JP6932467B2 - State change detection device, state change detection system and state change detection program - Google Patents

Description

The present invention relates to a state change detection device, a state change detection system, and a state change detection program.

For example, when the current, voltage, and vibration of a machine are measured to detect a state change such as an abnormality, the abnormality is generally determined based on past experience by a person or the like. That is, the threshold value of the state change is set by a person, and it is almost impossible for a machine having no abnormal data to perform the state change.

Patent Document 1 discloses a highly versatile anomaly detection device that does not depend on the type and number of objects and events that detect abnormal situations, and the space (place, time zone, etc.) that monitors those objects and events. ing.

Specifically, it is a device that detects that an abnormality has occurred in a target event. In this device, the time series of the input pattern, which is the data that changes depending on the event, is acquired, the characteristics of the transition of the acquired input pattern in the time series are analyzed, and the analyzed characteristics of the transition and the characteristics of the transition in advance are analyzed. It is compared with the determined reference values, and if they are not close to each other within a certain range, it is determined that an abnormality has occurred in the above-mentioned event. In this way, when it is determined that an abnormality has occurred as a result of the above comparison, an output means for performing an output operation to that effect is provided.

That is, an time series in which data that changes according to a change in an event such as the movement of a person or an object to be monitored is used as an input pattern is acquired, attention is paid to the transition of the pattern in the time series, and the feature amount in the transition is extracted. However, the occurrence of an abnormal situation is detected by comparing it with that in a normal case. Here, the "event" is a phenomenon that can be expressed as data that can be processed by a computer by a signal from a device sensor or the like, a report (data input) from a person, or the like, and "human behavior" in the house. It is also applicable to.

Patent Document 2 discloses a data prediction method or data prediction system that uses a prediction model that realizes highly accurate prediction and that allows the prediction process to be understood.

The invention of Patent Document 2 is a data prediction method for predicting the future based on past achievements, wherein the prediction model includes a first prediction model that outputs a prediction value for at least one feature amount on the prediction target date. It is composed of a second prediction model that includes the prediction value output from the first prediction model as an input factor and outputs the prediction value for each predetermined time on the prediction target day.

Then, a prediction model construction means for constructing a prediction model using the collected nearest actual data and past actual data, and prediction execution for obtaining a predicted value by inputting input data for prediction into the constructed prediction model. Means, predictive error calculation means for calculating prediction error or model error from collected closest actual data and predicted value, calculation of correction coefficient or correction amount to correct the predicted value based on prediction error or model error, and correction It is realized by the predicted value correction means for obtaining the predicted value.

References 3 include rainwater inflow forecasts, inflow water quality forecasts, river water level forecasts, dam storage volume forecasts, demand forecasts for water, gas, electricity, etc., natural phenomena and artificial phenomena / events. A prediction model system used for making online predictions is disclosed. In Patent Document 3, the method using a prediction model can be roughly classified into a method called a white box approach using a prediction model based on a mathematical formula based on the laws of physics and chemistry, and a method called a white box approach, which uses actually available data. , There is a description that it can be divided into a method called a black box approach that constructs a prediction model from actual data by performing statistical processing, learning (for example, neural network, etc.), system identification, and so on.

The invention of Patent Document 3 considers the physical law (conservation law) by the white box approach in the prediction by the system identification method using the model of the discrete time system, which is a typical method of the black box approach. Provide a predictive model system that can build a model. The prediction model system of Patent Document 3 includes a prediction model including a prediction model including a plurality of parameters, a prediction model structure unit for predicting a prediction target, a parameter value storage unit for storing parameter values, and input / output data for storing input and output time series data. Provide storage means. Constraints on parameters are determined in the static parameter identification means based on stationary input data and output data. Parameters that satisfy this constraint condition are determined by the dynamic parameter identification means based on the non-stationary input data and output data stored in the input / output data storage means, and sent to the parameter value storage unit.

Patent Document 4 forecasts electric power demand, water demand, heat demand, gas demand, steam demand and other demands, various loads, various sales volumes, various economic indicators, etc., and past results or related information. A data prediction method and a data prediction system for predicting the future are disclosed using.

Patent Document 4 solves the following problems in the prior art. In the first method using the similar pattern in the past, it is necessary to make a correction according to the situation of the prediction target day, and it is difficult to make an accurate correction. If the second method of creating a forecast model for each forecast target time is adopted, it is necessary to create a model for each forecast target time, which requires a great deal of modeling work. The prediction accuracy is poor because it cannot be modeled. In the third method of collectively predicting the prediction target by the neural network prediction model, the neural network is usually used as the prediction model, but it is difficult to analyze the inside of the neural network because the inside is a black box. , There is a problem that it is difficult to explain the reason for prediction about the prediction result.

In view of the above, the invention of Patent Document 4 aims to reduce a large amount of work and work time for constructing a prediction model, and is highly accurate in consideration of the time-series transition of daily data. The purpose is to provide a prediction method.

The invention of Patent Document 4 is a data prediction method for predicting the future based on past achievements.
The prediction model to be used includes a first prediction model that outputs a prediction value for at least one feature quantity on the prediction target date and a prediction value output from this first prediction model as input factors, and a predetermined prediction target date is used. It is composed of a second prediction model that outputs a prediction value for each time. The invention of Patent Document 4 is a prediction model construction means for constructing a prediction model using the collected closest actual data and past actual data, inputting input data for prediction into the constructed prediction model, executing prediction, and making a prediction. Prediction execution means for obtaining a value, prediction error calculation means for calculating a prediction error or model error from the collected nearest actual data and the predicted value, a correction coefficient or a correction amount for correcting the predicted value based on the prediction error or the model error. Is provided, and the predicted value correction means for obtaining the corrected predicted value is provided.

Patent Document 5 discloses a technique for predicting a synthesis target in speech synthesis, and in particular, a method for creating a prediction model for an acoustic feature amount of a synthesis target during speech synthesis. An object of the present invention of Patent Document 5 is to provide a method for creating a prediction model capable of constructing a prediction model of acoustic features of a synthesis target of speech synthesis more efficiently.

According to the invention of Patent Document 5, the method of creating a prediction model is a method of creating a prediction model for predicting a synthesis target value of an acoustic feature amount for a predetermined voice unit in speech synthesis, and is a predetermined prosody. -A step of preparing computer-readable training data in which prosodic information and linguistic information are attached for each predetermined speech unit and a label for a predetermined acoustic feature amount is attached, and a gradient boosting algorithm using the training data. This includes a step of training a predetermined prediction model.

Further, the invention described in Patent Document 6 relates to a prediction error evaluation device, a prediction error evaluation method, and a prediction error evaluation program. The invention described in Patent Document 6 describes, for example, how the error between the predicted value and the measured value changes (for example, in time series) in relation to the series of attribute values of the phenomenon to be observed. Take this into account so that the error can be evaluated.

In the invention described in Patent Document 6, for example, a predicted value predicted to be obtained by observing the phenomenon for each attribute value of a predetermined phenomenon and the phenomenon observed each time the attribute value of the phenomenon changes. The present invention includes a storage device that stores in advance error data indicating an error from the actually measured value obtained in association with the attribute value of the phenomenon. The invention described in Patent Document 6 has a statistical processing unit, and in a series of attribute values stored in the storage device, one sub-series and at least one of the beginning and the end of the sub-series are used. A plurality of sub-series obtained by shifting by a predetermined number is set as a processing target series, and for each processing target series, error data corresponding to the attribute value included in the processing target series is read from the storage device, and the read error data is obtained. The predetermined statistical processing is executed by the processing device. Further, the output unit includes a result output unit that outputs the result of the statistical processing executed by the statistical processing unit by the output device.

Japanese Unexamined Patent Publication No. 2003-256957 Japanese Unexamined Patent Publication No. 2015-127914 Japanese Unexamined Patent Publication No. 2004-62440 Japanese Unexamined Patent Publication No. 2004-94437 Japanese Unexamined Patent Publication No. 2006-89467 Japanese Unexamined Patent Publication No. 2011-95946

However, as described above, the conventional prediction model aims to improve the accuracy and reliability of prediction and to improve efficiency, and only from the data when there is no state change or it is unknown whether the state change occurs. A state change detection device capable of accurately capturing state change has not been known.

The present invention has been made in view of the current state of the state change detection device as described above, and an object of the present invention is to obtain state change from only data when there is no state change, or only data in which state change occurs. It is an object of the present invention to provide a state change detection device and a state change detection program capable of detecting no state change.

The state change detection device according to the present embodiment uses data obtained from a specific 1 data acquisition source in the data acquired from a plurality of data acquisition sources as an objective variable, and is used from a plurality of data acquisition sources other than the specific 1 data acquisition source. Using the obtained data as an explanatory variable , create a distribution map of the frequency of appearance for the value of the explanatory variable for each explanatory variable unit included in the teacher data for creating a prediction model that predicts the objective variable from the explanatory variable. The distribution map creation means, the average value calculation means for obtaining the average value of the explanatory variables in the explanatory variable unit of the test data, and the position on the distribution map of the corresponding explanatory variable unit in the teacher data. A state change calculation means that obtains position information indicating whether or not it is located, statistically processes all position information for each explanatory variable unit, obtains a tendency of the position information of the statistical processing result, and detects a state change based on this tendency. It is characterized by having a detection means.

The block diagram which shows the structure of the state change detection device or the computer system which realizes the state change detection system which concerns on embodiment of this invention. The figure which shows the means realized by the state change detection program provided in the computer system which realizes the 2nd state change detection apparatus which concerns on embodiment of this invention. The figure which shows an example of the teacher data used for the 2nd state change detection apparatus which concerns on embodiment of this invention. The figure which shows three distribution maps created for each explanatory variable unit of a teacher data in the 2nd state change detection apparatus which concerns on embodiment of this invention. The figure which shows the state which the average value of each is arranged with respect to the 3 distribution charts created for each explanatory variable unit of the teacher data in the 2nd state fluctuation detection apparatus which concerns on embodiment of this invention. The flowchart which shows the operation of the 2nd state change detection apparatus which concerns on embodiment of this invention. The figure of the pie chart which shows the process of determining the presence or absence of the state change by the 2nd state change detection apparatus which concerns on embodiment of this invention by a ratio. The figure which shows the means realized by the state change detection program provided in the computer system which realizes the state change detection apparatus which concerns on this invention using a prediction model. The flowchart for demonstrating the operation of the structure realized by using the prediction model about the state change detection apparatus which concerns on this invention. The figure which shows the teacher data used for the structure which uses the prediction model about the state change detection apparatus which concerns on this invention, the prediction data obtained by applying it to a prediction model, and the value of a standard error. The figure which shows the procedure until the prediction model is made using the teacher data in the state change detection apparatus which concerns on this invention. The figure which shows the procedure from inputting the teacher data to the prediction model in the state change detection apparatus which concerns on this invention, and obtaining the standard error. The figure which shows the measurement data of three times used for the composition realized by using the prediction model about the state change detection apparatus which concerns on this invention, the prediction data obtained by applying them to a prediction model, and the error value. The figure which shows the procedure until the error is obtained by inputting the measurement data of three time times used for the configuration realized by using the prediction model about the state change detection apparatus which concerns on this invention into a prediction model. The figure which conceptually showed the magnification of the error with respect to the standard error obtained from the measurement data of three time in the configuration realized by using the prediction model about the state fluctuation detection apparatus which concerns on this invention. In the configuration realized by using the prediction model for the state fluctuation detection device according to the present invention, a graph showing the difference between the peaks, which are the degrees of dissociation, is obtained by using the standard error distribution map and the error distribution map. .. In the configuration realized by using the prediction model for the state change detection device according to the present invention, it will be described that the state change is detected based on the dissociation degree and the deviation value which are the values obtained by standardizing the error for each measurement. figure. The flowchart for demonstrating the operation in embodiment of the state change detection system which concerns on this invention. The figure which shows the judgment result separated by the threshold value used in the state change presence / absence judgment processing using both one prediction result (ratio) and the other prediction result (error amount) in the embodiment of the state change detection system which concerns on this invention. ..

Hereinafter, embodiments of a state change detection device, a state change detection system, and a state change detection program according to the present invention will be described with reference to the accompanying drawings. In each figure, the same components are designated by the same reference numerals, and duplicate description will be omitted. The state change detection device according to the embodiment of the present invention can be configured by, for example, a personal computer, a workstation, or other computer system as shown in FIG. This computer system operates as a state change detection device by controlling each part based on a program or data stored in the main memory 11 or read into the main memory 11 by the CPU 10 and executing necessary processing. be.

An external storage interface 13, an input interface 14, a display interface 15, and a data input interface 16 are connected to the CPU 10 via a bus 12. A program such as a state change detection program and an external storage device 23 in which necessary data and the like are stored are connected to the external storage interface 13. An input device 24 such as a keyboard as an input device for inputting commands and data and a mouse 22 as a pointing device are connected to the input interface 14.

A display device 25 having a display screen such as an LED or an LCD is connected to the display interface 15. Sensors 26-1, 26-2, ..., 26-m for obtaining measurement data are connected to the data input interface 16. This computer system may be provided with other configurations, and the configuration of FIG. 1 is only an example.

In the CPU 10, each means and the like shown in FIG. 2 is realized by a state change detection program in the external storage device 23. That is, the distribution map creating means 51, the average calculating means 52, the position information calculating means 53, and the state change detecting means 54 are realized. Further, the teacher data T is stored in the external storage device 23.

The distribution map creating means 51 creates a distribution map of the frequency of appearance with respect to the value of the explanatory variable for each explanatory variable unit included in the teacher data T for creating a prediction model for predicting the objective variable from the explanatory variable. It is a thing. As shown in FIG. 3, for example, it is assumed that the teacher data T uses the data measured by the sensor A as the objective variable and the data measured by the sensors B, C, and D as the explanatory variables. In the example of FIG. 3, nine data are measured for each sensor, but the actual number can be much larger than nine.

In the above, "for each explanatory variable unit" means that each of the sensors B, C, and D is used as a unit. Therefore, in the present embodiment, as shown in FIG. 4, three distribution maps MAP1, MAP2, and MAP3 are created corresponding to the sensors B, C, and D.

The average calculation means 52 obtains the average value of the explanatory variables for each explanatory variable of the test data. The test data is data collected for detecting state fluctuations, and is data collected corresponding to sensors B, C, and D, which are explanatory variable units.

The position information calculation means 53 obtains position information indicating at which position the average value obtained by the average calculation means 52 is located on the distribution map of the corresponding explanatory variable unit in the teacher data. For example, when the distribution map MAP as shown in FIG. 5 is obtained, the average value indicated by the diamond mark M is the position of the explanatory variable in the part distribution map MAP. In this embodiment, this position information is represented by σ indicating how far away from the average value in the distribution map MAP, and this σ can be obtained by the following equation (1).

The position information σ used here indicates the degree of deviation of the average value from the average value in each distribution map, and although it differs in what the deviation is, the deviation value will be described later. It has the same technical implications as what is explained as σ. The above σ is a deviation, the root mean square of the deviation is the “variance”, and its positive square root is the “standard deviation”.

The state change detecting means 54 statistically processes all the position information for each explanatory variable unit, obtains the tendency of the position information of the statistical processing result, and detects the state change based on this tendency. That is, in the present embodiment, for each explanatory variable unit, the position information σB, σC, σD for each sensor B, C, D is statistically processed, and the tendency of the position information of the statistical processing result is obtained. In the present embodiment, the case where the number of explanatory variable units is three is taken as an example, but usually, the number of explanatory variable units is larger. Considering the case of such a large number, the position information σB, σC, σD, ... Is usually close to 0 (center of the distribution map), so all the position information σB, σC, σD, ... It is based on the theory that when a lot of (or most) position information in the inside is caught in a tendency to deviate greatly from the normal position, it can be inferred that a state change has occurred. Further, even if the number of explanatory variable units is not large (even if it is about three as in the present embodiment), it is considered that this theory holds when the number of explanatory variables included in one explanatory variable unit is large. Be done.

As statistical processing, the state change detecting means 54 detects whether or not the position information of a predetermined ratio exists in a region larger or smaller than the predetermined position of the distribution map. That is, when a predetermined ratio (80% or 90%, etc.) in the position information σB, σC, σD, ... Is closer to the right side (left side) than a certain position on the right side (or left side) of the distribution map. Concludes that a state change is occurring.

This embodiment is realized by a computer executing a program corresponding to the flowchart shown in FIG. 6A. In the present embodiment, it is assumed that the teacher data T for creating a prediction model for predicting the objective variable from the explanatory variables is prepared. For each explanatory variable unit included in the teacher data T, a distribution map of the frequency of appearance with respect to the value of the explanatory variable is created (S11). As a result, when the explanatory variable units are described as three, as shown in FIG. 4, three distribution maps MAP1, MAP2, and MAP3 are created corresponding to the sensors B, C, and D.

Next, the average value of the explanatory variables is obtained for each explanatory variable of the test data (S12). The position information σ indicating the position of the average value obtained above in the distribution map of the corresponding explanatory variable unit in the teacher data is obtained (S13). Here, as shown in FIG. 5, three σ corresponding to the sensors B, C, and D are obtained as σB = −2.5, σC = 1, and σD = 3.5.

Following step S13, the ratio of the position information σ exceeding the predetermined value (here, 3) is obtained (S14). Here, as shown in the pie chart of FIG. 6B, it is assumed that the position information σ of “3” or more is 10%. Assuming that the threshold value for the presence or absence of state change is, for example, 45%, the presence or absence of state change is determined by comparing the threshold value TH-P = 45 with the ratio obtained above (S15). Thus, it is appropriate even when the process of detecting whether the position information of a predetermined ratio exists in a region larger or smaller than the predetermined position of the above distribution map is performed and the value does not have a value when the state changes. It is possible to determine the presence or absence of state fluctuations.

As the statistical processing performed by the state change detecting means 54, the difference value between the previous position information and the current position information is obtained for all the position information for each explanatory variable unit, and the state change is calculated based on the number of explanatory variable units whose difference value is equal to or more than a predetermined value. It may be detected. That is, it is obtained by the ratio of the number of explanatory variable units in which the difference value between the previous and current position information is equal to or greater than a predetermined value and the number of all explanatory variable units. In the normal state, the difference value between the previous and current position information is greater than or equal to the specified value. It is based on the estimation that the number of explanatory variable units in which the difference value of the position information is equal to or more than a predetermined value will suddenly increase.

The statistical processing performed by the state change detecting means 54 is to detect the state change when even one position information unit whose position information exists in a region larger or smaller than a predetermined position of the distribution map occurs. Is also good. As a case where this processing is effective, in a normal state, the position information σ never exists in a region larger or smaller than a predetermined position in the distribution map. Therefore, it is presumed that the state change occurs when even one position information σ exists in a region larger or smaller than a predetermined position on the distribution map. In this case, "even one" is used, but it holds true even when the number is "predetermined".

Next, an embodiment of the state change detection system according to the present invention will be described. The state change detection system according to the present embodiment is a first state change detection device that detects state change by inputting data collected about an object for which state change should be detected into a prediction model created by machine learning as test data. To be equipped. Further, the state change detection device according to the embodiment described above is provided as the second state change detection device. That is, both the first state change detection device using the prediction model and the state change detection device according to the above embodiment are provided. Since the second state change detection device has been described above, the first state change detection device using the prediction model will be described here.

The CPU 10 shown in FIG. 1 realizes each means and the like shown in FIG. 7 by a state change detection program in the external storage device 23, and also operates as a first state change detection device. That is, the prediction model creating means 31, the prediction model 30, the teacher prediction data acquisition means 32, the standard error acquisition means 33, the measurement prediction data acquisition means 34, the error acquisition means 35, and the state change detection means 36 are realized. Further, teacher data is stored in the external storage device 23. The prediction model creating means 31 creates the prediction model 30 using the teacher data. The prediction model 30 predicts the objective variable from the explanatory variables by machine learning. Here, as a machine learning algorithm, a random forest of pattern mining can be mentioned, but in addition to this, an algorithm such as regression analysis or regression tree may be adopted.

Regarding the explanatory variable and objective variable prediction model, when y is the objective variable, xi is the explanatory variable, and f is the prediction model, y = f (x1, x2, ..., Xi, ..., Xn). Can be represented by. The teacher prediction data acquisition means 32 predicts the teacher explanatory variable by using the teacher data composed of the teacher objective variable which is the objective variable of the fixed value and the teacher explanatory variable which is the explanatory variable of the fixed value. It is input to the model 30 to obtain teacher prediction data. The standard error acquisition means 33 acquires a standard error which is a difference between the teacher prediction data and the teacher objective variable.

The measurement prediction data acquisition means 34 inputs the measurement explanatory variable into the prediction model 30 by using the measurement data composed of the measurement objective variable which is the measured objective variable and the measurement explanatory variable which is the measured explanatory variable. To obtain measurement prediction data. The measurement data can be the data obtained by the sensors 26-1, 26-2, ..., 26-m. The error acquisition means 35 acquires an error which is a difference between the measurement prediction data and the measurement objective variable.

The state change detecting means 36 obtains the degree of dissociation between the standard error and the above error, and detects the state change based on the degree of dissociation.

Since the state change detection device configured by the above means and the like executes the processing operation according to the flowchart shown in FIG. 8, the operation will be described with reference to this flowchart. First, the prediction model 30 is created using the teacher data (STEP 1). For example, as shown in FIG. 9, the teacher data includes the objective variable data to be obtained by the sensor A, the first explanatory variable data obtained by the sensor B, and the second explanatory variable obtained by the sensor C. It is composed of variable data. The number of times data is acquired (the number of data in the vertical direction in FIG. 9) is arbitrary. The teacher data shown in FIG. 9 is the same as that shown in FIG. 3, but explanatory variables are collected by two sensors, one for sensor B and one for sensor C, to simplify the explanation. It is shown as two data. As shown in FIG. 10, the prediction model creating means 31 creates the prediction model 30 using the teacher data T composed of the numerical values as illustrated in FIG.

Next, as shown in the flowchart of FIG. 8 and the operation sequence diagram of FIG. 11, explanatory variables (data of sensor B and data of sensor C of FIG. 9) are input to the prediction model 30 using the teacher data T, and prediction is performed. The data (teacher prediction data TF (teacher data prediction result in FIG. 11)) is obtained, and the error (standard error) which is the difference between the prediction data (teacher prediction data TF) and the objective variable of the teacher data T is obtained (standard error). STEP2). Here, the explanatory variables are the data of the sensor B and the sensor C in the teacher data of FIG. As shown in FIG. 11, if the objective variable of the teacher data T is TT and the objective variable TT is represented by, for example, a perfect circle, the teacher prediction data TF is distorted with respect to this objective variable TT. Therefore, the part that protrudes or retracts from the perfect circle becomes an error. Actually, as shown in the column of "predicted value standard / error of teacher data" in FIG. 9, the predicted data (teacher prediction data) and the error (standard error STER) are obtained as numerical values. Since these are obtained by calculating the same number as the number of times the data is acquired, the same quantity as the number of times the data is acquired can be obtained. Therefore, the average of these is calculated and obtained, and this is held as the average standard error (AVSTER).

Next, as shown in the flowchart of FIG. 8, the explanatory variables M1 and M2 are input to the prediction model 30 using the measurement data M by the sensors B and C, and the prediction data (measurement prediction data) is obtained, and the prediction data is obtained. The error, which is the difference between (measurement prediction data) and the actually measured objective variable of the measurement data M, is obtained (STEP 3).

Here, the measurements t1, t2, and t3 are performed at different times to obtain the explanatory variables t1M1 and t1M2 of the measurement t1, the explanatory variables t2M1 and t2M2 of the measurement t2 are obtained, and the explanatory variables t3M1 and t3M1 of the measurement t3 are obtained. Obtain t3M2 (Fig. 12). Here, since the teacher data shown in FIG. 9 has two explanatory variables, the collected measurement data (test data) is also set to two explanatory variables, but three explanatory variables or four or more explanatory variables should be used. You can also. As shown in FIG. 13, these are input to the same prediction model 30 to obtain prediction data (measurement prediction data) t1F, t2F, and t3F (in the right column in FIG. 12). In FIG. 13, three prediction models 30 are provided, but one prediction model 30 may be used for sequential prediction. Of course, three prediction models 30 may be provided. As shown in FIGS. 12 and 13, the difference between the predicted data (measurement predicted data) t1F, t2F, t3F and the measured objective variables t1MT, t2MT, t3MT of the measured data M is an error (individual error) t1ER, t2ER, t3ER. Ask as. Since the errors t1ER, t2ER, and t3ER are obtained by calculating the same number as the number of times the data is measured, the same quantity as the number of times of acquisition can be obtained. It is held as an average error (t1AVER, t2AVER, t3AVER).

Next, as shown in STEP 4 of FIG. 8, it is calculated how many times the amount of error obtained from each measurement data is compared with the reference error obtained from the teacher data. We conclude that the larger this magnification is, the more the state changes. FIG. 14 shows, as a comparison main body, “measurement t1 error (average) (t1AVER)”, “measurement t2 error (average) (t2AVER)”, and “measurement t3 error (average) (t3AVER)”. It is assumed that there is a value corresponding to the area of the circle, and there is a value corresponding to the area of the circle as shown in "Standard error (mean) (AVSTER)" as a comparison target. Assuming that the magnification is 1.2, 2.3, and 3.8, (measurement t1 error) <(measurement t2 error) <(measurement t3 error) holds, so measurement 3 has the most state fluctuation. It can be concluded that the state change is occurring, then the state change is occurring in the measurement 2, and the measurement 1 is the least probable that the state change is occurring.

Alternatively, instead of finding the average for the standard error or error, create a distribution chart graph of the standard error value, create a distribution chart graph of the error value, and detect the state change using these distribution chart graphs. .. The distribution chart graph of the standard error value is as shown in FIG. 15 (a), the distribution chart graph of the error value of the measurement t1 is as shown in FIG. 15 (b), and the distribution chart graph of the error value of the measurement t2 is shown. Is as shown in FIG. 15 (c), and the distribution graph of the error value of the measurement t3 is as shown in FIG. 15 (d).

In the above case, the difference between the peaks of the distribution chart is used as the degree of dissociation. The difference between the peak of the standard error value distribution chart graph and the peak of the measurement t1 error value distribution chart graph is as shown in FIG. 15 (b). The difference between the peak of the standard error value distribution chart graph and the peak of the measurement t2 error value distribution chart graph is as shown in FIG. 15 (c). The difference between the peak of the standard error value distribution chart graph and the peak of the measurement t3 error value distribution chart graph is as shown in FIG. 15 (d). Since the difference between the peaks is d ₁ <d ₂ <d ₃ , the determination of the state change is as described with reference to FIG.

Further, as shown in STEP 4 of FIG. 8, an abnormality determination threshold value is set for the magnification obtained above or for the value σ obtained by standardizing the error for each measurement number, so that the abnormality is detected. Is also good. For example, in the example of FIG. 14, if the threshold value is "3.5", it can be determined that the measurement t3 is abnormal.

An example of using the deviation value σ is shown in FIG. It is assumed that the above-mentioned error distribution map is shown in the graph of FIG. The average value of this error is obtained (S21). Using this average value, the deviation value σ is obtained (S22). Next, the state change is detected using the deviation value σ obtained above (S23). For example, since the standard error is also obtained for each measurement, the deviation value σ _{ST of the} standard error can be obtained from these distribution maps. On the other hand, the deviation values σ _t1 , σ _t2 , and σ _t3 are also obtained for the measurements t1, t2, and t3. Of the deviation values σ _t1 , σ _t2 , and σ _t3 , the larger the difference or ratio or magnification of the standard error deviation value σ _ST , the higher the probability of a state change. The threshold value σ _SH for the deviation values σ _t1 , σ _t2 , and σ _t3 can be set, and if it exceeds this, it can be considered that there is a state change (abnormality).

In the above, one objective variable is obtained from two explanatory variables, but it can be applied to a prediction model in which one objective variable is obtained from one or more explanatory variables. Further, the state change is not limited to the change to an abnormality, and can be applied to various state change detections such as an excess state, a shortage state, a low state, and a high state.

In the above embodiment, when the data is measured by the sensors A, B, and C, the sensor A prediction model in which the objective variable is the data obtained by the sensor A and the explanatory variable is the data obtained by the sensors B and C. Was made to configure only one prediction model. However, in such a system that measures data of three or more systems, a plurality of prediction systems can be constructed.

For example, when data is measured by sensors A, B, and C, sensor A predicts that the objective variable is the data measured by sensor A and the explanatory variable is the data measured by sensors B and C. The case where the model is called a model, the objective variable is the data measured by the sensor B, and the explanatory variables are the data measured by the sensors A and C is called the sensor B prediction model, and the objective variable is the data measured by the sensor C. When the explanatory variables are the data measured by the sensors A and B and are referred to as the sensor C prediction model, three prediction models such as the sensor A prediction model, the sensor B prediction model, and the sensor C prediction model are used. Can be created.

It is assumed that the time-series measured values obtained by measuring the data of the sensors A, B, and C are t1, t2, and t3. All of t1, t2, and t3 are composed of three types of data, sensors A, B, and C. In the sensor A prediction model, the measured values t1, t2, and t3 of the sensors B and C can be input to obtain three types of average errors of the sensor A prediction model, A_t1AVER, A_t2AVER, and A_t3AVER. In the sensor B prediction model, the measured values t1, t2, and t3 of the sensors A and C can be input to obtain three types of average errors of the sensor B prediction model, B_t1AVER, B_t2AVER, and B_t3AVER. In the sensor C prediction model, the measured values t1, t2, and t3 of the sensors A and B can be input to obtain three types of average errors of the sensor C prediction model, C_t1AVER, C_t2AVER, and C_t3AVER.

When the reference error of each model obtained from the teacher data is A_AVSTER, B_AVSTER, C_AVSTER, the magnification and error distribution map (deviation) of A_t1AVER, A_t2AVER, and A_t3AVER based on the reference error A_AVSTER are obtained. It is possible to obtain the magnification and error distribution map (deviation) of B_t1AVER, B_t2AVER, and B_t3AVER based on the reference error B_AVERSER, and the magnification of C_t1AVER, C_t2AVER, and C_t3AVER based on the reference error C_AVERSER. The error distribution map (deviation) can also be obtained.

When the deviation is obtained in the above, for example, the deviation value σA_t1 of the measured value t1 in the sensor A prediction model, the deviation value σB_t1 of the measured value t1 in the sensor B prediction model, and the deviation value t1 of the measured value t1 in the sensor C prediction model. By obtaining the average of σC_t1 and the like, a new feature amount of the measured value t1 can be generated, and by setting a threshold value for this, it is possible to detect an abnormality. The above is an example of the measured value t1, but the measured values t2 and t3 can be handled in the same manner.

The state change detection system according to the present embodiment has the above-mentioned configuration, and the first detection result obtained by the first state change detection device that performs the above-mentioned operation, and the configuration and operation are first described. The state change detection is performed using both of the second detection results obtained by the second state change detection device described. Hereinafter, the operation of the state change detection system according to this embodiment will be described with reference to the flowchart of FIG.

Using the first state change detection device configured as described above, the measurement data (test data) that is the target of the state change detection is input to obtain the first prediction result (error amount) (S31). Further, using the second state change detection device, the measurement data (test data) that is the target of the state change detection is input to obtain the second prediction result (ratio) (S32).

Next, in step S33, it is compared with the threshold value TH-A for the prediction result (ratio) obtained in step S32, and compared with the threshold value TH-B for the prediction result (error amount) obtained in step S31 ( S33), when both of them exceed the threshold value, it is determined that there is a state change (abnormality) (S34).

Regarding the determination result, as shown in FIG. 18, the case where only the prediction result (ratio) exceeds the threshold value TH-A and the case where only the prediction result (error amount) exceeds the threshold value TH-B are regarded as "quasi-abnormal". , When the prediction result (ratio) exceeds the threshold value TH-A and the prediction result (error amount) also exceeds the threshold value TH-B, it is regarded as "abnormal", and the predicted value (ratio) is lower than the threshold value TH-A. If the prediction result (error amount) is also below the threshold value TH-B, it can be regarded as "normal".

Here, the first prediction result is used as the error amount, but the average error value or the deviation value σ described above may be used. According to the state change detection device according to this embodiment, it is possible to accurately capture the state change only from the data when there is no state change or it is unknown whether the state change has occurred.

10 CPU 11 Main memory 12 Bus 13 External storage interface 14 Input interface 15 Display interface 16 Data input interface 22 Mouse 23 External storage device 24 Input device 25 Display device 26 Sensor 30 Prediction model 31 Prediction model creation means 32 Teacher Prediction data acquisition means 33 Standard error acquisition means 34 Measurement prediction data acquisition means 35 Error acquisition means 36 State change detection means 51 Distribution map creation means 52 Average calculation means 53 Position information calculation means 54 State change detection means

Claims (12)

The data obtained from a specific 1 data acquisition source in the data acquired from a plurality of data acquisition sources is used as the objective variable, and the data obtained from a plurality of data acquisition sources other than the specific 1 is used as the explanatory variable, and the explanatory variables are used. For each explanatory variable unit included in the teacher data for creating a prediction model for predicting the objective variable, a distribution map creation means for creating a distribution map of the frequency of appearance with respect to the value of the explanatory variable, and a distribution map creation means.
An average value calculation means for calculating the average value of the explanatory variables for each explanatory variable of the test data,
A position information calculation means for obtaining position information indicating the position of the obtained average value on the distribution map of the corresponding explanatory variable unit in the teacher data.
Explanatory variable A state change detection device provided with a state change detection means for statistically processing all position information for each variable unit, obtaining a tendency of the position information of the statistical processing result, and detecting a state change based on this tendency. .. The state change detection means according to claim 1, wherein the state change detecting means detects a state change when a predetermined ratio of position information exists in a region larger or smaller than a predetermined position in the distribution map. Device. The state change detecting means obtains an average value of position information for each explanatory variable unit, and detects state change when the average value exists in a region larger or smaller than a predetermined position in the distribution map. The state change detection device according to claim 1. The state change detecting means is characterized in that the difference value between the previous position information and the current position information is obtained for all the position information for each explanatory variable unit, and the state change is detected based on the number of explanatory variable units whose difference value is equal to or more than a predetermined value. The state change detection device according to claim 1. The claim is characterized in that the state change detecting means detects a state change when even one or a predetermined number of position information units exist in a region where the position information is larger or smaller than a predetermined position in the distribution map. Item 1. The state change detection device according to item 1. The data obtained from one specific data acquisition source among the data acquired from a plurality of data acquisition sources is used as the objective variable, and the data obtained from a plurality of data acquisition sources other than the specific one is used as an explanatory variable to control the state change. The first state change detection device for detecting state change by inputting the data collected for the target to be detected into a prediction model created by machine learning as test data, and the first item according to any one of claims 1 to 5. In the state change detection system composed of the second state change detection device which is the state change detection device of
It is characterized in that state change detection is performed using both the first detection result obtained by the first state change detection device and the second detection result obtained by the second state change detection device. State fluctuation detection system. Computer,
The data obtained from a specific 1 data acquisition source in the data acquired from a plurality of data acquisition sources is used as the objective variable, and the data obtained from a plurality of data acquisition sources other than the specific 1 is used as the explanatory variable, and the explanatory variables are used. Distribution map creation means that creates a distribution map of the frequency of appearance for the value of the explanatory variable for each explanatory variable unit included in the teacher data for creating a prediction model that predicts the objective variable.
An average value calculation method for calculating the average value of explanatory variables in units of explanatory variables of test data,
A position information calculation means for obtaining position information indicating the position of the obtained average value on the distribution map of the corresponding explanatory variable unit in the teacher data.
Explanatory variable A state change detection program characterized in that all position information for each variable unit is statistically processed, the tendency of the position information of the statistical processing result is obtained, and the state change detection means is used to detect the state change based on this tendency. .. A claim, wherein the computer is used as the state change detecting means to function to detect a state change when a predetermined ratio of position information exists in a region larger or smaller than a predetermined position of the distribution map. The program for detecting state fluctuations according to 7. Using the computer as the state change detecting means, the average value of the position information for each explanatory variable unit is obtained, and the state change is detected when the average value exists in a region larger or smaller than a predetermined position in the distribution map. The state change detection program according to claim 7, wherein the program is characterized by the functioning of the program. Using the computer as the state change detecting means, the difference value between the previous position information and the current position information is obtained for all the position information for each explanatory variable unit, and the state change is detected based on the number of explanatory variable units whose difference value is equal to or more than a predetermined value. The state change detection program according to claim 7, wherein the program is made to function as such. Using the computer as the state change detecting means, the function is to detect the state change when even one or a predetermined number of position information units exist in a region where the position information is larger or smaller than the predetermined position in the distribution map. The state change detection program according to claim 7, wherein the program is used. Computer,
The data obtained from one specific data acquisition source among the data acquired from a plurality of data acquisition sources is used as the objective variable, and the data obtained from a plurality of data acquisition sources other than the specific one is used as an explanatory variable to control the state change. A first state change detection device that detects state change by inputting the data collected for the target to be detected as test data into a prediction model created by machine learning.
It functions as a second state change detection device, which is the state change detection device according to any one of claims 1 to 5, and also functions as a second state change detection device.
A first detection result obtained by the computer functioning as the first state change detection device, and a second detection result obtained by the computer functioning as the second state change detection device. A state change detection program characterized in that it functions to perform state change detection using both of the above.

Images