Information processing device, information processing method, and information processing program

Abstract

To effectively detect an abnormal condition of a system.SOLUTION: An information processing device 10 disclosed herein is configured to: acquire a normal dataset included in each of multiple sections of measurement data measured by a device; combine the acquired normal datasets to generate combined data, and train a machine learning model designed to output a score indicative of the normal state of a system having the device provided therein from an input of the measurement data through machine learning using the combined data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] A conventional technique is known in which a machine learning model is created by displaying recorded past data files collected from a device on a display screen (viewer), selecting "normal data" from all of the data for a single period, and performing machine learning on the data as a single group. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-094037 Summary of the Invention [Problem to be solved by the invention]

[0004] However, with conventional technology, it is difficult to efficiently generate learning data for normal conditions to detect abnormal conditions in a system. For example, in a plant, for a device that operates normally in the morning, stops operation during a lunch break, and then operates normally in the afternoon, it is desirable to learn both the morning and afternoon data, but it is necessary to select one of them.

[0005] The present invention has been made in view of the above, and has as its object to effectively detect an abnormal state of a system. [Means for solving the problem]

[0006] The present invention provides an information processing device comprising: an acquisition unit that acquires, for each interval, normal data included in multiple intervals from measurement data measured by a device; a generation unit that combines the acquired normal data to generate synthetic data; and a training unit that trains, through machine learning using the synthetic data, a machine learning model that outputs a score indicating the normal state of a system in which the device is installed in response to input of the measurement data.

[0007] The present invention also provides an information processing method in which a computer executes processing, in which each normal data included in a plurality of intervals from measurement data measured by a device is acquired for each of the intervals, the acquired normal data is synthesized to generate synthetic data, and a machine learning model is trained using the synthetic data to output a score indicating the normal state of a system in which the device is installed in response to input of the measurement data.

[0008] The present invention also provides an information processing program that causes a computer to execute a process of acquiring, for each interval, normal data included in a plurality of intervals from measurement data measured by a device, synthesizing the acquired normal data to generate synthetic data, and training, through machine learning using the synthetic data, a machine learning model that outputs a score indicating the normal state of a system in which the device is installed in response to input of the measurement data. [Effects of the Invention]

[0009] According to the present invention, an abnormal state of the system can be effectively detected. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating an example of the configuration of an information processing system according to an embodiment. [Figure 2] FIG. 4 is a diagram illustrating an example of measurement data according to the embodiment. [Figure 3] 1 is a block diagram illustrating an example of the configuration of an information processing device according to an embodiment. [Figure 4] FIG. 2 is a diagram illustrating an example of a designated data storage unit of the information processing device according to the embodiment. [Figure 5] FIG. 2 is a diagram illustrating an example of a composite data storage unit of the information processing apparatus according to the embodiment. [Figure 6] FIG. 10 is a diagram showing an example of a display screen for measurement data according to the embodiment. [Figure 7] FIG. 10 is a diagram showing an example of a measurement data specification screen according to the embodiment. [Figure 8] FIG. 10 is a diagram showing measurement data designation processing 1 according to the embodiment. [Figure 9] FIG. 10 is a diagram showing measurement data designation processing 2 according to the embodiment. [Figure 10] FIG. 10 is a diagram showing measurement data designation processing 3 according to the embodiment. [Figure 11] 10 is a flowchart illustrating an example of the overall flow of information processing according to the embodiment. [Figure 12] FIG. 2 is a diagram illustrating an example of a hardware configuration. DETAILED DESCRIPTION OF THE INVENTION

[0011] An information processing device, an information processing method, and an information processing program according to an embodiment of the present invention will be described in detail below with reference to the accompanying drawings. Note that the present invention is not limited to the embodiment described below.

[0012] [Embodiment] The configuration of the information processing system 100 according to the embodiment, the configuration of the information processing device 10, and the processing flow will be described below in order, and finally the effects of the embodiment will be described.

[0013] 1. Configuration of Information Processing System 100 The configuration of an information processing system 100 according to an embodiment will be described in detail using Fig. 1. Fig. 1 is a diagram showing an example of the configuration of the information processing system 100 according to an embodiment. Below, an example of the overall configuration of the information processing system 100, processing by the information processing system 100, and problems with the information processing system of the reference technology will be described in this order, and finally, the effects of the information processing system 100 will be described. Note that in the embodiment, production monitoring using plant equipment, which is a device installed in a plant, will be described as an example, but the location and field of use are not limited, and the information processing system can also be applied to environmental monitoring and power monitoring that detect signs of system abnormalities.

[0014] (1-1. Example of the overall configuration of the information processing system 100) The information processing system 100 includes an information processing device 10, which is an information processing device, and a sensor device 20, which is a measuring device. The information processing device 10 and the sensor device 20 are connected to each other so that they can communicate with each other via a predetermined communication network. The information processing device 10 is used by a user who manages plant equipment such as the sensor device 20. The sensor device 20 is installed in the plant. The information processing system 100 shown in FIG. 1 may include multiple information processing devices 10.

[0015] In the information processing system 100, data input to the information processing device 10 involves real-time measurement data IP collected from sensor devices 20. In addition, in the information processing system 100, data output by the information processing device 10 involves a health score OP indicating the normal state of the system.

[0016] (1-2. Overall processing of the information processing system 100) In the information processing system 100, first, the information processing device 10 executes a learning process to train the calculation model 14c from the acquired past measurement data DA (see FIG. 1(1)). Next, the user of the information processing device 10 executes an input process of the measurement data IP of the sensor device 20 (see FIG. 1(2)). Finally, the information processing device 10 executes an output process of the health score OP (see FIG. 1(3)). Below, the processing of the information processing system 100 will be explained in the order of the learning process, input process, and output process.

[0017] (1-2-1. Learning process) 1(1), the information processing device 10 executes a learning process. The learning process will be described below in the order of data specification process, data synthesis process, and model learning process.

[0018] (1-2-1-1. Data specification processing) 1(1-1), the information processing device 10 specifies data for a plurality of periods. For example, the information processing device 10 specifies data groups DA-1 and DA-2, which are normal data included in different periods, from past measurement data DA acquired from the sensor device 20.

[0019] (1-2-1-2. Data synthesis processing) 1(1-2), the information processing device 10 synthesizes data from multiple periods. For example, the information processing device 10 generates a data group DB, which is synthesized data corresponding to one period, by linking a data group DA-1 and a data group DA-2 included in different specified periods.

[0020] (1-2-1-3. Model learning process) 1(1-3), the information processing device 10 performs learning of the calculation model 14c using the synthetic data. For example, the information processing device 10 inputs a data group DB, which is the generated synthetic data, to the calculation model 14c, thereby training the calculation model 14c that outputs a health score OP indicating a normal state from the measurement data IP.

[0021] (1-2-2. Input processing) 1(2), the information processing device 10 receives input of measurement data IP, which is real-time data from a sensor device 20, which is a plant device. For example, the information processing device 10 receives input of temperature data, acceleration data, and velocity data, each of which has a time stamp, as the measurement data IP transmitted by the sensor device 20.

[0022] (1-2-3. Output processing) 1(3), the information processing device 10 outputs a health score OP. For example, the information processing device 10 outputs a health score OP indicating the normal state of the plant in which the sensor device 20 is installed, with the health score OP being a positive score if the plant is in a normal state and a negative score if the plant is in an abnormal state.

[0023] (1-3. Measurement data used for learning process) Here, the measurement data used in the learning process will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of measurement data D according to the embodiment. As shown in Fig. 2, the measurement data D used in the learning process is a data group to which a timestamp is added, and includes a plurality of different sensor values ​​(e.g., temperature data, acceleration data, speed data).

[0024] 2(1) is a waveform graph showing a data group including multiple different sensor values ​​collected over consecutive periods, including a data group that is in a normal state in the first half of the period and in an abnormal state in the second half of the period. In this case, the data group marked "normal data" is used as the measurement data D to be used in the learning process, and the data group marked "abnormal data" is not used.

[0025] Figure 2(2) is a waveform graph showing a data group including multiple different sensor values ​​collected over a continuous period, including a data group that is in a normal state throughout the period. In this case, the entire data group marked "normal data" is used as the measurement data D to be used in the learning process.

[0026] Figure 2(3) is a waveform graph showing a data group containing multiple different sensor values ​​collected over consecutive periods. Both the first and second halves of the period contain data groups that are normal, but there are "stepping stone" periods where no sensor values ​​are collected. In this case, the measurement data D used for the learning process is a combination of multiple data groups indicated as "normal data 1" and "normal data 2."

[0027] (1-4. Reference Technology Information Processing System) After explaining the outline of the information processing as the reference technology, the problems of the reference technology will be explained.

[0028] (1-4-1. Overview of information processing in reference technology) The information processing of the reference technology is performed as follows. First, the information processing device of the reference technology displays a graph of the waveform of the sensor value, etc., on the screen, as shown in Figure 2. Second, a user of the information processing device of the reference technology specifies only one data group for the period they wish to use for learning processing on the displayed screen. Third, the information processing device of the reference technology inputs the specified data group into a calculation model and performs model learning. Fourth, the information processing device of the reference technology inputs real-time measurement data collected from sensor devices into a trained calculation model, thereby calculating a health score that indicates the normal state of the plant.

[0029] (1-4-2. Problems with information processing in reference technology) The following describes the problems with information processing in the reference technology. First, the information processing device of the reference technology specifies only one learning period, acquires a group of data for that period, and uses it for model learning. That is, the information processing device of the reference technology can use the "normal data" in FIG. 2(1) and the "normal data" in FIG. 2(2) among the measurement data shown in FIG. 2, but can only use either "normal data 1" or "normal data 2" in FIG. 2(3) for model learning (Problem 1). The example of FIG. 2(3) shows a plant in which the "morning shift" from morning to before noon is when operation has started and is operating normally ("normal data 1"); the "lunch break" is when operation is temporarily stopped; the "afternoon shift" from late afternoon to evening is when operation has resumed and is operating normally ("normal data 2"); and the "end of work" is when operation is stopped from evening to morning. Therefore, the information processing device of the reference technology can only select one period (section), making it difficult to secure a sufficient number of learning data, and making it difficult to effectively utilize the sensor values ​​of plant equipment such as motors that operate intermittently.

[0030] Second, the information processing device of the reference technology does not combine multiple data groups and use them for model learning, making it difficult to combine future predicted values ​​or simulation values ​​for which results have not yet been obtained with actual values ​​up to the present (Problem 2).

[0031] Third, the information processing device of the reference technology does not combine data groups from multiple periods or data groups including predicted values, etc., and use them for model learning, making it difficult to perform the calculation process of the health score OP with high accuracy (Problem 3). In other words, when normal data intervals are "stepping stones," the information processing device of the reference technology has difficulty improving the accuracy of the calculation process even if it uses data groups from multiple intervals as learning data.

[0032] (1-5. Effects of the information processing system 100) In the following, an overview of the information processing system 100 according to the embodiment will be described, and then the effects of the information processing system 100 will be described.

[0033] (1-5-1. Overview of Information Processing System 100) First, the information processing device 10 specifies data for multiple periods. Second, the information processing device 10 combines the data for multiple periods. Third, the information processing device 10 uses the combined data to train the calculation model 14c. Fourth, the information processing device 10 collects measurement data IP. Fifth, the information processing device 10 outputs a health score OP from the measurement data IP using the trained calculation model 14c.

[0034] (1-5-2. Effects of the information processing system 100) The above-described information processing system 100 has the following advantages. First, the information processing system 100 allows multiple periods to be selected via a UI (User Interface), and enables learning to be performed by arbitrarily selecting multiple "stepping stone" normal data intervals, thereby ensuring a satisfactory number of learning data. Therefore, the information processing system 100 contributes to solving the above-described problem 1, which is that it is difficult to effectively utilize sensor values ​​from devices that operate intermittently.

[0035] Second, in the information processing system 100, the multiple data groups do not need to be composed only of actual values ​​up to the present, whose results have been confirmed, but can also include predicted values ​​obtained by simulating future values, making it applicable to predictions such as "If the future values ​​are like this, then we should operate like this from now into the future." Therefore, the information processing system 100 contributes to solving the above-mentioned problem 2, which is that it is difficult to select and learn multiple data groups including future predicted values.

[0036] Third, the information processing system 100 can improve the accuracy of AI analysis by internally linking multiple data groups and learning them as a single data group, thereby obtaining better AI (Artificial Intelligence) learning results. Therefore, the information processing system 100 contributes to solving the above-mentioned problem 3, that is, it is difficult to improve prediction accuracy with multiple data groups in normal data sections that become "stepping stones."

[0037] As described above, the information processing system 100 can effectively detect abnormal states in the system by ensuring a sufficient amount of learning data, enabling the use of data groups including future predicted values, and improving the accuracy of AI analysis.

[0038] 2. Configuration of Information Processing Device 10 3 to 10, the functional configuration of the information processing device 10 included in the information processing system 100 shown in Fig. 1 will be described in detail below. An example of the configuration of the information processing device 10 according to the embodiment, a specific example of a display screen of the information processing device 10, and data designation processing of the information processing device 10 will be described in this order.

[0039] (2-1. Configuration Example of Information Processing Device 10) An example of the configuration of the information processing device 10 according to the embodiment will be described with reference to Fig. 3. Fig. 3 is a block diagram showing an example of the configuration of the information processing device 10 according to the embodiment. The information processing device 10 has an input unit 11, an output unit 12, a communication unit 13, a storage unit 14, and a control unit 15.

[0040] (2-1-1. Input section 11) The input unit 11 is realized by, for example, a keyboard, a mouse, etc. The input unit 11 accepts various operations from a user of the information processing device 10. For example, the input unit 11 accepts an input from the user of the information processing device 10 specifying measurement data D to be used in the learning process.

[0041] (2-1-2. Output section 12) Output unit 12 is realized by, for example, a liquid crystal display, etc. Output unit 12 displays various information. For example, output unit 12 displays health score OP calculated by control unit 15 of information processing device 10.

[0042] (2-1-3. Communications Department 13) The communication unit 13 is realized by, for example, a network interface card (NIC), etc. The communication unit 13 is connected to a predetermined communication network by wire or wirelessly, and transmits and receives information to and from various devices.

[0043] (2-1-4. Storage section 14) The storage unit 14 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. As shown in Fig. 3, the storage unit 14 according to the embodiment has a designated data storage unit 14a, a combined data storage unit 14b, and a calculation model 14c. The storage unit 14 stores various types of information referenced when the control unit 15 operates and various types of information acquired when the control unit 15 operates.

[0044] (2-1-4-1. Designated data storage unit 14a) The designated data storage unit 14a stores a group of measurement data DA acquired by the acquisition unit 15a of the control unit 15, the data group being included in a section designated by the user of the information processing device 10. An example of information stored in the designated data storage unit 14a will now be described with reference to FIG. 4. FIG. 4 is a diagram showing an example of the designated data storage unit 14a of the information processing device 10 according to the embodiment. In the example of FIG. 4, the designated data storage unit 14a has items such as "time zone," "temperature data," "acceleration data," and "speed data."

[0045] "Time period" indicates the time period during which the acquired data group was measured, for example, the start time and end time. "Temperature data" indicates the temperature of the device among the sensor values ​​measured by the sensor device 20, for example, the temperature of a motor expressed in degrees Celsius. "Acceleration data" indicates information about the vibration of the device among the sensor values ​​measured by the sensor device 20, for example, the acceleration of a motor expressed in m / s 2 The "acceleration data" indicates information relating to the vibration of the device among the sensor values ​​measured by the sensor device 20, and is, for example, the speed of a motor expressed in m / s.

[0046] 4 shows an example in which, in a time period indicated as "time period #1," the temperature data is "temperature data #1," the acceleration data is "acceleration data #1," the speed data is "speed data #1," etc., in a time period indicated as "time period #2," the temperature data is "temperature data #2," the acceleration data is "acceleration data #2," the speed data is "speed data #2," etc., and in a time period indicated as "time period #3," the temperature data is "temperature data #3," the acceleration data is "acceleration data #3," the speed data is "speed data #3," etc. In the above example, the "temperature data," "acceleration data," and "speed data" included in each time period include multiple pieces of data with time stamps attached for each measurement time.

[0047] (2-1-4-2. Composite data storage unit 14b) The composite data storage unit 14b stores the composite data DB generated by the generation unit 15b of the control unit 15. An example of information stored in the composite data storage unit 14b will now be described with reference to FIG. 5. FIG. 5 is a diagram showing an example of the composite data storage unit 14b of the information processing device 10 according to the embodiment. In the example of FIG. 5, the composite data storage unit 14b has items such as "time zone," "temperature data," "acceleration data," and "speed data."

[0048] "Time period" indicates the time period in which the synthesized data group is included, for example, the start time and end time. "Temperature data" indicates the temperature of the device among the sensor values ​​measured by the sensor device 20, for example, the temperature of a motor expressed in degrees Celsius. "Acceleration data" indicates information about the vibration of the device among the sensor values ​​measured by the sensor device 20, for example, the acceleration of a motor expressed in m / s 2 The "acceleration data" indicates information relating to the vibration of the device among the sensor values ​​measured by the sensor device 20, and is, for example, the speed of a motor expressed in m / s.

[0049] 5 shows an example in which the time period indicated by the synthesized "time period #1 to #3" includes temperature data "temperature data #1 to #3," acceleration data "acceleration data #1 to #3," and speed data "speed data #1 to #3." Note that in the above example, the "temperature data," "acceleration data," and "speed data" included in each time period include multiple pieces of data with labels assigned corresponding to the timestamps for each measurement time.

[0050] (2-1-4-3. Calculation model 14c) The calculation model 14c is a machine learning model trained by a training unit 15c of the control unit 15. For example, the calculation model 14c is a trained model trained by machine learning using real-time data such as temperature data, acceleration data, and speed data transmitted by the sensor device 20 as explanatory variables and a health score OP (normal state: positive score, abnormal state: negative score) indicating the normal state of the plant in which the sensor device 20 is installed as a response variable. For example, the calculation model 14c is a machine learning model trained using classification based on the Random Forest technology and a binary classification method such as a Boost Decision Tree, which is a recursive partitioning method, as the classification method.

[0051] (2-1-5. Control unit 15) The control unit 15 is responsible for overall control of the information processing device 10. The control unit 15 has an acquisition unit 15a, a generation unit 15b, a training unit 15c, a collection unit 15d, and a calculation unit 15e. Here, the control unit 15 can be realized by, for example, an electronic circuit such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit), or an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0052] (2-1-5-1. Acquisition part 15a) The acquiring unit 15a acquires normal data included in a plurality of sections from the measurement data measured by the device for each section, and stores the acquired measurement data in the specified data storage unit 14a.

[0053] The following describes the types of data to be acquired as processing by the acquisition unit 15a. The acquisition unit 15a acquires, for each section, normal data included in multiple sections from measurement data collected by plant devices installed in the plant.

[0054] To explain a specific example, the acquisition unit 15a acquires temperature data, acceleration data, and speed data for each time period as data measured by the sensor device 20, such as {time period #1: temperature data #1, acceleration data #1, speed data #1}, {time period #2: temperature data #2, acceleration data #2, speed data #2}, {time period #3: temperature data #3, acceleration data #3, speed data #3}, ...

[0055] The following describes the process of acquiring data specified using the data display screen as a process of the acquiring unit 15a. The acquiring unit 15a outputs a data display screen that displays measurement data using a predetermined graph, and acquires normal data included in multiple intervals specified by the user on the output data display screen for each interval.

[0056] For example, the acquiring unit 15a displays a plurality of section candidates on the data display screen and acquires each normal data included in a section selected from the plurality of section candidates. As a specific example, as will be described later in (2-3. Specific example of designation processing) (2-3-2. Designation processing 2), the acquiring unit 15a acquires a data group included in a section designated by a user by clicking from a plurality of section candidates on a data display screen that displays waveform graphs of temperature data, acceleration data, and speed data measured by the sensor device 20.

[0057] Furthermore, the acquiring unit 15a displays sections containing measurement data exceeding a predetermined threshold as multiple section candidates, and acquires normal data contained in a section selected from the multiple section candidates. To explain a specific example, as will be described later in (2-3. Specific example of designation process) (2-3-3. Designation process 3), the acquiring unit 15a acquires a data group contained in a section designated by the user by checking a checkbox from multiple section candidates that exceed the threshold 10.0 on a data display screen that displays waveform graphs of temperature data, acceleration data, and velocity data measured by the sensor device 20.

[0058] The following describes the process of acquiring data specified using the data display screen as a process of the acquiring unit 15a. The acquiring unit 15a outputs a data designation screen that displays an input or selectable section for acquiring normal data from among the measurement data, and acquires normal data included in multiple sections input or selected by the user on the output data designation screen for each of the sections.

[0059] To explain a specific example, as will be described later in (2-3. Specific example of designation process) (2-3-1. Designation process 1), the acquisition unit 15a acquires a data group included in a section that is designated by the user clicking a start point and an end point on a data display screen that displays waveform graphs of temperature data, acceleration data, and speed data measured by the sensor device 20, and that is confirmed by the user clicking an "OK" button on the data designation screen. At this time, the acquisition unit 15a can also acquire a data group included in a section designated by the user directly inputting a "designated time" on the data designation screen, without using the data display screen.

[0060] The following describes the process of acquiring data determined to be normal or abnormal as a process performed by the acquiring unit 15a. The acquiring unit 15a acquires, as normal data, each piece of measurement data included in a section in which all of the measurement data is determined to be normal data, from among multiple different types of correlated measurement data measured by devices.

[0061] To give a specific example, when all the values ​​of temperature data, acceleration data, speed data, etc. measured by the sensor device 20, which are known to have a correlation with each other or are estimated to have a correlation with each other, are within the range of the upper and lower thresholds set for each data type, the acquisition unit 15a acquires the measurement data included in the section within that range as a normal data group.

[0062] The following describes the process of the acquiring unit 15a to acquire not only measurement data measured by a device but also predicted data including predicted values. The acquiring unit 15a further acquires predicted normal data, which is normal data, from the predicted data including predicted values ​​of the measurement data.

[0063] To give a specific example, the acquisition unit 15a acquires, as a normal data group, numerical values ​​predicted by extrapolating graphs of temperature data, acceleration data, and speed data measured by the sensor device 20, or numerical values ​​simulated by statistically processing past data, which are within the range of upper and lower thresholds set for each data type.

[0064] (2-1-5-2. Generation unit 15b) The generating unit 15b combines the acquired normal data to generate combined data. The acquiring unit 15a stores the acquired measurement data in the designated data storage unit 14a.

[0065] The following describes the types of data generated by the generation unit 15b as part of the processing performed by the generation unit 15b. The generation unit 15b generates synthesized data by synthesizing the normal data collected by the plant devices included in multiple sections.

[0066] To explain a specific example, the generation unit 15b combines data groups measured by the sensor device 20 for three time periods {Time period #1: temperature data #1, acceleration data #1, speed data #1}, {Time period #2: temperature data #2, acceleration data #2, speed data #2}, and {Time period #3: temperature data #3, acceleration data #3, speed data #3}, to generate combined data {Time periods #1 to #3: temperature data #1 to #3, acceleration data #1 to #3, speed data #1 to #3}.

[0067] At this time, the generating unit 15b can also re-add a timestamp to the data group. For example, the generation unit 15b can generate synthesized data {time T001: data 005, time T002: data 006, time T003: data 007, time T004: data 009, time T005: data 010, time T006: data 011, time T007: data 007}, data group 2 {time T009: data 009, time T010: ​​data 010, time T011: data 011}, and data group 3 {time T015: data 015, time T016: data 016, time T017: data 017} by re-adding timestamps to the synthesized data {time T001: data 005, time T002: data 006, time T003: data 007, time T004: data 009, time T005: data 010, time T006: data 011, time T007: data 015, time T008: data 016, time T009: data 017}.

[0068] The following describes the process of generating predicted normal data including predicted values ​​as the process of the generating unit 15b. The generating unit 15b combines each normal data and the predicted normal data to generate predicted combined data.

[0069] To explain a specific example, the generation unit 15b combines data groups for three time periods measured by the sensor device 20, {Time period #1: temperature data #1, acceleration data #1, speed data #1}, {Time period #2: temperature data #2, acceleration data #2, speed data #2}, and {Time period #3: temperature data #3, acceleration data #3, speed data #3}, with one data group {Time period #4: temperature data #4, acceleration data #4, speed data #4} that includes only normal data (predicted normal data) predicted to be measured by the sensor device 20, to generate predicted combined data {Time periods #1 to #4: temperature data #1 to #4, acceleration data #1 to #4, speed data #1 to #4}. At this time, the generation unit 15b can also combine only the predicted normal data to generate predicted combined data.

[0070] (2-1-5-3. Training Department 15c) The training unit 15c trains a machine learning model that outputs a score indicating the normal state of a system in which a device is installed in response to input of measurement data, by machine learning using the synthetic data. The training unit 15c trains the calculation model 14c stored in the storage unit 14.

[0071] The following describes the types of data used for training as processing by the training unit 15c. The training unit 15c trains a machine learning model that outputs a score indicating the normal state of a plant in which plant equipment is installed, in response to input of measurement data, by machine learning using synthetic data.

[0072] To explain a specific example, the training unit 15c trains the calculation model 14c by inputting the synthetic data {time periods #1 to #3: temperature data #1 to #3, acceleration data #1 to #3, speed data #1 to #3} generated by the generation unit 15b.

[0073] The following describes a training process using predicted normal data including predicted values ​​as the process of the training unit 15c. The training unit 15c trains a machine learning model by machine learning using predicted synthetic data.

[0074] To explain a specific example, the calculation model 14c is trained by inputting the predicted synthetic data {time periods #1 to #4: temperature data #1 to #4, acceleration data #1 to #4, speed data #1 to #4} generated by the generation unit 15b.

[0075] (2-1-5-4. Collection section 15d) The collection unit 15d collects the measurement data transmitted by the device. The collection unit 15d may store the collected measurement data in the storage unit .

[0076] To explain the types of data to be collected, the collection unit 15d collects temperature data, acceleration data, and speed data of the motor transmitted by the sensor devices 20 installed in the plant.

[0077] To give a specific example, the real-time data transmitted by the sensor device 20 includes sensor values ​​of each type with a time stamp attached, such as {time T101: temperature data 101, acceleration data 101, speed data 101}.

[0078] (2-1-5-5. Calculation unit 15e) The calculation unit 15e calculates a score indicating the normal state of the system in which the device is installed, based on the result obtained by inputting the collected measurement data into a trained machine learning model. Note that the calculation unit 15e may store the calculated score in the storage unit 14.

[0079] To explain the types of scores to be calculated, the calculation unit 15e outputs a positive score if the plant is in a normal state and a negative score if the plant is in an abnormal state as the health score OP indicating the normal state of the plant in which the sensor device 20 is installed. In this case, the calculation unit 15e outputs a score as the health score OP indicating that the larger the absolute value of the positive score is, the more normal the state is, and that the larger the absolute value of the negative score is, the more abnormal the state is.

[0080] To explain a specific example, the calculation unit 15e calculates the score at which the health score OP indicates an abnormal value for each time (timestamp), such as {Time T101: health score "+1.0"}, {Time T102: health score "-1.0"}, {Time T103: health score "-2.0"}, etc.

[0081] (2-2. Specific examples of display screens) 6 and 7, specific examples of display screens displayed by the information processing device 10 will be described. Below, the data display screen and the data designation screen will be described in that order.

[0082] (2-2-1. Data display screen) A data display screen that displays measurement data D using a predetermined graph will be described with reference to FIG. 6. FIG. 6 is a diagram showing an example of a display screen for measurement data D according to an embodiment. As shown in the example of FIG. 6, the information processing device 10 displays a waveform graph showing the intensity per time of a plurality of pieces of measurement data D. For example, the information processing device 10 can display data in different colors to make it easier to distinguish the types of data, such as coloring the temperature data (see FIG. 6(1)) transmitted by the sensor device 20 in green, the acceleration data (see FIG. 6(2)) in red, and the speed data (see FIG. 6(3)) in blue. In addition, the information processing device 10 can change the units of the vertical axis (°C for temperature data, m / s for acceleration data) so that the waveforms of each data type do not overlap. 2 The scale of the velocity data (m / s) can also be changed.

[0083] (2-2-2. Data specification screen) FIG. 7 illustrates a data specification screen that displays a section for acquiring normal data from the measurement data D and allows the user to input or select the section. FIG. 7 illustrates an example of a specification screen for measurement data D according to an embodiment. As illustrated in the example of FIG. 7, the information processing device 10 displays a screen on which a user selects the "study period" as a "specified time" and inputs "2020 / 05 / 22 8:30" as the "start time" and "2020 / 05 / 22 11:45" as the "end time" for the "anomaly detection learning period" that specifies the measurement data D to be used as learning data. The user can specify the "specified time" shown in FIG. 7 by specifying the start and end points of the waveform graph on the data display screen shown in FIG. 6. Furthermore, when the user clicks the "add period" button, the period specified as "period 1" is displayed. Similarly, when the user specifies a different period and clicks the "add period" button, the periods specified as "period 2" and "period 3" are displayed. The user then clicks the "OK" button to confirm the specification of the periods "period 1" to "period 3."

[0084] (2-3. Specific examples of designated processing) 8 to 10, a specific example of the designation process of the measurement data D of the information processing device 10 will be described. The following describes designation process 1, which uses the data display screen and the data designation screen, designation process 2, which uses the data display screen to display candidate periods for continuous data, and designation process 3, which uses the data display screen to display candidate periods for "stepping stone" data, in that order.

[0085] (2-3-1. Designation process 1) Using FIG. 8, a description will be given of a designation process 1 for measurement data D using a data display screen and a data designation screen. FIG. 8 is a diagram showing the designation process 1 for measurement data D according to an embodiment. As shown in FIG. 8(1), a user of the information processing device 10 designates a data group DA-1 by designating the start and end points of a waveform graph on the data display screen. At this time, when the user clicks the "Add Period" button, the period designated as "Period 1" is displayed as "2020 / 05 / 22 8:30 to 11:45." As shown in FIG. 8(2), a user of the information processing device 10 designates a data group DA-2 by designating the start and end points of a waveform graph on the data display screen. At this time, when the user clicks the "Add Period" button, the period designated as "Period 2" is displayed as "2020 / 05 / 22 12:30 to 17:15." As shown in Figure 8 (3), by clicking the "OK" button, the user confirms the specification of the period "2020 / 05 / 22 8:30~11:45" for "Period 1" and the period "2020 / 05 / 22 12:30~17:15" for "Period 2."

[0086] In the case of the above-mentioned Designation Process 1, if only one of the "stepping stone" periods is trained, there may be cases where the amount of data is insufficient for sufficient learning, and if only one data group is trained, the results of subsequent AI analysis may not be satisfactory. Here, in the above-mentioned Designation Process 1, a two-stage "stepping stone" pattern was used, but in reality, if the device operates intermittently and irregularly, there will be many more stepping stones. In the above-mentioned Designation Process 1, multiple periods can be selected up to three periods, but the number of selectable periods can be increased, allowing for flexible selection even with an infinite number of "stepping stones." As described above, the above-mentioned Designation Process 1 makes it possible to perform sufficient AI learning even from data where the normal data sections are "stepping stones," which can contribute to increasing the number of operational patterns in which AI analysis can be implemented.

[0087] Note that while the above-mentioned specification process 1 targets "past performance data recorded by the recorder," the target data is not limited to past data; for example, if the viewer had a function that "can display text data written in a certain format in the same way as recorder data files," the user could simulate future predicted values ​​and display continuous data combined with past data, making it possible to "select multiple data groups including future predicted values ​​and execute learning." Therefore, the above-mentioned specification process 1 can also be expected to be used in ways that proactively simulate abnormalities and enable analysis and countermeasures, such as, "When considered in conjunction with the current operating status, at what point in the future will the value of this channel need to exceed a threshold to cause an abnormal state?"

[0088] (2-3-2. Designation process 2) Using FIG. 9, a description will be given of designation process 2 for displaying candidate periods of continuous data using the data display screen. FIG. 9 is a diagram showing designation process 2 for measurement data D according to an embodiment. As shown in FIGS. 9(1) to (3), the information processing device 10 displays a plurality of candidate intervals (shaded areas) on the data display screen. At this time, the user clicks on the area of ​​the candidate interval they wish to designate from among the candidate intervals in FIGS. 9(1) to (3), thereby finalizing the designation of the interval. The user can also change the start point (start time) and end point (end time) by dragging the area of ​​the displayed candidate.

[0089] That is, in the above-mentioned designation process 2, the selection UI can be set not as a dialog box (data designation screen) but so that multiple selections can be made simultaneously at the time of the selection operation on the waveform display. On the other hand, in the above-mentioned designation process 2, when the selection UI is displayed as a dialog box, the multiple selection period may be preset and already selected.

[0090] (2-3-3. Designation process 3) Using FIG. 10, a description will be given of designation process 3 for displaying candidate periods for "stepping stone" data using a data display screen. FIG. 10 is a diagram showing designation process 3 for measurement data D according to an embodiment. As shown in FIG. 10(1), the information processing device 10 displays multiple candidate intervals (shaded areas) on the data display screen. At this time, the information processing device 10 displays multiple candidate intervals (shaded areas) on the data display screen that exceed a predetermined threshold (10.0). Then, as shown in FIG. 10(2), the user checks the checkbox of the candidate they wish to designate from among the displayed candidate intervals, thereby finalizing the designation of the interval (candidate 1, candidate 3, candidate 4, candidate 5, candidate 6, candidate 8).

[0091] That is, in the above-mentioned designation process 3, since manual selection by the user becomes cumbersome when there are many items to select, the "stepping stone" part can be automatically searched and listed as study period candidates, and the user can select any period from that list. For example, in the above-mentioned designation process 3, when determining whether or not a period is a "stepping stone" period, it is possible to accept a condition setting from the user in advance, such as "if it is below the lower limit of 10.0, it is considered a stepping stone down period and other periods are listed."

[0092] 3. Processing flow of information processing system 100 The processing flow of the information processing system 100 according to the embodiment will be described with reference to Fig. 11. Fig. 11 is a sequence diagram showing an example of the overall flow of information processing according to the embodiment. Note that the processing of steps S101 to S106 below can also be executed in a different order. Also, some of the processing of steps S101 to S106 below may be omitted. Below, the processing flow of the information processing system 100 will be described in the order of learning processing and input / output processing.

[0093] (3-1. Learning process) First, the information processing device 10 executes a data acquisition process (step S101). For example, the information processing device 10 acquires past measurement data DA measured by the sensor device 20.

[0094] Second, the information processing device 10 executes a data designation process (step S102). For example, the information processing device 10 designates data groups DA-1 and DA-2, which are normal data included in different time periods, from the past measurement data DA.

[0095] Third, the information processing device 10 executes a data synthesis process (step S103). For example, the information processing device 10 generates a data group DB, which is synthetic data corresponding to one period, by linking a data group DA-1 and a data group DA-2 included in different specified periods.

[0096] Fourth, the information processing device 10 executes a model learning process (step S104). For example, the information processing device 10 inputs the data group DB, which is the generated composite data, to the calculation model 14c, thereby training the calculation model 14c that outputs a health score OP indicating a normal state from the measurement data IP.

[0097] (3-2. Input / Output Processing) First, the information processing device 10 executes a data collection process (step S105). For example, the information processing device 10 receives input of time-stamped temperature data, acceleration data, and speed data as measurement data IP transmitted by the sensor device 20.

[0098] Second, the information processing device 10 executes a score calculation process (step S106). For example, the information processing device 10 outputs a health score OP indicating the normal state of the plant in which the sensor device 20 is installed, with a positive score if the state is normal and a negative score if the state is abnormal.

[0099] 4. Effects of the embodiment Finally, the effects of the embodiment will be described below: Effects 1 to 9 corresponding to the processing according to the embodiment will be described below.

[0100] (4-1. Effect 1) First, in the process according to the embodiment described above, the information processing device 10 acquires normal data included in multiple intervals from among measurement data measured by the device, synthesizes the acquired normal data for each interval, generates synthetic data, and trains a machine learning model using the synthetic data to output a score indicating the normal state of the system in which the device is installed in response to input of measurement data. Therefore, this process can effectively detect an abnormal state of the system.

[0101] (4-2. Effect 2) Second, in the process according to the embodiment described above, the information processing device 10 outputs a data display screen that displays measurement data using a predetermined graph, and acquires normal data included in multiple intervals specified by the user on the output data display screen for each interval. Therefore, in this process, the measurement data specified on the data display screen is used in the learning process, thereby making it possible to effectively detect abnormal states of the system.

[0102] (4-3. Effect 3) Third, in the process according to the embodiment described above, the information processing device 10 displays multiple section candidates on the data display screen and acquires normal data included in a section selected from the multiple section candidates. Therefore, in this process, the measurement data specified from the candidates displayed on the data display screen is used for the learning process, thereby making it possible to effectively detect an abnormal state of the system.

[0103] (4-4. Effect 4) Fourth, in the process according to the embodiment described above, the information processing device 10 displays sections containing measurement data exceeding a predetermined threshold as multiple section candidates, and acquires normal data contained in a section selected from the multiple section candidates. Therefore, in this process, the measurement data specified from the candidates displayed on the data display screen based on the threshold is used for the learning process, thereby making it possible to effectively detect an abnormal state of the system.

[0104] (4-5. Effect 5) Fifth, in the process according to the embodiment described above, the information processing device 10 outputs a data specification screen that displays an input or selectable section for acquiring normal data from among the measurement data, and acquires normal data included in the multiple sections input or selected by the user on the output data specification screen for each section. Therefore, in this process, the measurement data specified on the data specification screen is used in the learning process, thereby making it possible to effectively detect abnormal states of the system.

[0105] (4-6. Effect 6) Sixth, in the process according to the embodiment described above, the information processing device 10 acquires, as normal data, each piece of measurement data included in a section in which all of the measurement data is determined to be normal, from among multiple different types of correlated measurement data measured by devices. Therefore, in this process, multiple different types of measurement data are efficiently determined to be normal data and used in the learning process, thereby making it possible to effectively detect an abnormal state of the system.

[0106] (4-7. Effect 7) Seventh, in the process according to the embodiment described above, the information processing device 10 further acquires predicted normal data, which is normal data, from predicted data including predicted values ​​of measurement data, combines each normal data with the predicted normal data to generate predicted combined data, and trains a machine learning model by machine learning using the predicted combined data. Therefore, in this process, by using not only past measurement data but also future measurement data in the learning process, it is possible to effectively detect an abnormal state of the system.

[0107] (4-8. Effect 8) Eighth, in the process according to the above-described embodiment, the information processing device 10 collects measurement data transmitted by the device, inputs the collected measurement data into a trained machine learning model, and calculates a score indicating the normal state of the system in which the device is installed based on the results. Therefore, in this process, the normal or abnormal state of the system is appropriately visualized using real-time data, thereby making it possible to effectively detect an abnormal state of the system.

[0108] (4-9. Effect 9) Ninth, in the processing according to the above-described embodiment, the information processing device 10 acquires normal data included in multiple intervals from among the measurement data collected by plant equipment installed in the plant, synthesizes the normal data collected by the plant equipment included in the multiple intervals to generate synthetic data, and trains a machine learning model using the synthetic data to output a score indicating the normal state of the plant in which the plant equipment is installed in response to input of the measurement data. Therefore, in this processing, the measurement data collected from the plant equipment in the plant is used in the learning process, thereby making it possible to effectively detect abnormal states of the plant.

[0109] 〔system〕 The information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified.

[0110] Furthermore, the components of each device shown in the figure are functional concepts and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown. In other words, all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0111] Furthermore, all or any part of the processing functions performed by each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware using wired logic.

[0112] [Hardware] Next, an example of the hardware configuration of the information processing device 10 will be described. Note that other devices such as the information processing device 10 may also have a similar hardware configuration. FIG. 12 is a diagram illustrating an example of the hardware configuration. As shown in FIG. 12, the information processing device 10 has a communication device 10a, an HDD (Hard Disk Drive) 10b, a memory 10c, and a processor 10d. Furthermore, the components shown in FIG. 12 are connected to each other via a bus or the like.

[0113] The communication device 10a is a network interface card or the like, and communicates with other servers. The HDD 10b stores programs and DBs that operate the functions shown in FIG.

[0114] The processor 10d reads out a program that executes the same processes as the respective processing units shown in FIG. 3 from the HDD 10b or the like and loads it into the memory 10c, thereby operating a process that executes each function described in FIG. 3 or the like. For example, this process executes the same functions as the respective processing units of the information processing device 10. Specifically, the processor 10d reads out a program that has the same functions as the acquisition unit 15a, the generation unit 15b, the training unit 15c, the collection unit 15d, the calculation unit 15e, etc. from the HDD 10b or the like. Then, the processor 10d executes a process that executes the same processes as the acquisition unit 15a, the generation unit 15b, the training unit 15c, the collection unit 15d, the calculation unit 15e, etc.

[0115] In this way, the information processing device 10 operates as a device that executes various processing methods by reading and executing a program. The information processing device 10 can also realize functions similar to those of the above-described embodiment by reading the program from a recording medium using a medium reading device and executing the read program. Note that the program in these other embodiments is not limited to being executed by the information processing device 10. For example, the present invention can also be applied in the same way to cases where another computer or server executes a program, or where these execute a program in cooperation with each other.

[0116] This program can be distributed via a network such as the Internet. In addition, this program can be recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, a magneto-optical disk (MO), or a digital versatile disk (DVD), and can be executed by being read from the recording medium by a computer.

[0117] 〔others〕 Some examples of combinations of the disclosed technical features are set out below.

[0118] (1) An information processing device comprising: an acquisition unit that acquires, for each interval, normal data included in multiple intervals from measurement data measured by a device; a generation unit that synthesizes the acquired normal data to generate synthetic data; and a training unit that trains, through machine learning using the synthetic data, a machine learning model that outputs a score indicating the normal state of a system in which the device is installed in response to input of the measurement data.

[0119] (2) The information processing device described in (1), wherein the acquisition unit outputs a data display screen that displays the measurement data using a predetermined graph, and acquires each of the normal data included in the multiple intervals specified by the user on the output data display screen for each of the intervals.

[0120] (3) The information processing device described in (2), wherein the acquisition unit displays a plurality of section candidates on the data display screen and acquires each of the normal data included in the section selected from the plurality of section candidates.

[0121] (4) The information processing device described in (3), wherein the acquisition unit displays sections containing the measurement data exceeding a predetermined threshold as candidates for the plurality of sections, and acquires each of the normal data contained in the section selected from the plurality of candidate sections.

[0122] (5) An information processing device described in any one of (1) to (4), wherein the acquisition unit outputs a data designation screen that displays the section in which each of the normal data among the measurement data is to be acquired in an input or selectable manner, and acquires each of the normal data included in the multiple sections input or selected by the user on the output data designation screen for each of the sections.

[0123] (6) An information processing device described in any one of (1) to (5), wherein the acquisition unit acquires, as the normal data, each of the measurement data included in a section in which all of the measurement data are determined to be normal data, from among each of a plurality of different types of correlated measurement data measured by a device.

[0124] (7) The information processing device described in any one of (1) to (6), wherein the acquisition unit further acquires predicted normal data, which is normal data from predicted data including predicted values ​​of the measurement data, the generation unit combines each of the normal data with the predicted normal data to generate predicted synthesized data, and the training unit trains the machine learning model by machine learning using the predicted synthesized data.

[0125] (8) The information processing device described in any one of (1) to (7), further comprising: a collection unit that collects the measurement data transmitted by the device; and a calculation unit that calculates a score indicating the normal state of the system in which the device is installed based on the results obtained by inputting the collected measurement data into the trained machine learning model.

[0126] (9) The acquisition unit acquires, for each of the sections, the normal data included in the plurality of sections from the measurement data collected by plant equipment installed in the plant, and the generation unit The information processing device according to any one of (1) to (8), wherein the normal data collected by the plant equipment included in the plurality of sections is synthesized to generate the synthesized data, and the training unit trains the machine learning model by machine learning using the synthesized data, which outputs a score indicating the normal state of the plant in which the plant equipment is installed in response to input of the measurement data.

[0127] (10) An information processing method in which a computer executes a process in which it acquires normal data included in multiple intervals from measurement data measured by a device, synthesizes the acquired normal data to generate synthetic data, and trains a machine learning model using the synthetic data to output a score indicating the normal state of a system in which the device is installed in response to input of the measurement data.

[0128] (11) An information processing program that executes a process on a computer to acquire normal data included in multiple intervals from measurement data measured by a device, synthesize the acquired normal data to generate synthetic data, and train a machine learning model using the synthetic data to output a score indicating the normal state of a system in which the device is installed in response to input of the measurement data. [Explanation of symbols]

[0129] 10. Information processing equipment 11 Input section 12 Output section 13 Communications Department 14 Storage section 14a Designated data storage section 14b Synthetic data storage unit 14c calculation model 15 Control Unit 15a Acquisition part 15b Generator 15c Training Department 15d Collection Department 15e Calculation section 20 Sensor Equipment 100 Information Processing Systems

Claims

1. an acquisition unit that acquires normal data included in a plurality of sections from measurement data measured by a device, for each of the sections; a generating unit that combines the acquired normal data to generate combined data; a training unit that trains a machine learning model by machine learning using the synthetic data, and outputs a score indicating a normal state of a system in which the device is installed in response to input of the measurement data; and Equipped with The acquisition unit outputting a data display screen that displays, using a predetermined graph, the measurement data including continuous data in which recorded past data and simulated future predicted values ​​are combined, and outputting a data designation screen that displays the intervals in the measurement data from which each normal data is to be acquired so that the intervals can be input or selected, and acquiring each normal data included in the plurality of intervals designated by a user on the output data display screen or the plurality of intervals input or selected by a user on the output data designation screen for each interval; Information processing device.

2. The acquisition unit displaying a plurality of section candidates on the data display screen, and acquiring each of the normal data included in a section selected from the plurality of section candidates; The information processing device according to claim 1 .

3. The acquisition unit displaying sections including the measurement data exceeding a predetermined threshold as candidates for the plurality of sections, and acquiring each of the normal data included in the section selected from the plurality of candidate sections; The information processing device according to claim 2 .

4. The acquisition unit Among a plurality of different types of correlated measurement data measured by a device, each of the measurement data included in a section in which all of the measurement data is determined to be normal data is acquired as the normal data. The information processing device according to claim 1 .

5. The acquisition unit Further acquiring predicted normal data, which is normal data, from predicted data including predicted values ​​of the measurement data; The generation unit synthesizing each normal data and the predicted normal data to generate predicted synthesized data; The training department: training the machine learning model by machine learning using the predicted synthetic data; The information processing device according to claim 1 .

6. a collection unit that collects the measurement data transmitted by the device; a calculation unit that calculates a score indicating the normal state of the system in which the device is installed based on a result obtained by inputting the collected measurement data into the trained machine learning model; and The information processing device according to claim 1 , further comprising:

7. The acquisition unit acquiring, for each of the sections, the normal data included in the plurality of sections from among the measurement data collected by plant equipment installed in the plant; The generation part is synthesizing the normal data collected by the plant devices included in the plurality of sections to generate the synthesized data; The training department: training the machine learning model by machine learning using the synthetic data, the machine learning model outputting a score indicating a normal state of the plant in which the plant device is installed in response to input of the measurement data; The information processing device according to claim 1 .

8. The computer acquiring normal data included in a plurality of sections from the measurement data measured by the device for each of the sections; synthesizing the acquired normal data to generate synthesized data; training a machine learning model by machine learning using the synthetic data, the machine learning model outputting a score indicating the normal state of a system in which the device is installed in response to input of the measurement data; Execute the process, outputting a data display screen that displays, using a predetermined graph, the measurement data including continuous data in which recorded past data and simulated future predicted values ​​are combined, and outputting a data designation screen that displays the intervals in the measurement data from which each normal data is to be acquired so that the intervals can be input or selected, and acquiring each normal data included in the plurality of intervals designated by a user on the output data display screen or the plurality of intervals input or selected by a user on the output data designation screen for each interval; Information processing methods.

9. On the computer, acquiring normal data included in a plurality of sections from the measurement data measured by the device for each of the sections; synthesizing the acquired normal data to generate synthesized data; training a machine learning model by machine learning using the synthetic data, the machine learning model outputting a score indicating the normal state of a system in which the device is installed in response to input of the measurement data; Execute the process, outputting a data display screen that displays, using a predetermined graph, the measurement data including continuous data in which recorded past data and simulated future predicted values ​​are combined, and outputting a data designation screen that displays the intervals in the measurement data from which each normal data is to be acquired so that the intervals can be input or selected, and acquiring each normal data included in the plurality of intervals designated by a user on the output data display screen or the plurality of intervals input or selected by a user on the output data designation screen for each interval; Information processing program.

Images