary signal and the binary signal has been converted by the conversion unit 122, the prediction unit 123 calculates prediction values, which are stable values ​​of a signal value to be output next, using the prediction model 133. All inputs to the prediction model 133 are binary signals. Hereinafter, the binary signal converted by the conversion unit 122 from the multi-level signal in the operating data, that is, the binary signal output by the conversion unit 122, is referred to as the converted binary signal.

[0047] Note that the judgment unit 124 obtains the operation data from the operation database 131 and calls the conversion unit 122 and the prediction unit 123, so that a conversion process is executed by the conversion unit 122 and a prediction process is executed by the prediction unit 123.

[0048] The judgment unit 124 compares the binary signal in the operating data and an actual measured value of the converted binary signal with the prediction values ​​output by the prediction unit 123. Based on the comparison result, the judgment unit 124 judges whether the operating data is stable or not, that is, whether the operating data conforms to the learned normal signal pattern. The judgment unit 124 outputs a judgment result as instability judgment information. If it is determined that the operating data is unsteady, the judgment unit 124 calls the identification unit 125 and identifies an unstable part with the identification unit 125. The judgment unit 124 also calls the display unit 127 and displays the judgment result on the display device through the display unit 127.

[0049] The identification unit 125 identifies which signal was unstable and when it was unstable based on the binary signal in the operating data and the converted binary signal, as well as the prediction value of the binary signal and the prediction value of the converted binary signal. The identification unit 125 outputs the identified information as instability identification information.

[0050] The range determining unit 126 determines, based on the prediction value of the converted binary signal, a stable range of the signal value in the multi-level signal as before conversion into the converted binary signal.

[0051] The display unit 127 can determine the unstable region in the multi-level signal by calling the region determination unit 126.

[0052] Using the stable region in the multi-level signal, the display unit 127 visibly displays information such as the actual measured value of the operating data, the prediction values ​​outputted by the prediction unit 123, the instability judgment information outputted by the judgment unit 124, and the instability identification information outputted by the identification unit 125 on the display device so that they can be easily recognized. *** Description of functions ***

[0053] The following describes the operation of the stable area determination system 500 according to the present embodiment. An operation of the stable area determination system 500 corresponds to the stable area determination method. A program that implements the operations of the stable area determination system 500 corresponds to the stable area determination program that causes the computer to execute a stable area determination process. The operations of the stable area determination system 500 are operations of the individual devices of the stable area determination system 500. <Prozess zur Bestimmung eines stabilen Bereichs >

[0054] Fig. 5 is a general flowchart of a stable region determination process by the stable region determination apparatus 100 according to the present embodiment.

[0055] With reference to Fig. 5, details of step S107 “calculation process of an existence probability of a signal value of a multi-level signal” and step S108 “determination process of a stable range of the signal value of the multi-level signal” will be described later. < <erwerbungsprozess>>

[0056] In step S101, the acquisition unit 121 copies the operation data from the data collection server 200 to the operation database 131 via the communication device 950. For example, if the operation data output from the data collection server 200 includes a binary signal expressing ON and OFF of a sensor and a multi-level signal expressing a torque value of a robot hand, both the binary signal and the multi-level signal are stored in the operation database 131 as operation data.

[0057] The prediction process by the prediction unit 123 requires operating data covering a specified period in the past. Therefore, the operating data required for the prediction process for the past, specified period is stored in the operating database 131.

[0058] It should be noted that the acquisition unit 121 copies the operation data from the data collection server 200 to the operation database 131 in real time as much as possible. < <umwandlungsprozess>>

[0059] In step S102, the conversion unit 122 converts signal data of the multi-level signal from the operating data stored in the operating database 131 into signal data of the binary signal. The conversion unit 122 sets at least one threshold value for the multi-level signal contained in the operating data and converts the multi-level signal into at least one binary signal using the threshold value.

[0060] Specifically, the conversion unit 122 retrieves the threshold value from the threshold group database 132. Based on the threshold value, the conversion unit 122 converts the signal data of the multi-level signal into the signal data of the binary signal from the operation data stored in the operation database 131. The details of the conversion process will be described later. < <vorhersageprozess>>

[0061] In step S103, the prediction unit 123 predicts the next signal values ​​from the past binary signal stored in the operational database 131 and from the converted binary signal converted from the past multi-level signal stored in the operational database 131. The prediction model 133 created in advance by the model generation unit 110 is used for the prediction.

[0062] The prediction unit 123 inputs the original binary signal contained in the operational data and the converted binary signal to the prediction model 133 and outputs prediction values ​​that are stable signal values ​​of the signal contained in the operational data. Specifically, with respect to the binary signal converted by the conversion unit 122 (converted binary signal), the prediction unit 123 inputs the converted binary signal to the prediction model 133 and outputs a prediction value of the converted binary signal as the prediction value of the converted binary signal. < <beurteilungsprozess>>

[0063] In step S104, the judgment unit 124 compares the predicted value of the operation data calculated in step S103 with the actual measured value of the signal of the operation data stored in the operation database 131 and calculates an abnormality degree.

[0064] In step S105, the judging unit 124 judges whether the operation data is stable or unstable based on the abnormality degree calculated in step S104.

[0065] If it is determined that the operating data is unstable, the judging unit 124 proceeds to step S106. If it is determined that the operating data is stable, the judging unit 124 proceeds to step S107. < <bereich-bestimmungsprozess>>

[0066] In step S106, the identification unit 125 determines which signal was unstable and when. Specifically, the identification unit 125 can identify an unstable part by extracting a signal whose prediction value and actual measurement value differ by a specified value or more, and by extracting a time point at which the difference occurred. < <bereich-bestimmungsprozess>>

[0067] With reference to step S107 and step S108, an area determination process by the area determination unit 126 will now be described.

[0068] In step S107, the range determination unit 126 calculates a probability that the signal value of the multi-level signal lies within the range from the prediction value of the signal of the operating data calculated in step S103. Specifically, the range determination unit 126 calculates a probability that the signal value of the multi-level signal included in the operating data lies within a range determined based on the threshold value, based on the converted binary signal prediction value and the threshold value.

[0069] The converted binary signal prediction value is the prediction value of the converted binary signal obtained by inputting the binary signal converted by the conversion unit 122 to the prediction model 133. The threshold value is the threshold value used in converting the multi-level signal into the binary signal.

[0070] In step S108, the range determining unit 126 determines the stable range of the multi-level signal included in the operation data based on the probability calculated in step S107 that the signal value of the multi-level signal is in the range.

[0071] In step S109, the display unit 127 presents the user with a judgment result of the binary signal or multi-level signal included in the operational data. In the example of the present embodiment, the judgment result is presented to the user by displaying it on the display device. Alternatively, the judgment result may be presented to the user by another method, such as outputting the result to a printer or outputting the result as electronic data.

[0072] The display unit 127 displays the signal behavior in the form of a time series. If the signal is a binary signal, the display unit 127 displays a predicted value of the binary signal that expresses the normal behavior.

[0073] In the case of a multi-level signal, the display unit 127 displays the signal value of the multi-level signal by superimposing it on the range including the stable range determined based on the threshold value. For example, the display unit 127 may display a background of the stable range determined in step S108 in a first color (e.g., green), display the background deviating from the stable range in a second color (e.g., yellow) or a third color (e.g., red) according to the degree of deviation from the stable range, and superimpose it on the signal value of the multi-level signal. Furthermore, the display unit 127 may display a line indicating the signal value deviating from the stable range in a second color (e.g., yellow) or a third color (e.g., red), depending on the degree of deviation.

[0074] Each process is described in detail.

[0075] Fig. 6 is a diagram showing a specific example of the conversion process according to the present embodiment.

[0076] The conversion unit 122 converts the multi-level signal into at least one binary signal using at least one threshold value. The multi-level signal does not always need to be converted into a binary signal if multiple threshold values ​​are used. The multi-level signal is converted into binary signals having a certain number of threshold values. If two threshold values ​​are used for the multi-level value as in Fig. 6, the multi-level signal is converted into two binary signals.

[0077] Specifically, the conversion unit 122 converts the multi-level signal into a binary signal, which takes the value 1 when the signal value of the multi-level signal exceeds the threshold value at any time, and takes the value 0 otherwise.

[0078] Fig. 7 is a diagram showing an example of the input / output of the prediction model 133 according to the present embodiment.

[0079] The prediction model 133 learns a signal pattern of a normal binary signal and outputs a prediction value of the signal. The prediction value is a real number value from 0 or more to 1 or less, as shown in Fig. 7, and corresponds to the probability that the signal value will be 1 at the next time point. The output does not show a temporal change pattern of the binary signal, but expresses a prediction value of each binary signal for only one next time point.

[0080] Suppose that signal 1 takes the values ​​0, 0, 1, 1, 1 as previous signal data, and signal 2 takes the values ​​1, 1, 1, 1, 0. When these values ​​are input into the prediction model, a value of 0.8 is output as the prediction value of signal 1, and a value of 0.2 is output as the prediction value of signal 2. This means that at this time, the probability that the value of signal 1 will be 1 at the next time is 0.8, and the probability that the value of signal 2 will be 1 at the next time is 0.2.

[0081] Fig. 8 is a diagram showing an example in which the prediction values ​​of a signal in the prediction process according to the present embodiment are output in a time series.

[0082] In Fig. 8, the prediction is performed repeatedly, and the prediction values ​​of the individual time points are arranged in a time series. For simplicity, Fig. 8 the input / output is a single signal, ie a binary signal obtained from a threshold value.

[0083] Fig. 9 is a diagram showing an example in which prediction values ​​in three signals in a time series are output in the prediction process according to the present embodiment.

[0084] In Fig. Figure 9 shows the prediction values ​​of three signals in a time series. A large number of signal values ​​from the same time point are output all at once by the prediction model. That is, prediction value 1 to prediction value 4 in Fig. 9 are all output together at once by the prediction model.

[0085] Fig. 10 is a diagram illustrating an example of calculating the probability that the signal value of the multi-level signal is in the range according to the present embodiment.

[0086] As described above, the prediction value output by the prediction unit 123 is a real number value from 0 or more to 1 or less, and corresponds to the probability that the signal value becomes 1 at any time. Therefore, the prediction value of the binary signal converted from the multi-level signal, which becomes 1 when the signal value exceeds a threshold value and 0 otherwise, corresponds to the probability that the signal value exceeds this threshold value. The probability that the signal value is in a range between two threshold values ​​is given by the following formula (1). (Probability that the signal value lies in a range between two thresholds) = (Probability that the signal value exceeds a lower threshold) − (Probability that the signal value exceeds an upper threshold)

[0087] The probability that the signal value lies in a range above a maximum threshold and the probability that the signal value lies in a range below a maximum threshold are given by formula (2) and formula (3), respectively. (Probability that the signal value lies in a range above a maximum threshold) = (Probability that the signal value exceeds the maximum threshold) (Probability that the signal value lies in a range below a minimum threshold) = 1 − (Probability that the signal value exceeds a minimum threshold)

[0088] As described above, the probability that the signal value of the multi-level signal is within the range is calculated from the prediction value of the converted binary signal by setting a threshold for the multi-level signal. The probability is a real numerical value from 0 or more to 1 or less.

[0089] In the conversion unit 122, multiple threshold values ​​can be set for the multi-level signal, and the multi-level signal can be converted into a binary signal that takes 1 when the signal value exceeds a threshold value and takes 0 otherwise. In this case, the prediction value of the binary signal corresponds to the probability that a signal value falls below the threshold value at any time. The probability that the signal value lies in a range between two threshold values, the probability that the signal value lies in a range above a maximum threshold value, and the probability that the signal value lies in a range below a minimum threshold value are given by formula (4), formula (5), and formula (6), respectively. (Probability that the signal value lies in a range between two thresholds) = (Probability that the signal value falls below an upper threshold) − (Probability that the signal value falls below a lower threshold) (Probability that the signal value lies in a range above a maximum threshold) = 1 − (Probability that the signal value falls below the maximum threshold) (Probability that the signal value lies in a range below a minimum threshold) = (Probability that the signal value falls below a minimum threshold)

[0090] Fig. 11 is a detailed flowchart of a method for calculating an in-range probability of the signal value of the multi-level signal according to the present embodiment.

[0091] In step S201, the range determining unit 126 selects a non-selected threshold value from the plurality of threshold values ​​used in converting the multi-level signal into the binary signal.

[0092] In step S202, the range determination unit 126 judges whether or not a threshold value smaller than the selected threshold exists. If such a threshold value exists, the range determination unit 126 proceeds to step S203. If such a threshold value does not exist, the range determination unit 126 proceeds to step S204.

[0093] If a threshold value with a smaller value than the selected threshold value exists, then the range determining unit 126 calculates a probability that the signal value exists in a range between the selected threshold value and a lower threshold value besides the selected threshold value in step S203.

[0094] If there is no threshold value with a value smaller than the selected threshold value, the range determining unit 126 calculates a probability that the signal value is in a range below the minimum threshold value in step S204.

[0095] In steps S205 and S206, the range determination unit 126 judges whether there is a non-selected threshold. If there is a non-selected threshold, the range determination unit 126 returns to step S201 and repeats the processing until there is no non-selected threshold.

[0096] If there is no unselected threshold, the range determining unit 126 calculates a probability that the signal value is in a range above the maximum threshold in step S207.

[0097] The following describes the procedure for determining a stable region of the multi-level signal. < First determination method for the process of determining a stable range >

[0098] Fig. 12 is a diagram showing a specific example of a first determination method of the stable region determination process according to the present embodiment.

[0099] According to the first determination method, the range determination unit 126 determines, from the ranges each determined based on the threshold value, a range in which the probability is a certain value or more than the stable range. The determined value is a predetermined value.

[0100] Specifically, the range determining unit 126 takes a range in which the probability of the signal value at the same point has a fixed value or more as the stable range. Fig. Figure 12 shows an example in which an area where the probability is 0.5 or more is set as a stable area. < Second determination method for the process of determining a stable range >

[0101] According to the second determination method, the area determination unit 126 determines, from the areas each determined based on the threshold value, an area in which the probability is maximum as the stable area.

[0102] Specifically, the range determining unit 126 takes a range in which the probability of the signal value at the same point is maximum as the stable range.

[0103] Fig. 13 is a flowchart showing an example of the second determination method of the stable region determination process according to the present embodiment.

[0104] Fig. 13 shows a determination procedure after selecting a range with descending probability order.

[0105] In the determination process after selecting an area in descending order of probability, the area determination unit 126 selects areas in descending order of probability from the areas each determined based on the threshold value, and determines the selected areas as the stable area until the probabilities in total reach a certain value or more.

[0106] Specifically, the range determination unit selects 126 ranges in descending order of the probability of the same time point and takes the selected ranges as the stable range until the probabilities sum to a set value or more.

[0107] In step S301, the range determining unit 126 selects a non-selected range in which the probability of the value is maximum.

[0108] In step S302, the area determining unit 126 repeats step S301 until the probabilities of the selected areas as a whole reach the specified value or more.

[0109] In step S303, when the probabilities of the selected areas reach the set value or more, the area determining unit 126 determines the selected areas each as the stable area.

[0110] Fig. 14 is a flowchart showing another example of the second determination method of the stable region determination process according to the present embodiment.

[0111] Fig. 14 illustrates a determination method according to the selection of a neighboring maximum probability region.

[0112] In the determination method according to the maximum probability selection of adjacent regions, the region determination unit 126 repeatedly selects a region where the probability is maximum from the regions each determined based on the threshold value, and selects a region where the probability is higher from the regions adjacent to the selected region. The region determination unit 126 repeatedly determines selected regions until the probabilities of the selected regions add up to a certain value or more, each as a stable region.

[0113] Specifically, the range determining unit 126 repeatedly selects a range in which the probability of the same time point is maximum, and selects a range in which the probability is larger from the ranges adjacent to the selected range; and takes the selected ranges as the stable range until the probabilities sum to a set value or more.

[0114] In step S401, the range determining unit 126 determines a range in which the probability of the value is maximum as the stable range, respectively.

[0115] In step S402, if the total probabilities of the stable regions do not reach the specified value or more, the region determining unit 126 proceeds to step S403. If the probabilities of the stable regions reach the specified value or more, the region determining unit 126 terminates the processing.

[0116] In step S403, the region determining unit 126 determines, from the regions adjacent to the stable region, a region in which the probability is higher than the stable region. The region determining unit 126 repeats step S402 and step S403 until the probabilities of the stable regions reach the specified value or more. < Third Determination Method for the Stable Range Determination Process >

[0117] Fig. 15 is a diagram showing a specific example of a third determination method of the stable region determination process according to the present embodiment.

[0118] According to the third method, the area determining unit 126 determines, from the areas each determined based on the threshold value, an area in which a probability density, which is a value obtained by dividing the probability by a width of the area, has a certain value or more, as the stable area.

[0119] Specifically, the range determining unit 126 takes a range in which a probability density of the signal value at the same time point has a fixed value or more as the stable range. Fig. Figure 15 shows a case where probability densities are calculated and areas where the probability densities are 0.0100 or more are determined as stable areas.

[0120] When determining the stable range, the wider the range, the higher the probability of the value. Therefore, the stable range is determined based on the probability density, so that it is possible to highly evaluate the degree of stability of a range with a narrow width and correspondingly low probability. < Fourth Determination Method for the Stable Range Determination Process >

[0121] The fourth determination method describes variants of the determination method that use the probability density.

[0122] The area determining unit 126 may determine an area in which the probability is highest as the stable area from the areas each determined based on the threshold value.

[0123] Specifically, the range determining unit 126 takes a range in which the probability density of the same time point is maximum as a stable range.

[0124] Alternatively, the area determining unit 126 may select areas in descending order of probability density from the areas each determined based on the threshold values, and may determine the selected areas as the stable area until the total probability densities reach a certain value or more.

[0125] Specifically, the range determining unit 126 selects a range in descending order of the probability of the same time point, and takes the selected ranges as the stable range until the probability densities sum up to a set value or more.

[0126] Alternatively, the area determination unit 126 repeatedly selects an area where the probability density is maximum from the areas each determined based on the threshold value, and selects an area where the probability density is larger from the areas adjacent to the selected area. The area determination unit 126 may then determine areas that are repeatedly selected until the probability densities add up to a certain value or more, each as a stable area.

[0127] Specifically, the area determining unit 126 repeatedly selects an area where the probability density at the same time point is maximum, and selects an area where the probability density is larger from the areas adjacent to the selected area; and takes the selected areas as the stable area until the probability densities in total reach a set value or more. < Fifth Determination Method for the Stable Range Determination Process >

[0128] In a fifth determination method of the stable region determination process, the region determination unit 126 may gradually determine an unstable region when determining a stable region of a multi-level signal.

[0129] In a first method for determining an unstable region step by step, the region determining unit 126 determines, with respect to the regions each determined based on the threshold value, an instability degree of a region that is unstable according to the probability.

[0130] Specifically, the range determination unit 126 determines the degree of instability of the range based on the probability of the value at the same time point. For example, if a range where the probability is 0.5 or more is determined as stable, a range where the probability is 0.2 or more to less than 0.5 is determined as slightly unstable, and a range where the probability is less than 0.2 is determined as severely unstable. Three or more levels of instability can be defined.

[0131] The area determining unit 126 may determine, with respect to the areas each determined based on the threshold value, an instability degree of an area that is unstable according to the probability density instead of the probability.

[0132] Fig. 16 is a diagram showing a specific example of a second method for stepwise determining the unstable region of the stable region determining process according to the present embodiment.

[0133] Fig. Figure 16 shows an example of the step-by-step determination of a stable area according to the separation efficiency from the stable area.

[0134] According to the second method for determining the unstable region step by step, the region determining unit 126 determines, with respect to the regions each determined based on the threshold value, an instability degree of a region that is unstable according to the degree of separation from the stable region.

[0135] In Fig. 16, the area determination unit 126 determines the degree of instability based on the separation degree between the area and the stable area. An area adjacent to the stable area is classified as slightly unstable, and an area two or more areas away from the stable area is classified as severely unstable. *** Other configurations ***

[0136] In the present embodiment, the functions of the model generation unit 110 and the determination unit 120 are implemented by software. In a modification, the functions of the model generation unit 110 and the determination unit 120 may be implemented by hardware.

[0137] Specifically, a stable region determining device 100 is equipped with an electronic circuit 909 instead of a processor 910.

[0138] Fig. 17 is a diagram showing a configuration example of the stable region determining apparatus 100 according to the modification of the present embodiment.

[0139] The electronic circuit 909 is a special electronic circuit that implements the functions of the model generation unit 110 and the determination unit 120. The electronic circuit 909 is, in particular, a single circuit, a composite circuit, a programmed processor, a parallel-programmed processor, a logic IC, a GA, an ASIC, or an FPGA. Note that GA stands for Gate Array, ASIC stands for Application Specific Integrated Circuit, and FPGA stands for Field-Programmable Gate Array.

[0140] The functions of the model generation unit 110 and the determination unit 120 may be implemented by one electronic circuit or by a plurality of electronic circuits through distribution.

[0141] According to another modification, some of the functions of the model generation unit 110 and the determination unit 120 may be implemented by an electronic circuit, and the remaining functions may be implemented by software. Some or all of the functions of the model generation unit 110 and the determination unit 120 may be implemented by firmware.

[0142] The processor and electronic circuitry are also called processing circuits. That is, the functions of the model generation unit 110 and the determination unit 120 are implemented by processing circuits. *** Description of the effect of the present embodiment ***

[0143] As described above, in the stable region determination device 100 according to the present embodiment, a stable range of a signal value of a multi-level signal is calculated based on the probability that the signal value of the multi-level signal lies between two threshold values. Therefore, with the stable region determination device 100 according to the present embodiment, it is possible to clearly show the operator how the signal value of the multi-level signal differs from the steady-state range.

[0144] In addition, with the stable range determining apparatus 100 according to the present embodiment, it is possible to calculate the stable range of the signal value of the multi-level signal based on the probability density in the range.

[0145] It is assumed that the probability that the signal value of the multi-level signal is within the range increases with the increase in the range width. Therefore, in the stable range determination device 100 according to the present embodiment, the stable range is determined based on the probability density, so that it is possible to appropriately evaluate the degree of stability of a range with a narrow width and correspondingly low probability.

[0146] In Embodiment 1 above, each stable region determination device is described as an independent functional block. However, the stable region determination device does not need to be configured as in the above-described embodiment. The functional block of the stable region determination device can have any configuration as long as it can perform the function described in the above-described embodiment. Also, the stable region determination device does not need to be composed of a single device, but may be a system composed of a plurality of devices.

[0147] Several parts of Embodiment 1 may be implemented in combination. Alternatively, a part of this embodiment may also be implemented. This embodiment may also be implemented in whole or in part by any combination.

[0148] That is, in Embodiment 1, various embodiments can be freely combined, any element of each embodiment can be changed, or any element can be omitted from each embodiment. LIST OF REFERENCE SYMBOLS

[0149] 31: Operational data; 100: Stable range determination device; 110: Model generation unit; 111, 121: Acquisition unit; 112: Threshold group calculation unit; 113, 122: Conversion unit; 114: Learning unit; 120: Determination unit; 123: Prediction unit; 124: Judgment unit; 125: Identification unit; 126: Range determination unit; 127: Display unit; 130: Storage unit; 131: Operational database; 132: Threshold group database; 133: Prediction model; 200: Data collection server; 300: Target system; 301, 302, 303, 304, 305: Equipment; 401, 402: Network; 500: Stable region determination system; 909: Electronic circuit; 910: Processor; 921: Random access memory; 922: Auxiliary storage device; 930: Input interface; 940: Output interface; 950: Communication device. < / beurteilungsprozess> < / vorhersageprozess> < / umwandlungsprozess> < / erwerbungsprozess>

Claims

[1] A stable region determination system (500) that determines a stable region of a multi-level signal in operating data including the multi-level signal, the stable region determination system comprising: a conversion unit (122) for setting at least one threshold value for the multi-level signal contained in the operating data and converting the multi-level signal into at least one binary signal using the threshold value; a prediction unit (123) for inputting the binary signal converted by the conversion unit into a prediction model (133) that predicts a stable signal value of the operating data, and for calculating a prediction value of the binary signal converted by the conversion unit as a converted binary signal prediction value; and a range determining unit (126) for calculating, based on the converted binary signal prediction value and the threshold value, a probability that a signal value of the multi-level signal included in the operation data exists in a range determined based on the threshold value, and for determining the stable range of the multi-level signal included in the operation data based on the probability. [2] A stable region determining system (500) according to claim 1, comprising a display unit (127) for displaying the signal value of the multi-level signal included in the operating data by superimposing it over the region including the stable region and determined based on the threshold value. [3] The stable region determining system (500) according to claim 1 or 2, wherein the region determining unit (126) determines, from the regions each determined based on the threshold value, a region in which the probability has a certain value or more as the stable region. [4] The stable region determining system (500) according to claim 1 or 2, wherein the region determining unit (126) determines a region in which the probability is maximum as the stable region from the regions each determined based on the threshold value. [5] The stable area determination system (500) according to claim 4, wherein the area determination unit (126) selects areas in a descending order of probability from the areas each determined based on the threshold value, and determines selected areas as the stable area until the probabilities reach a certain value or more. [6] The stable area determination system (500) according to claim 4, wherein the area determination unit (126) repeatedly selects an area in which the probability is maximum from areas each determined based on the threshold value, and selects an area in which the probability is larger from areas adjacent to the selected area; and determines selected areas, until the probabilities sum to a certain value or more, each as the stable area. [7] The stable area determination system (500) according to claim 1 or 2, wherein the area determination unit (126) determines, from areas each determined based on the threshold value, an area in which a probability density, which is a value obtained by dividing the probability by a width of the area, has a certain value or more, as the stable area. [8] The stable area determination system (500) according to claim 1 or 2, wherein the area determination unit (126) determines, from the areas each determined based on the threshold value, an area in which a probability density, which is a value obtained by dividing the probability by a width of the area, is maximum, as the stable area. [9] The stable area determination system (500) according to claim 8, wherein the area determination unit (126) selects areas in a descending order of the probability density, which is a value obtained by dividing the probability by a width of the area, from areas each determined based on the threshold value, and determines selected areas until the probability densities sum to a certain value or more, each as the stable area. [10] The stable area determination system (500) according to claim 8, wherein the area determination unit (126) repeatedly selects, from areas each determined based on the threshold value, an area in which a probability density, which is a value obtained by dividing the probability by a width of the area, is maximum, and selects, from areas adjacent to the selected area, an area in which the probability density is larger; and determines selected areas until the probabilities sum to a certain value or more, each as the stable area. [11] The stable region determining system (500) according to claim 3, wherein the region determining unit (126) determines, with respect to the regions each determined based on the threshold value, a degree of instability of a region that is not stable according to the probability. [12] The stable region determining system (500) according to claim 7, wherein the region determining unit (126) determines, with respect to the regions each determined based on the threshold value, an instability degree of a region that is not stable according to the probability density which is the value obtained by dividing the probability by the width of the region. [13] The stable region determining system (500) according to claim 1 or 2, wherein the region determining unit (126) determines, with respect to the regions each determined based on the threshold value, a degree of instability of a region that is not stable according to a degree of separation from the stable region. [14] A method for determining a stable region (500) used in a stable region determination system that determines a stable region of a multi-level signal in operational data including the multi-level signal, the method for determining a stable region comprising: by a computer, setting at least one threshold value for the multi-level signal contained in the operating data and converting the multi-level signal into at least one binary signal using the threshold value; by the computer, inputting the converted binary signal into a prediction model (133) that predicts a stable signal value of the operating data, and calculating a prediction value of the converted binary signal; and by the computer, calculating, based on the converted binary signal prediction value and the threshold value, a probability that a signal value of the multi-level signal included in the operating data exists in a range determined based on the threshold value, and determining the stable range of the multi-level signal included in the operating data based on the probability. [15] A stable region determination program (500) used in a stable region determination system that determines a stable region of a multi-level signal in operational data including the multi-level signal, the stable region determination program causing a computer to execute: a conversion process in which at least one threshold value is set for the multi-level signal contained in the operating data and the multi-level signal is converted into at least one binary signal using the threshold value; a prediction process in which the binary signal converted by the conversion process is input to a prediction model (133) that predicts a stable signal value of the operating data, and a prediction value of the binary signal converted by the conversion process is calculated as a converted binary signal prediction value; and a range determining process of calculating, based on the converted binary signal prediction value and the threshold value, a probability that a signal value of the multi-level signal included in the operation data exists in a range determined based on the threshold value, and determining the stable range of the multi-level signal included in the operation data based on the probability.

Images