Anomaly sign detection system, anomaly sign detection model generation method, and anomaly sign detection model generation program

Abstract

To provide an abnormality sign detection technology which suppresses erroneous detection to improve the reliability of an abnormality sign detection function.SOLUTION: A computer 5 is configured to: select a reference process value 31 as reference for calibration; calculate, based on an actual process value 30 and the reference process value 31, a calibration necessity determination coefficient 32 for determining whether or not the actual process value 30 correlates with the reference process value 31; calculate, based on the actual process value 30 and the reference process value 31, a calibration value 34 for calibrating the actual process value 30; determine, based on the calibration necessity determination coefficient 32, whether or not each actual process value 30 correlates with the reference process value 31; calibrate the actual process value 30, which has been determined to correlate with the reference process value 31, with the calibration value 34; generate learning input data including a calibrated process value 35 obtained through the calibration with the calibration value 34; and input the learning input data into an abnormality sign detection model to perform machine learning.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] FIELD An embodiment of the present invention relates to an abnormality sign detection technology. [Background technology]

[0002] In recent years, advances have been made in the practical application of artificial intelligence (AI) to detect signs of abnormalities in plant equipment. There are two main methods for detecting signs of abnormalities.

[0003] For example, there is supervised learning, in which if an abnormal state can be identified, that abnormal state is used as an index (teacher), and if the target for which an abnormality is to be detected contains data that matches the index state, it is detected as a sign of an abnormality.

[0004] Furthermore, when it is not possible to clearly identify an abnormal state in advance due to the wide variety of possible abnormal states, there is unsupervised learning, which does not use an abnormal state as an indicator (teacher) and cannot be explained from the data that the AI ​​has learned. For example, data that differs from the learning data can be detected as a sign of an abnormality.

[0005] In the case of unsupervised learning, if there is insufficient data of normal conditions to train the AI, it will end up identifying conditions that should be treated as normal as abnormal. However, there is generally not a large amount of data to train from. For this reason, a method of enhancing data by modifying existing data, known as data augmentation, has been developed mainly in the field of diagnostic imaging.

[0006] Common data enhancement techniques in the field of diagnostic imaging include adding noise components to image data, rotating or flipping the data, stretching or compressing the data, and converting the frequency components (phase and amplitude) of the image.

[0007] In fields other than diagnostic imaging, a method has been proposed for regenerating data by changing the phase of frequency components in order to compensate for the timing of acquiring discrete time-series data, i.e., the sampling interval. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] International Publication No. 2014 / 091955 [Patent Document 2] International Publication No. 2020 / 178936 Summary of the Invention [Problem to be solved by the invention]

[0009] For example, when applying AI-based anomaly detection technology to data acquired from a plant to detect signs of anomalies in time-series data, there are countless abnormal conditions, and it is not possible to identify all of them in advance. For this reason, unsupervised learning is used.

[0010] However, the normal states that the time-series data used as learning data can take vary depending on various conditions and can take on countless states, so in reality, it is often very difficult to obtain all of the actual data.

[0011] For example, conditions caused by seasonal fluctuations in outside air and seawater temperatures never take the same values. Also, the timing of transient states, such as when plant output is changed in response to these fluctuations, also differs.

[0012] Furthermore, there are cases where the plant is operated with its output power and other parameters changed based on external constraints such as changes in the main operating conditions, for example, power grid requirements.

[0013] Furthermore, there are countless combinations of operation patterns for redundant equipment, for example, but the data associated with each individual piece of equipment may be completely different.

[0014] Furthermore, when a new sensor is installed on an existing facility, new data is added, and the data to be correlated may change before and after the sensor is added.

[0015] Therefore, even if the driving condition is normal, due to insufficient learning, the driving condition may be recognized as a previously unknown driving condition, i.e., abnormal data, and may be unnecessarily detected as a sign of an abnormality. Also, when new data is added, it is not due to a lack of data, but rather to a lack of data itself, and the data cannot be used as a target for abnormality sign detection.

[0016] Data augmentation can be considered as a solution to these issues. However, while conventional techniques such as "adding noise components" and "changing frequency components" to time series data are useful for data augmentation, other methods cannot be utilized. For example, outside air temperature and seawater temperature cannot take on any value, but rather change within a certain range and time period. Therefore, adding noise components or changing frequency components cannot augment normal data.

[0017] Therefore, even when data is augmented using conventional technology, many anomalies will be detected unnecessarily. This places an unnecessary workload on personnel who respond when an anomaly is detected in order to identify the anomaly. Furthermore, if true anomalies are mixed in with the numerous detected anomalies, they may be overlooked or not addressed, reducing the reliability of the anomaly sign detection function. Furthermore, anomaly sign detection cannot be performed on data that is not included in the training data.

[0018] An embodiment of the present invention has been made in consideration of these circumstances, and aims to provide an abnormality sign detection technology that can suppress false detections and improve the reliability of the abnormality sign detection function. [Means for solving the problem]

[0019] an anomaly sign detection system according to an embodiment of the present invention includes one or more computers that perform machine learning of an anomaly sign detection model that detects at least one anomaly or a sign of the anomaly in a monitored facility, wherein the computer is configured to: acquire a plurality of actual process values ​​generated in the facility; select at least one reference process value that serves as a reference for correction from the plurality of actual process values; calculate, from at least one of the actual process values ​​and the reference process value, a correction necessity determination coefficient for determining whether the actual process value is correlated with the reference process value; calculate a correction value for correcting the actual process value from at least one of the actual process value and the reference process value; determine, based on the correction necessity determination coefficient, whether each of the actual process values ​​is correlated with the reference process value; correct at least one of the actual process values ​​that is determined to be correlated with the reference process value, with the correction value; generate learning input data including at least one corrected process value corrected with the correction value; and input the learning input data to the anomaly sign detection model to perform the machine learning. The correction necessity determination coefficient is calculated using a coefficient of determination. . [Effects of the Invention]

[0020] According to an embodiment of the present invention, an abnormality sign detection technique is provided that can reduce false detections and improve the reliability of the abnormality sign detection function. [Brief explanation of the drawings]

[0021] [Figure 1] FIG. 1 is a block diagram showing an abnormality sign detection system according to a first embodiment. [Figure 2] FIG. 2 is a block diagram showing a preprocessing unit according to the first embodiment. [Figure 3] 10 is a graph showing how one corrected process value is calculated from an actual process value. [Figure 4] 10 is a graph showing a manner in which a plurality of corrected process values ​​are calculated from an actual process value. [Figure 5] FIG. 10 is a block diagram showing an abnormality sign detection system according to a second embodiment. [Figure 6]FIG. 10 is a block diagram showing a preprocessing unit according to a second embodiment. [Figure 7] FIG. 10 is a block diagram showing a setting reception unit according to the second embodiment. [Figure 8] FIG. 1 is a configuration diagram showing a system having redundant devices. [Figure 9] 10 is a graph showing a manner in which corrected process values ​​of redundant devices are calculated. [Figure 10] FIG. 10 is a block diagram showing an abnormality sign detection system according to a third embodiment. [Figure 11] FIG. 10 is a block diagram showing a preprocessing unit according to a third embodiment. [Figure 12] 10 is a graph showing a manner in which a corrected process value is calculated from additional data for learning. DETAILED DESCRIPTION OF THE INVENTION

[0022] (First embodiment) Hereinafter, embodiments of an anomaly sign detection system, an anomaly sign detection model generation method, and an anomaly sign detection model generation program will be described in detail with reference to the drawings. First, a first embodiment will be described with reference to FIGS. 1 to 4.

[0023] Reference numeral 1 in Fig. 1 denotes an abnormality sign detection system according to a first embodiment. This abnormality sign detection system 1 uses data obtained from a target facility as input data and detects abnormalities or signs of abnormalities in the target facility. Target facilities to be monitored include, for example, nuclear power plants, thermal power plants, factory facilities, or production facilities. A plant 2 of this kind is exemplified as a target facility in the first embodiment.

[0024] The plant 2 is also equipped with a large number of sensors 3. These sensors 3 are, for example, specified measuring instruments attached to specified equipment such as pipes, pumps, and valves. The sensors 3 also acquire measurement values ​​(actual measurements) that include information indicating the status of these equipment. The large number of measurement values ​​obtained from these sensors 3 are referred to as monitored data. Furthermore, each measurement value converted into a format that can be handled by machine learning is referred to as an actual process value 30 (Figure 2). Note that control signals output from a control device that controls the equipment are also included in the actual process value 30. Valve openings and the like are also included in the actual process value 30.

[0025] In the abnormality sign detection technology, abnormalities or signs thereof are detected by detecting slight changes in the actual process value 30. To do this, it is necessary to determine the normal state of the plant 2 with high accuracy. Incorrect determinations will cause false positives and result in unnecessary work for operators.

[0026] Furthermore, the actual process values ​​30 acquired in the plant 2 are enormous in volume. Therefore, in this embodiment, artificial intelligence (AI) realized by machine learning is used to determine whether an abnormality or a sign thereof exists from the enormous amount of actual process values ​​30.

[0027] For example, it is possible to use a learning model generated by machine learning using a neural network, a learning model generated by other machine learning, a deep learning algorithm, a mathematical algorithm such as regression analysis, etc. Furthermore, forms of machine learning include forms such as clustering and deep learning.

[0028] For example, this abnormality sign detection system 1 may be configured with one computer equipped with a neural network, or may be configured with multiple computers equipped with neural networks.

[0029] Here, a neural network is a mathematical model that represents the characteristics of brain function through computer simulation. For example, it refers to a model in which artificial neurons (nodes) that form a network through synaptic connections change the strength of synaptic connections through learning, thereby acquiring problem-solving ability. Furthermore, neural networks acquire problem-solving ability through deep learning.

[0030] For example, a neural network may have multiple intermediate layers. Each intermediate layer is composed of multiple units. By training the multi-layer neural network in advance using training data (teacher data), it is possible to automatically extract certain features from patterns of change in the state of a circuit or system. The multi-layer neural network can be configured with any number of intermediate layers, any number of units, any learning rate, any number of learning iterations, and any activation function via a user interface.

[0031] In this embodiment, an anomaly sign detection technology using an autoencoder (encoder-decoder network) will be described. The learning model of this embodiment is realized by this autoencoder. Note that algorithms other than the autoencoder may also be applied to the machine learning of this embodiment.

[0032] Furthermore, during machine learning, a false correlation occurs in which the trends of change in time series are similar by chance, even though there is no physical correlation between the multiple actual process values ​​30. When the AI ​​learns this false correlation, it can cause false detection. However, in this embodiment, learning of the false correlation is avoided, and anomalies in the plant 2 can be predicted with high accuracy.

[0033] 1, the anomaly sign detection system 1 includes a data input computer 4, a learning computer 5, and a detection computer 6. These are configured as computers having hardware resources such as a CPU, ROM, RAM, and HDD, and in which software-based information processing is realized using the hardware resources as the CPU executes various programs. Furthermore, the anomaly sign detection model generation method of this embodiment is realized by having the computers execute various programs.

[0034] The data input computer 4 collects monitoring target data, which is a large number of measurement values ​​acquired by sensors 3 installed in the plant 2. This data input computer 4 is, for example, a server for storing the monitoring target data. The collected monitoring target data is sent to the learning computer 5 or the detection computer 6.

[0035] The learning computer 5 generates an abnormality sign detection model that detects at least one of an abnormality or a sign thereof in the plant 2. The generated abnormality sign detection model is sent to the detection computer 6.

[0036] The detection computer 6 detects at least one of an abnormality or a sign of an abnormality in the plant 2 using the abnormality sign detection model.

[0037] Although each component of the abnormality sign detection system 1 is installed on a separate computer, these components do not necessarily need to be realized by multiple computers. For example, each component of the abnormality sign detection system 1 may be realized by a single computer.

[0038] The learning computer 5 includes an input unit 7, an output unit 8, a communication unit 9, a memory unit 10, and a processing circuit 11. However, the learning computer 5 does not necessarily have to include the input unit 7, the output unit 8, and the memory unit 10.

[0039] Predetermined information is input to the input unit 7 in response to operations by the user of the learning computer 5. The input unit 7 includes input devices such as a mouse or a keyboard. In other words, predetermined information is input to the input unit 7 in response to operations of these input devices.

[0040] The output unit 8 outputs predetermined information. For example, the learning computer 5 includes a device for displaying images, such as a display that outputs analysis results. In other words, the output unit 8 controls the images displayed on the display. The display may be separate from the computer main body, or may be integrated with it.

[0041] The communication unit 9 communicates with the data input computer 4 or the detection computer 6 via a predetermined communication line. In the first embodiment, the data input computer 4, the learning computer 5, and the detection computer 6 are connected to each other via a LAN (Local Area Network). The data input computer 4, the learning computer 5, and the detection computer 6 may also be connected to each other via the Internet, a WAN (Wide Area Network), or a mobile communication network. Furthermore, each device may be connected to each other via a bus.

[0042] The storage unit 10 stores various information required for generating an abnormality sign detection model. For example, the storage unit 10 stores the actual process value 30 sent from the data input computer 4.

[0043] The processing circuit 11 is a circuit including, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a dedicated or general-purpose processor. This processor realizes various functions by executing various programs stored in the storage unit 10. The processing circuit 11 may also be configured with hardware such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). Various functions can also be realized by such hardware. The processing circuit 11 can also realize various functions by combining software processing by a processor and a program with hardware processing.

[0044] The processing circuit 11 of the first embodiment includes a preprocessing unit 12 and a learning model generation unit 13. These are realized by a CPU executing a program stored in a memory or a HDD.

[0045] The preprocessing unit 12 performs preprocessing of learning input data (input signals) to be input to the neural network. For example, the learning input data is generated based on a plurality of actual process values ​​30 generated in the plant 2. Here, a process is executed to correct at least one of the plurality of actual process values ​​30.

[0046] The pre-processing unit 12 includes an actual process value acquiring unit 20, a reference process value selecting unit 21, a correction necessity determination coefficient calculating unit 22, a correction necessity determining unit 23, a correction value calculating unit 24, an arithmetic unit 25, and a signal selecting unit 26 (see FIG. 2). The learning input data generated by the pre-processing unit 12 is input to the learning model generating unit 13.

[0047] Although each component of the pre-processing unit 12 is described as a minimum configuration, there may be a plurality of each component. In addition, the pre-processing unit 12 may have a function of interpolating missing data in the time-series data generated in the plant 2.

[0048] The learning model generation unit 13 performs machine learning by inputting the learning input data generated by the preprocessing unit 12 into the anomaly sign detection model. The anomaly sign detection model generated by the learning model generation unit 13 is set in the detection computer 6.

[0049] The detection computer 6 performs anomaly sign detection using an autoencoder. For example, the detection computer 6 acquires monitored data sent from the data input computer 4. The detection computer 6 inputs a plurality of actual process values ​​30 converted from the monitored data as determination input data into an input layer of a trained anomaly sign detection model. The detection computer 6 then acquires determination output data that is output from the output layer of the anomaly sign detection model in response to the input of the determination input data and that restores the normal state of the plurality of actual process values ​​30. The detection computer 6 determines the presence or absence of at least one of an anomaly or a sign of an anomaly in the plant 2 based on the difference between the determination input data and the determination output data. In this embodiment, even when anomaly sign detection is performed using an autoencoder under conditions in which sufficient training data cannot be acquired, false detections can be suppressed and accuracy can be improved.

[0050] This anomaly sign detection model includes an input layer, an intermediate layer, and an output layer. Learning input data or judgment input data is input to the input layer. The output layer outputs judgment output data in response to the input of judgment input data. The intermediate layer's parameters are machine-learned using the learning input data. The anomaly sign detection model causes the detection computer 6 to function so as to determine the presence or absence of at least one of an anomaly or a sign of an anomaly in the plant 2 based on the difference between the judgment input data and the judgment output data.

[0051] Next, the flow of the processing executed by the abnormality sign detection system 1 up to the generation of learning input data will be described with reference to FIGS.

[0052] Note that the arrows in FIG. 2 are an example showing the flow of processes executed in the preprocessing unit 12, and there may be other process flows besides those shown by the arrows. Furthermore, the order of the processes is not necessarily fixed, and the order of some processes may be reversed. Furthermore, some processes may be executed in parallel with other processes. Furthermore, the anomaly sign detection system 1 may include components other than those shown in FIG. 2, or some of the components shown in FIG. 2 may be omitted.

[0053] As shown in Fig. 1, first, each sensor 3 installed in a plant 2 acquires each measurement value generated in the plant 2. Then, the measurement values ​​acquired by these sensors 3 are collected in a data input computer 4 and sent to a learning computer 5 as monitoring target data.

[0054] As shown in FIG. 2, in the preprocessing unit 12 of the training computer 5, the actual process value acquiring unit 20 acquires the monitoring target data from the data input computer 4.

[0055] The monitored data is time-series data and may be either analog data or digital data. The actual process value acquisition unit 20 converts the monitored data into digital data that can be input to the learning model generation unit 13, i.e., actual process values ​​30.

[0056] For example, the actual process value acquiring unit 20 associates a sensor management number, a sensor name, and the like with each of the multiple measurement values ​​included in the monitoring target data, and outputs them as the actual process value 30. In other words, the actual process value acquiring unit 20 converts each of the multiple measurement values ​​related to the plant 2 into a format used for machine learning of the anomaly sign detection model.

[0057] These actual process values ​​30 are used for multiple purposes. For example, these actual process values ​​30 are input to the reference process value selection unit 21, the correction value calculation unit 24, and the calculation unit 25, and are also input to the learning model generation unit 13.

[0058] The reference process value selection unit 21 selects at least one reference process value 31 that serves as a reference for data correction from a plurality of actual process values ​​30. The selected reference process value 31 is input to the correction necessity determination coefficient calculation unit 22 and the correction value calculation unit 24.

[0059] For example, the reference process value 31 is assumed to have correlations among multiple actual process values ​​30. When automatically determining whether correction is necessary, it is sufficient to use the reference process value 31 that has the greatest correlation among the actual process values ​​30. For example, in a plant 2 having a turbine generator, the feedwater / condensate system that cools steam for power generation back to water and supplies this water back to the heat source may be monitored. In this case, the output of the turbine generator and the seawater temperature, for example, would be the value that has the greatest correlation among the multiple actual process values ​​30 in the feedwater / condensate system. Note that the reference process value 31 can be set individually for each actual process value 30. The same reference process value 31 can also be set for multiple actual process values ​​30.

[0060] The correction necessity determination coefficient calculation unit 22 calculates a correction necessity determination coefficient 32 for determining whether or not the actual process value 30 is correlated with the reference process value 31, from at least one actual process value 30 and the reference process value 31. The calculated correction necessity determination coefficient 32 is input to the correction necessity determination unit 23.

[0061] Here, not all actual process values ​​30 may be corrected, but only the actual process values ​​30 that are correlated with the reference process value 31 need to be corrected. Therefore, in order to determine whether or not each actual process value 30 needs to be corrected, the correction necessity determination coefficient calculation unit 22 calculates a correction necessity determination coefficient 32. Note that when calculating the correction necessity determination coefficient 32, a plurality of actual process values ​​30 may be used.

[0062] Here, a method using a coefficient of determination or the like can be used to calculate the correction necessity determination coefficient 32. In this way, the correction necessity determination coefficient 32 can be calculated on a statistically appropriate scale.

[0063] The correction necessity determination unit 23 determines whether each actual process value 30 is correlated with the reference process value 31 based on the correction necessity determination coefficient 32. If it is determined that correction is necessary, the correction necessity determination unit 23 sets a correction determination flag 33 indicating the determination result. The set correction determination flag 33 is input to the signal selection unit 26.

[0064] Here, one method of determination is to set an arbitrary threshold value in advance for the correction necessity determination coefficient 32. For example, if the correction necessity determination coefficient 32 is equal to or greater than the threshold value, it is determined that the actual process value 30 needs to be corrected.

[0065] Such a determination method, for example, the threshold value used for the determination, does not need to be set uniquely and may be changed in accordance with the reference process value 31. For example, in a plant 2 having a turbine generator, if the seawater temperature is set as the reference process value 31, the degree of vacuum in the condenser is linked to the seawater temperature. For this reason, the correction necessity determination coefficient 32 is set high, and it is determined that correction is necessary. In contrast, the steam temperature input to the turbine generator is not linked to the seawater temperature. For this reason, the correction necessity determination coefficient 32 is set low, and it is determined that correction is not necessary.

[0066] The correction value calculation unit 24 calculates a correction value 34 for correcting at least one actual process value 30 from at least one actual process value 30 and the reference process value 31. The calculated correction value 34 is input to the calculation unit 25.

[0067] Here, one method for calculating the correction value 34 is to set it based on a linear regression coefficient between at least one actual process value 30 and at least one reference process value 31. Note that when calculating the correction value 34, multiple actual process values ​​30 may be used.

[0068] For example, in a plant 2 having a turbine generator, if seawater temperature is set as the reference process value 31, there is a percentage of the condenser vacuum that fluctuates with fluctuations in this seawater temperature. This percentage of fluctuation is calculated, and a correction value 34 for the condenser vacuum is calculated based on the seasonal fluctuation range of the seawater temperature. For example, if the seawater temperature changes by 1°C, the condenser vacuum will fluctuate by 1 mmHg. In this case, if the seasonal fluctuation range of the seawater temperature is ±5°C, the correction value 34 for the condenser vacuum will be ±5 mmHg.

[0069] The calculation unit 25 corrects at least one actual process value 30 determined to be correlated with the reference process value 31 with the correction value 34, and calculates a corrected process value 35. The calculated corrected process value 35 is input to the signal selection unit 26.

[0070] Here, the calculation method for correcting the actual process value 30 may be, for example, arithmetic operations. Alternatively, a function of the correction value 34 and the actual process value 30 may be calculated.

[0071] The correction value 34 is calculated using linear regression coefficients, and the corrected process value 35 is generated by adding, subtracting, integrating, or dividing this correction value 34. In this way, the correction value 34 can be calculated using a value close to the actual operating state, and the corrected process value 35 that approximates the actual operation can be generated.

[0072] The signal selection unit 26 generates learning input data including at least one corrected process value 35 corrected with the correction value 34.

[0073] For example, the signal selection unit 26 selects whether to output a signal including information on the actual process value 30 for which the correction determination flag 33 is set, that is, information on the corrected process value 35. The signal including information on this corrected process value 35 is output from the signal selection unit 26 and added to the signal including information on the actual process value 30 output from the actual process value acquisition unit 20. Learning input data including the corrected process value 35 and the actual process value 30 is input to the learning model generation unit 13.

[0074] If no correction is required, that is, if the signal contains information about the actual process value 30 for which the correction determination flag 33 is not set, the signal is input to the learning model generation unit 13 as is.

[0075] In this way, when the learning input data is input from the preprocessing unit 12 to the learning model generation unit 13, the learning model generation unit 13 inputs the learning input data into the anomaly sign detection model and performs machine learning.

[0076] An example of the correction image in the calculation unit 25 is shown in Fig. 3. The calculation unit 25 performs calculation on the actual process value 30 using the correction value 34 to calculate the corrected process value 35. In the case of the waveform in Fig. 3, the corrected process value 35 is calculated by subtracting it from the actual process value 30.

[0077] This corrected process value 35 is selected by the signal selection unit 26 and input to the learning model generation unit 13. This allows additional data (corrected process value 35) that is not included in the actual data (actual process value 30) acquired from the plant 2 to be added as data to be used for machine learning.

[0078] Next, another example of the correction image in the calculation unit 25 is shown in Fig. 4. The calculation unit 25 calculates multiple corrected process values ​​35, 35' using multiple correction values ​​34, 34' for one actual process value 30. In the case of the waveform in Fig. 4, two corrected process values ​​35, 35' are calculated by adding and subtracting the respective correction values ​​34, 34' to and from one actual process value 30.

[0079] The anomaly sign detection system 1 of the first embodiment can appropriately enhance normal data for time-series data during machine learning. The anomaly sign detection model can identify the waveform of the corrected process value 35 as a normal state, in addition to the waveform of the actual process value 30. Therefore, even if the actual process value 30 fluctuates to the corrected process value 35 during operation of the plant 2, the possibility of detecting it as an anomaly can be reduced. This reduces the total number of detections, improving the detection performance (true anomaly detections / total number of detections). In other words, it is possible to prevent the detection computer 6 from unnecessarily detecting "different" operation during operation of the plant 2, i.e., making false detections.

[0080] (Second embodiment) Next, a second embodiment will be described with reference to Figures 5 to 9. Note that the same components as those shown in the above-described embodiment will be assigned the same reference numerals and redundant description will be omitted.

[0081] 5, the processing circuit 11 of the abnormality sign detection system 1A of the second embodiment includes a setting reception unit 14 in addition to a preprocessing unit 12 and a learning model generation unit 13. These are realized by the CPU executing a program stored in a memory or an HDD.

[0082] As shown in FIG. 7, the setting reception unit 14 includes a reference process value setting unit 40, a coefficient setting unit 41, a judgment setting unit 42, a correction value setting unit 43, a correction value calculation assisting unit 44, and a calculation method specifying unit 45.

[0083] The reference process value setting unit 40 receives an external input, for example, an input operation by a user, for selecting the reference process value 31. In this way, the user can select the reference process value 31 as desired.

[0084] Here, the reference process value 31 can be selected automatically, or the user can arbitrarily select the reference process value 31 through input operations. For example, when the seawater temperature is automatically selected as the reference process value 31, if the user wants to select the heat source output of plant 2 as the reference process value 31, the user can set the reference process value 31, and this heat source output will be preferentially selected as the reference process value 31 in the reference process value selection unit 21 (FIG. 6).

[0085] The coefficient setting unit 41 receives an external input, for example, an input operation by a user, for calculating the correction necessity determination coefficient 32. In this way, the user can arbitrarily perform settings for calculating the correction necessity determination coefficient 32.

[0086] Here, the correction necessity determination coefficient 32 can be calculated automatically, but the correction necessity determination coefficient 32 can also be arbitrarily set by a user's input operation. By the user setting it in this way, the correction necessity determination coefficient 32 set by the user is given priority in the correction necessity determination coefficient calculation unit 22 (FIG. 6).

[0087] The determination setting unit 42 receives an external input for determining whether correction is necessary, for example, an input operation by the user. In this way, the user can arbitrarily perform settings for determining whether correction is necessary.

[0088] Here, the determination of whether correction is necessary can be performed automatically, or can be performed arbitrarily by a user's input operation. By performing this setting by the user, the correction necessity determination unit 23 (FIG. 6) can give priority to the determination result set by the user.

[0089] The correction value setting unit 43 receives an external input, for example, an input operation by the user, for calculating the correction value 34. In this way, the user can arbitrarily perform settings for calculating the correction value 34.

[0090] Here, the correction value 34 can be calculated automatically, but the user can also arbitrarily set the correction value 34 through input operations. By setting this by the user, the correction value 34 set by the user is given priority in the correction value calculation unit 24 (FIG. 6).

[0091] The correction value calculation assistant unit 44 accepts an external input, for example, an input operation by a user, for auxiliary adjustment of the correction value 34. In this way, the user can arbitrarily adjust the correction value 34 in the correction value calculation unit 24 (FIG. 6).

[0092] The calculation method designation unit 45 receives an external input, for example, a user input operation, for designating a calculation method to be used when correcting at least one actual process value 30 with the correction value 34. In this way, the user can arbitrarily designate a calculation method.

[0093] Here, the calculation method can be automatically calculated, but the user can also arbitrarily set the calculation method by input operation. By setting this by the user, the calculation method set by the user is set preferentially in the calculation unit 25 (FIG. 6).

[0094] In the second embodiment, various settings are made by external input, and this is effective when correcting the operation pattern of the redundant equipment, for example.

[0095] First, an example of erroneous determination of redundant equipment will be described. As shown in Fig. 8, for example, assume that there is a system in which a first pipe 51 and a second pipe 52 join together midway to form a third pipe 53. The first pipe 51 is provided with a first pump 54 and a first pressure gauge 55 that measures its discharge pressure. The second pipe 52 is provided with a second pump 56 and a second pressure gauge 57 that measures its discharge pressure. The third pipe 53 is provided with a flow meter 58 that measures its flow rate.

[0096] It is assumed that the system is made redundant by the first pump 54 and the second pump 56. The operator selects the redundant operation conditions for the first pump 54 and the second pump 56. For example, during a learning period, only the first pump 54 (the pump on the operating side) is always in operation. Also, during this learning period, the other pump, the second pump 56 (the pump on the stopped side), remains stopped.

[0097] If only the first pump 54 is operating during the learning period, the correlation between the discharge pressure of the first pump 54, which indicates the operating state, and the flow rate of the third pipe 53, which indicates the behavior of the entire system, is learned. On the other hand, the discharge pressure of the second pump 56, which was stopped during the learning period, does not fluctuate, and therefore the correlation between the discharge pressure of the first pump 54 and the flow rate of the third pipe 53 is not learned.

[0098] For example, during the monitoring period, the conditions for redundant operation may change due to an operator's selection, causing the first pump 54 to be stopped and only the second pump 56 to operate. In this case, the flow rate of the third pipe 53, which is linked to the discharge pressure of the second pump 56, causes a mis-linking of the predicted value of the discharge pressure of the first pump 54, whose correlation has been learned. The predicted value of the flow rate of the third pipe 53 is also affected, causing a mis-linking.

[0099] That is, the predicted values ​​(output data for determination) of the discharge pressure of the first pump 54 and the flow rate of the third pipe 53 differ from the actually measured values ​​(input data for determination). As a result, it is erroneously determined that there is an abnormality in the discharge pressure of the first pump 54 and the flow rate of the third pipe 53, even though there is no actual abnormality.

[0100] 9, in the second embodiment, the actual data acquired from the first pressure gauge 55 is set as the actual process value 30A to be used for correction, and the actual data acquired from the second pressure gauge 57 is set as the actual process value 30B to be corrected. Here, it is necessary to supplement the actual data when the second pump 56 is operating. Therefore, the actual process value 30B to be corrected is corrected with the correction value 34, and a corrected process value 35 is calculated.

[0101] In this case, the coefficient setting unit 41 increases the correction necessity determination coefficient 32 corresponding to the actual process value 30B to be corrected. For example, the correction necessity determination coefficient 32 is set to "1." Alternatively, the determination setting unit 42 sets the correction determination flag 33 corresponding to the actual process value 30B to be corrected to "correction required." In this way, it is possible to determine that the actual process value 30B to be corrected is to be corrected.

[0102] Here, the correction value setting unit 43 performs setting to calculate the correction value 34 based on the actual process value 30A used for correction. That is, the difference between the actual process value 30A used for correction and the actual process value 30B to be corrected is set as the correction value 34. Then, the calculation method designation unit 45 designates "addition" as the calculation method. Then, the actual process value 30B to be corrected is corrected as the corrected process value 35.

[0103] The second pump 56 that was stopped is considered to have been operating, as shown in the corrected process value 35. This corrected process value 35 is used in machine learning as additional data.

[0104] As shown in FIG. 7, in the second embodiment, an effect of providing the correction value calculation auxiliary unit 44 is that the actual data required to calculate the correction value 34 can be obtained from sources other than the plant 2 to be monitored.

[0105] For example, in the first embodiment, the correction value 34 is calculated based on the fluctuation range of the seawater temperature. However, it may be impossible to obtain the fluctuation range of the seawater temperature from the actual data of the plant 2. In this case, the correction value 34 is set based on external data such as data released by a national organization. By doing so, a more appropriate correction value 34 can be set.

[0106] According to the second embodiment, it is possible to incorporate external knowledge other than that of the plant 2, and it is possible to more appropriately enhance normal data for time-series data.

[0107] (Third embodiment) Next, a third embodiment will be described with reference to Figures 10 to 12. Note that the same components as those shown in the above-described embodiments will be assigned the same reference numerals and redundant description will be omitted.

[0108] 10 and 11, the abnormality sign detection system 1B of the third embodiment includes an external database 60 in addition to the configuration of the second embodiment. The configuration of the processing circuit 11 of the third embodiment is the same as that of the processing circuit 11 of the second embodiment. For an explanation of the setting reception unit 14, please refer to the above-mentioned FIG. 7 as appropriate.

[0109] The training computer 5 of the third embodiment acquires additional training data generated to simulate the actual process values ​​30 from the external database 60, and generates training input data based on this additional training data. In this way, any design value, a model-based theoretical value, a predicted value newly calculated by AI, or the like can be used as training data.

[0110] For example, in the operation of Plant 2, the monitored data itself may be added for the purpose of improving functionality and monitorability. For example, a bypass line may be added to the system to improve operability, and the flow rate of the bypass line may be added and measured. In this case, since there is no learning input data for the added data, the detection computer 6 cannot monitor for anything "unusual."

[0111] In this case, it is conceivable to use arbitrary design values, model-based theoretical values, or newly calculated predicted values ​​using AI as input data for learning. However, because it is not possible to take into account correlations in actual equipment, particularly the effects of transient changes and noise, there is a possibility that the input data for learning may not necessarily be correct.

[0112] The third embodiment can solve this problem by increasing the amount of data. For example, any design value, a model-based theoretical value, or a predicted value newly calculated by AI is set as additional learning data. This additional learning data is then input to the actual process value acquisition unit 20. The reference process value setting unit 40 (FIG. 7) performs setting to select a reference process value 31 having characteristics to be corrected for the additional learning data.

[0113] An image of the correction of the additional data for learning is shown in Figure 12. The reference process value 31 selected by the setting in the reference process value setting unit 40 (Figure 7) is the process value used for correction. The additional data for learning is shown as a constant value 36, assuming a design value.

[0114] The correction judgment flag 33 is set to "correction required" by at least one of the coefficient setting unit 41 (FIG. 7) and the judgment setting unit 42 (FIG. 7). The correction value calculation unit 24 or the correction value setting unit 43 (FIG. 7) sets the correction value 34 as the fluctuation range (difference) from the average value of the reference process value 31 used for correction. The calculation method designation unit 45 (FIG. 7) designates "addition" as the calculation method. Then, the correction value 34 is added to the additional data for learning (constant value 36), thereby calculating the corrected process value 35 107.

[0115] In the third embodiment, it is possible to add targets for abnormality sign detection, and to appropriately augment additional data having characteristics unique to an actual machine as normal data for learning.

[0116] The abnormality precursor detection system 1 (1A, 1B) has been described based on the first to third embodiments, but the configuration applied in any of the embodiments may be applied to other embodiments, and the configurations applied in each embodiment may be combined.

[0117] In the above-described embodiment, the function of the actual process value acquiring unit 20 is provided in the learning computer 5, but other configurations are also possible. For example, the function of the actual process value acquiring unit 20 may be provided in the data input computer 4. In this case, the learning computer 5 does not need to have the function of the actual process value acquiring unit 20, and the multiple actual process values ​​30 processed by the data input computer 4 are input to the learning computer 5 for subsequent processing.

[0118] The abnormality sign detection system 1 described above includes a control device that is a highly integrated processor such as an FPGA (Field Programmable Gate Array), GPU (Graphics Processing Unit), CPU (Central Processing Unit), or dedicated chip, storage devices such as ROM (Read Only Memory) and RAM (Random Access Memory), external storage devices such as HDD (Hard Disk Drive) and SSD (Solid State Drive), a display device such as a monitor, input devices such as a mouse and keyboard, and a communication interface. This abnormality sign detection system 1 can be realized with a hardware configuration that uses a normal computer.

[0119] The program executed by the abnormality sign detection system 1 is provided in advance as a program installed in a ROM or the like. Additionally or alternatively, the program is provided as a file in an installable or executable format stored on a computer-readable non-transitory storage medium such as a CD-ROM, CD-R, memory card, DVD, or flexible disk (FD).

[0120] The program executed by this anomaly sign detection system 1 may be stored on a computer connected to a network such as the Internet and provided by downloading it via the network. In other words, the program may be provided via the cloud. Alternatively, a server on the cloud may execute the program, and only the processing results may be provided via the cloud. Furthermore, this anomaly sign detection system 1 may be configured by combining separate modules that independently perform the functions of the components and interconnect them via a network or dedicated lines.

[0121] According to at least one of the embodiments described above, by correcting at least one actual process value 30 determined to be correlated with the reference process value 31 with the correction value 34, it is possible to suppress false detections and improve the reliability of the abnormality precursor detection function.

[0122] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations can be made without departing from the spirit of the invention. These embodiments and modifications thereof are intended to be included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents. [Explanation of symbols]

[0123] 1, 1A, 1B... anomaly detection system, 2... plant, 3... sensor, 4... data input computer, 5... learning computer, 6... detection computer, 7... input section, 8... output section, 9... communication section, 10... memory section, 11... processing circuit, 12... preprocessing section, 13... learning model generation section, 14... setting reception section, 20... actual process value acquisition section, 21... reference process value selection section, 22... correction necessity determination coefficient calculation section, 23... correction necessity determination section, 24... correction value calculation section, 25... calculation section, 26... signal selection section, 30, 30A, 30 B...actual process value, 31...reference process value, 32...coefficient for determining whether correction is necessary, 33...determination flag for correction, 34, 34'...correction value, 35, 35'...process value after correction, 36...constant value, 40...reference process value setting section, 41...coefficient setting section, 42...determination setting section, 43...correction value setting section, 44...correction value calculation auxiliary section, 45...calculation method designation section, 51...first piping, 52...second piping, 53...third piping, 54...first pump, 55...first pressure gauge, 56...second pump, 57...second pressure gauge, 58...flow meter, 60...external database.

Claims

1. one or more computers that perform machine learning of an anomaly sign detection model that detects at least one of an anomaly or a sign of the anomaly in a monitored facility, The computer Acquire a plurality of actual process values ​​occurring in the target facility; selecting at least one reference process value from the plurality of actual process values ​​as a reference for correction; calculating a correction necessity determination coefficient for determining whether the actual process value is correlated with the reference process value from at least one of the actual process value and the reference process value; calculating a correction value for correcting the actual process value from at least one of the actual process value and the reference process value; determining whether each of the actual process values ​​is correlated with the reference process value based on the correction necessity determination coefficient; correcting at least one of the actual process values ​​determined to be correlated with the reference process value with the correction value; generating learning input data including at least one corrected process value corrected with the correction value; The learning input data is input into the anomaly sign detection model to perform the machine learning. It is structured as follows: The correction necessity determination coefficient is calculated using a coefficient of determination. Anomaly detection system.

2. the computer receives an external input for selecting the reference process value; The abnormality sign detection system according to claim 1 .

3. the computer receives an external input for calculating the correction necessity determination coefficient; The abnormality sign detection system according to claim 1 .

4. the computer accepts an external input for calculating the correction value; The abnormality sign detection system according to claim 1 .

5. The computer receives an external input for auxiliary adjustment of the correction value. The abnormality sign detection system according to claim 1 .

6. the computer receives an external input for specifying an algorithm to be used when correcting at least one of the actual process values ​​with the correction value; The abnormality sign detection system according to claim 1 .

7. the computer acquires additional learning data generated by simulating the actual process value from an external database, and generates the learning input data based on the additional learning data; The abnormality sign detection system according to claim 1 .

8. one or more computers that perform machine learning of an anomaly sign detection model that detects at least one of an anomaly or a sign of the anomaly in a monitored facility, The computer Acquire a plurality of actual process values ​​occurring in the target facility; selecting at least one reference process value from the plurality of actual process values ​​as a reference for correction; calculating a correction necessity determination coefficient for determining whether the actual process value is correlated with the reference process value from at least one of the actual process value and the reference process value; calculating a correction value for correcting the actual process value from at least one of the actual process value and the reference process value; determining whether each of the actual process values ​​is correlated with the reference process value based on the correction necessity determination coefficient; correcting at least one of the actual process values ​​determined to be correlated with the reference process value with the correction value; generating learning input data including at least one corrected process value corrected with the correction value; The learning input data is input into the anomaly sign detection model to perform the machine learning. It is structured as follows: The correction value is calculated using a linear regression coefficient, and the corrected process value is generated by adding, subtracting, integrating, or dividing the correction value. Anomaly detection system.

9. The method is carried out using one or more computers that perform machine learning of an anomaly sign detection model that detects at least one of an anomaly or a sign of the anomaly in a monitored facility, Acquire a plurality of actual process values ​​occurring in the target facility; selecting at least one reference process value from the plurality of actual process values ​​as a reference for correction; calculating a correction necessity determination coefficient for determining whether the actual process value is correlated with the reference process value from at least one of the actual process value and the reference process value; calculating a correction value for correcting the actual process value from at least one of the actual process value and the reference process value; determining whether each of the actual process values ​​is correlated with the reference process value based on the correction necessity determination coefficient; correcting at least one of the actual process values ​​determined to be correlated with the reference process value with the correction value; generating learning input data including at least one corrected process value corrected with the correction value; The learning input data is input into the anomaly sign detection model to perform the machine learning. The processing is executed by the computer, The correction necessity determination coefficient is calculated using a coefficient of determination. A method for generating an anomaly detection model.

10. The method is carried out using one or more computers that perform machine learning of an anomaly sign detection model that detects at least one of an anomaly or a sign of the anomaly in a monitored facility, Acquire a plurality of actual process values ​​occurring in the target facility; selecting at least one reference process value from the plurality of actual process values ​​as a reference for correction; calculating a correction necessity determination coefficient for determining whether the actual process value is correlated with the reference process value from at least one of the actual process value and the reference process value; calculating a correction value for correcting the actual process value from at least one of the actual process value and the reference process value; determining whether each of the actual process values ​​is correlated with the reference process value based on the correction necessity determination coefficient; correcting at least one of the actual process values ​​determined to be correlated with the reference process value with the correction value; generating learning input data including at least one corrected process value corrected with the correction value; The learning input data is input into the anomaly sign detection model to perform the machine learning. The processing is executed by the computer, The correction value is calculated using a linear regression coefficient, and the corrected process value is generated by adding, subtracting, integrating, or dividing the correction value. A method for generating an anomaly detection model.

11. a program executed on one or more computers that performs machine learning of an anomaly sign detection model that detects at least one of an anomaly or a sign of the anomaly in a monitored facility, Acquire a plurality of actual process values ​​occurring in the target facility; selecting at least one reference process value from the plurality of actual process values ​​as a reference for correction; calculating a correction necessity determination coefficient for determining whether the actual process value is correlated with the reference process value from at least one of the actual process value and the reference process value; calculating a correction value for correcting the actual process value from at least one of the actual process value and the reference process value; determining whether each of the actual process values ​​is correlated with the reference process value based on the correction necessity determination coefficient; correcting at least one of the actual process values ​​determined to be correlated with the reference process value with the correction value; generating learning input data including at least one corrected process value corrected with the correction value; The learning input data is input into the anomaly sign detection model to perform the machine learning. causing the computer to execute a process; The correction necessity determination coefficient is calculated using a coefficient of determination. Anomaly prediction detection model generation program.

12. a program executed on one or more computers that performs machine learning of an anomaly sign detection model that detects at least one of an anomaly or a sign of the anomaly in a monitored facility, Acquire a plurality of actual process values ​​occurring in the target facility; selecting at least one reference process value from the plurality of actual process values ​​as a reference for correction; calculating a correction necessity determination coefficient for determining whether the actual process value is correlated with the reference process value from at least one of the actual process value and the reference process value; calculating a correction value for correcting the actual process value from at least one of the actual process value and the reference process value; determining whether each of the actual process values ​​is correlated with the reference process value based on the correction necessity determination coefficient; correcting at least one of the actual process values ​​determined to be correlated with the reference process value with the correction value; generating learning input data including at least one corrected process value corrected with the correction value; The learning input data is input into the anomaly sign detection model to perform the machine learning. causing the computer to execute a process; The correction value is calculated using a linear regression coefficient, and the corrected process value is generated by adding, subtracting, integrating, or dividing the correction value. Anomaly prediction detection model generation program.

Images