Equipment diagnosis device and equipment diagnosis method

Abstract

To facilitate diagnosis using a plurality of sets of sensor data by performing diagnosis according to evaluation of correlation between the sets of sensor data.SOLUTION: An acceleration sensor 12 and a current sensor 13 are installed in a rotating machine 11, and time series data of any time width on the sensors 12 and 13 are stored in a data measuring device 14 as measurement data. An equipment diagnosis apparatus 10 divides each measurement data into a predetermined number of classes of random variables in a distribution section at regular time intervals and converts a distribution into a probability distribution according to the number of measurement values in each class. The apparatus includes an entropy conversion unit that converts the probability distribution of each sensor at each given time into entropy by using specific formulas (1) and (2); a mutual information conversion unit that calculates a mutual information amount at the same timing from the probability distribution of each sensor by using a specific formula (3), and diagnoses equipment based on correlation changes in the mutual information; and a KLD converting unit that converts the probability distribution of each sensor into a KLD using specific equations (4) and (5); and a monitoring unit that monitors the entropy, the mutual information, and the KLD and diagnoses monitoring data.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a technique for diagnosing equipment based on measurement data from a plurality of sensors installed in the equipment to be diagnosed. [Background technology]

[0002] The failure of infrastructure facilities that support society, such as those for electricity and transportation, can have a significant impact on society, making it necessary to prevent unplanned outages.Failures of production facilities in factories can also have a significant impact on corporate production activities.

[0003] These facilities are deteriorating over time, and there is a need to reduce costs amid a labor shortage due to the declining birthrate and aging population, as well as increased global competition. Therefore, there is a need for unmanned and remote monitoring systems that can detect equipment failures and abnormalities early on.

[0004] Therefore, in order to monitor facilities 24 hours a day, 356 days a year, a diagnostic technology has been proposed in which many sensors are attached to the facilities to be monitored and the diagnostic technology is based on the multidimensional time series signals detected by each sensor.

[0005] For example, Patent Document 1 proposes measuring the vibration of a rotating machine with a vibration sensor and diagnosing it based on the measurement results. Also, Patent Document 2 achieves diagnosis by measuring the current flowing through a grounding wire. According to these diagnoses, signs of an abnormality appear in vibrations and currents, and measuring these makes it possible to detect an abnormality.

[0006] Patent Document 3 proposes that increasing the number of types of state quantities (physical quantities) used to determine anomalies enables more accurate anomaly detection and failure prediction. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Patent Publication No. 2021-144054 [Patent Document 2] Patent Publication No. 2021-50921 [Patent Document 3] Patent Publication No. 2021-177074 [Non-patent literature]

[0008] [Non-Patent Document 1] "Introduction to the world of Divergence (KL-Divergence) and Entropy (Cross-Entropy) used in Machine Learning" [online] Retrieved December 27, 2021, Internet<URL:https: / / qiita.com / harmegiddo / items / 2a24a36418fade0eaf44> [Non-patent document 2] Shota Ako, "Machine Learning and Differential Geometry", Mathematical Sciences, Vol. 56, no. 8, pp. 14-19, Aug. 2018 [Non-patent document 3] S. Kullback and RA Leibler,”On information and sufficiency Ann. Math. Statist”, PP. :79-,1951 [Non-patent document 4] S.Amari and H.Nagaoka, “Methods of Information Geometry” A co-publication of the AMS and Oxford University Press, 2000. [Non-Patent Document 5] Takuya Ajiki, Masami Nakamura, Shunji Maeda, Takashi Komatsu, and Tetsuji Taniguchi, "Proposal of an Anomaly Detection Method Using KLD Duality," 72nd Chugoku Branch Joint Conference of Electrical and Information Engineering Societies, 2021 Summary of the Invention [Problem to be solved by the invention]

[0009] However, when using data from multiple sensors, if the physical quantities measured are different, the characteristics of the data (such as units and statistics) will change, making it difficult to understand the relationships between them and potentially making correlation analysis difficult.

[0010] The present invention aims to solve the problem of facilitating diagnosis using a plurality of sensor data by performing diagnosis according to an evaluation of correlations between the sensor data. [Means for solving the problem]

[0011] (1) One aspect of the present invention is A device that installs a plurality of sensors in a facility to be diagnosed, and diagnoses the facility using time-series data of an arbitrary time width of each of the sensors as measurement data, a probability distribution conversion unit that sets a distribution interval for each fixed time period for the measurement data of each sensor, divides the set distribution interval into a number of classes that indicate a preset random variable, and converts the set distribution interval into a probability distribution according to the number of measurement values ​​of each class; a mutual information conversion unit that calculates mutual information for the same time from the probability distribution of each sensor converted by the probability distribution conversion unit using equation (3);

[0012]

number

[0013] P X (x): Random variable X={x1, x2,...x n Probability distribution of P Y (y): The random variable Y of the other sensor = {y1, y2, ... y n Probability distribution of However, random variable X = random variable Y I(X;Y):P X (x),P Y Mutual information of (y) P XY (x, y): Joint probability (the relative frequency when data from one sensor and data from another sensor are included in each class at the same time) Equipped with The equipment is diagnosed based on a correlation change in the mutual information calculated by the mutual information transform unit.

[0014] (2) Another aspect of the present invention is a method for diagnosing equipment by installing a plurality of sensors in the equipment to be diagnosed and using time-series data of an arbitrary time width of each of the sensors as measurement data, the method comprising: a probability distribution conversion step of setting a distribution interval for each fixed time period for the measurement data of each sensor, dividing the set distribution interval into a number of classes representing a preset random variable, and converting the set distribution interval into a probability distribution according to the number of measurement values ​​of each class; a mutual information conversion step of calculating mutual information for the same time from the probability distribution of each sensor converted in the probability distribution conversion step using equation (3);

[0015]

number

[0016] P X (x): Random variable X={x1, x2,...x n Probability distribution of P Y (y): The random variable Y of the other sensor = {y1, y2, ... y n Probability distribution of However, random variable X = random variable Y I(X;Y):P X (x),P Y Mutual information of (y) P XY (x, y): Joint probability (the relative frequency when data from one sensor and data from another sensor are included in each class at the same time) a diagnosis step of diagnosing the equipment based on a correlation change of the mutual information calculated by the mutual information transform unit; It is characterized by having:

[0017] (3) In each of the above aspects, From the probability distribution, a reference probability distribution P X * Generate (x), The probability distribution P X * The probability distribution P of each sensor obtained at the same interval as (x) X (x),P Y (y) from the probability distribution P X * (x) is the same random variable "X=Y={x1,x2,...x n} is compared with the probability distribution P Y Get (x), The probability distribution P X * (x),P Y (x) to the KLD (Kullback-Leibler divergence / D) in equation (4) KL (X||Y)" and

[0018]

number

[0019] The equipment can also be diagnosed based on whether the transformed KLD is detected as an outlier away from a normal distribution in a scatter plot.

[0020] In this case, the probability distribution P X * (x) and the above P Y (x) and KLD in equation (5) KL (Y||X)" and

[0021]

number

[0022] The equipment may be diagnosed based on whether or not both of the KLDs in equations (4) and (5) are detected as outliers away from the normal distribution in the scatter plot.

[0023] (4) In each of the above aspects, converting the probability distribution of each sensor converted by the probability distribution conversion unit into entropy at any time interval using equations (1) and (2);

[0024]

number

[0025]

number

[0026] H(X): Entropy of one sensor H(Y): Entropy of the other sensor The equipment can also be diagnosed based on whether or not the entropy of each sensor has changed by a threshold value or more. [Effects of the Invention]

[0027] According to the present invention, it is possible to facilitate diagnosis using data from a plurality of sensors. [Brief explanation of the drawings]

[0028] [Figure 1] 1A is a schematic diagram of the diagnostic information amount (entropy) of an equipment diagnostic system according to an embodiment of the present invention, FIG. 1B is a schematic diagram of mutual information, and FIG. 1C is a schematic diagram of "KLD". [Figure 2] FIG. 2 is an overall diagram showing the processing steps of the diagnostic method. [Figure 3] FIG. 2 is a diagram showing a diagnostic status of a target facility according to an embodiment. [Figure 4] FIG. 2 is an overall diagram showing the processing steps of the diagnostic method. [Figure 5] (a) is a sensor data diagram for sensor 12, and (b) is a sensor data diagram for sensor 13. [Figure 6] 10(a) is a diagram of time-series sensor data of the sensor 12 input to the probability distribution conversion unit, and FIG. 10(b) is a diagram of data output from the probability distribution conversion unit after the time-series sensor data of FIG. 10(a) is converted into a probability distribution. [Figure 7](a) is a data diagram showing the probability distribution of the sensor 12 at regular intervals input to the entropy conversion unit, and (b) is a data diagram output from the entropy conversion unit after the probability distribution of (a) is converted to entropy. [Figure 8] (a) is a data diagram showing the probability distribution of sensor 12 input to the mutual information conversion unit, (b) is a data diagram showing the probability distribution of sensor 13, and (c) is a data diagram output from the mutual information conversion unit after the probability distributions of (a) and (b) are converted into mutual information. [Figure 9] (a) is a data diagram of the reference probability distribution (sensor 12) input to the KLD conversion unit, (b) is a data diagram of the probability distribution (sensor 13) to be compared, and (c) is an outlier detection diagram of the output results of the KLD conversion unit obtained from the probability distributions of (a) and (b). DETAILED DESCRIPTION OF THE INVENTION

[0029] An equipment diagnostic device (equipment diagnostic method) according to an embodiment of the present invention will be described below. This diagnostic device converts the phenomenon to be diagnosed into numerical values ​​(data) as physical quantities, and expresses these numerical values ​​as "statistics such as mean and variance." This is converted into the value of information, and the "rarity" of the information is used as an index for diagnosis.

[0030] Specifically, the diagnostic device treats sensor data having various units from multiple sensors attached to the equipment to be diagnosed as a probability distribution based on occurrence frequency, and focuses on the entropy and mutual information calculated from each sensor data to diagnose whether or not there is an abnormality / sign of an abnormality in the equipment to be diagnosed.

[0031] As evaluation methods based on such multiple sensor data, "JSD (Jensen-Shannon Divergence)" in Non-Patent Document 1, "Wasserstein" in Non-Patent Document 2, and KLD "Kullback-Leibler divergence" in Non-Patent Documents 3 to 5 have been proposed.

[0032] In response to this, the diagnostic device combines three types of information (entropy, mutual information, and KLD) to detect abnormalities and signs of abnormalities that could not be detected by "JSD," "Wasserstein," or "KLD" alone.

[0033] In other words, when using multiple sensor data, each sensor data changes in a variety of ways and has different units, making it difficult to handle. Therefore, the diagnostic device converts sensor data of different physical quantities collected from various sensors attached to the equipment to be diagnosed into a probability distribution based on occurrence frequency, and converts it into a common unit such as entropy or mutual information, which indicates the amount of information, thereby unifying the unit system, and further evaluating the correlation between the sensor data under the unified unit system, thereby facilitating diagnosis using multiple sensor data.

[0034] (1)Amount of information An outline of diagnosis based on each amount of information will be explained with reference to FIG.

[0035] A: As shown in Figure 1(a), the entropy (average amount of information) H of each sensor data is calculated at regular intervals. At this time, abnormal conditions in each sensor data are detected based on the temporal changes in each calculated entropy H, and the presence or absence of abnormalities / signs of abnormalities in the target is diagnosed. Here, equipment abnormalities and the like are regarded as changes in entropy H, thereby unifying the unit system.

[0036] B: As shown in Figure 1(b), the mutual information I between sensor data is calculated at regular intervals. Sensor data showing abnormal conditions is detected from the correlation based on the calculated mutual information I, and the presence or absence of abnormalities / signs of abnormalities in the target to be diagnosed is diagnosed. In other words, by calculating the mutual information between sensor data, the equipment is diagnosed based on the dependency between sensor data under normal conditions and the temporal change in the relationship when changing from normal to abnormal. This diagnosis is not affected by physical quantities, so stable detection is possible.

[0037] C: As shown in Figure 1(c), the "KLD" between a pre-recorded reference probability distribution (reference data) and the probability distribution of the object to be diagnosed is calculated at regular intervals. Here, abnormalities in the sensor data are detected from changes in the "KLD," and the presence or absence of abnormalities / signs of abnormalities in the object to be diagnosed is diagnosed. In this case, the reference probability distribution and the probability distribution of the object to be compared do not need to be sensor data from the same sensor.

[0038] (2) Diagnostic Procedure An overall outline of the diagnostic procedure (diagnosis steps) by the diagnostic device will be described with reference to Fig. 2. First, information that digitizes the physical phenomenon of the diagnostic object, that is, sensor data from the sensors 1 and 2 of the diagnostic object, is acquired (S01).

[0039] Next, to express each acquired sensor data as a statistical index, we divide each sensor data into probability distributions, P X (x),P Y (y)” (S02). The converted probability distribution “P X (x),P Y (y)” is expressed as an index of “information value.”

[0040] Here, the probability distribution "P X (x),P Y (y)" Entropy H(X), H(Y) ·Mutual information I(X;Y) ·KLD "Bivariate D KL (X||Y),D KL (Y||X)」 (S03 to S05). After that, the numerical values ​​of each piece of information converted in S03 to S05 are monitored (S06 to S08).

[0041] That is, in S06, "Monitoring entropy H (self-condition diagnosis)," the increase or decrease in the value of entropy H is monitored. As a result of the monitoring, it is determined that "increase in value = increased variance" or "decrease in value = increased deviation in value" and it is determined that an abnormality or a sign of an abnormality has occurred in the object to be diagnosed.

[0042] Furthermore, in S07, "Monitoring the value of mutual information I," the system determines that "increase in value = strong relationship between sensors 1 and 2" and "decrease in value = weak relationship (independence)" for mutual information I, and monitors whether the relationship between sensors 1 and 2 has been disrupted. If the monitoring result shows that the relationship has been disrupted, it is determined that an abnormality / sign of abnormality has occurred in the object to be diagnosed.

[0043] Furthermore, in S08, "KLD Numerical Value Monitoring," it is monitored whether a change in distribution has occurred in the two variables. If a change in distribution has occurred as a result of the monitoring, it is determined that an abnormality or a sign of an abnormality has occurred in the object to be diagnosed. [Example]

[0044] An embodiment of the diagnostic device will be described with reference to Figures 3 to 9. Here, a rotating machine will be described as an example of equipment.

[0045] <Configuration example> 3, reference numeral 11 denotes a rotating machine that is the equipment to be diagnosed. An acceleration sensor 12 is installed near the bearing of this rotating machine 11 to measure bearing vibration (acceleration).

[0046] A current sensor 13 is installed in the housing 11a of the rotating machine 11 to measure the power supply current and leakage current of the rotating machine 11. Measurement data (sensor data) measured by these sensors 12 and 13 is collected and stored in a data measuring device 14 equipped with a storage device (e.g., SSD, HDD, etc.).

[0047] In this example, an equipment diagnosis device 10 is connected to a data measuring instrument 14, and collected and accumulated measurement data is input to the equipment diagnosis device 10. This equipment diagnosis device 10 generates a probability distribution based on the input measurement data and diagnoses the rotating machine 11 using the above-mentioned information quantities (entropy, mutual information, KLD). The diagnosis results can be displayed on a display device (monitor) (not shown) connected to the diagnosis device 10, allowing the user to check them.

[0048] Specifically, the diagnostic device 10 is configured by a computer and includes the usual hardware resources of a computer (for example, a CPU, storage devices such as RAM and ROM, etc.).

[0049] As a result of the cooperation of these hardware resources and software resources (OS, applications, etc.), the diagnostic device 10 is equipped with a probability distribution conversion unit, entropy conversion unit, mutual information conversion unit, KLD conversion unit, and monitoring unit (diagnosis unit), all of which are not shown.

[0050] <Diagnostic Procedure> The diagnostic procedure (diagnostic steps) of the embodiment will be described with reference to FIG.

[0051] S11: The sensors 12 and 13 continuously measure physical information (acceleration and current value) of the rotating machine 11 to be diagnosed. The sensor data measured here is input to the data measuring instrument 14.

[0052] At this time, the data measuring instrument 14 periodically records each captured sensor data as time-series data of an arbitrary time width. Figure 5 shows an example of time-series data from sensors 12 and 13, which is recorded with one measurement per second. In this example, the sampling frequency of the time-series data from sensors 12 and 13 is 51,200 Hz, and 51,200 samples of data are recorded per second.

[0053] As a result, time-series data for diagnosis is accumulated every second in the data measuring instrument 14, and the accumulated time-series data is output to the diagnostic device 10. Note that D1 in Fig. 5 indicates an abnormality (100 Hz contamination) in the time-series data (4.0 to 5.0) of the sensor 12.

[0054] S12: The time series data of the sensors 12 and 13 accumulated in the data measuring instrument 14 is input to the diagnostic device 10. Based on the input time series data, the probability distribution conversion unit generates (calculates) a probability distribution P for each fixed time period. At this time, the following parameters are prepared:

[0055] (1) Distribution interval [a, b] The distribution interval [a, b] indicates the value that determines the range of the probability distribution. It is desirable to set this range wider than the range from the maximum value to the minimum value of the measurement value. For example, in the time-series sensor data of Figure 6(a), the measurement value is included in "±0.3", so the distribution interval [a, b] = [-0.4, 0.4] (±0.4).

[0056] (2) Number of classes "n" The number of classes "n" indicates a value that determines the resolution when dividing the measurement values ​​into several divisions.

[0057] The distribution interval [a, b] is divided into subintervals (hereafter referred to as "classes") by the number of classes "n". For example, if the distribution interval is [-0.4, 0.4] and the number of classes "n" = 40, the series of classes will be 40 divisions: {[-0.4, -0.38], [-0.38, -0.36}, ..., [+0.38, +0.36].

[0058] Then, the number of measurement values ​​included in each class (hereinafter referred to as frequency) is calculated. Taking the measurement values ​​of the sensor 12 as an example, the classes are expressed as "X={x1, x2, ..., x n}" and set it as a random variable. At this time, the probability distribution P X is expressed as the value obtained by dividing the frequency of each class by the number of sample data.

[0059] For example, if the order of frequencies is {0,0,0,1,6,15,33,72,...,0} and the number of sample data is "51200", then P X (x)={0,0,0,0,0.0001,0.0003,0.0006,0.0014,0.0031,0.0055,...,0}, and the probability distribution P X (x) is obtained.

[0060] By applying the same processing to the measurement value of the sensor 13, the probability distribution P Y (y) is obtained. The process of S12 is executed for the sensor data of each of the sensors 12 and 13 every time the process of S11 is executed.

[0061] As mentioned above, the sensor data is collected with one measurement per second, so the latest probability distribution P X (x),P Y (y) is calculated. Here, the calculated probability distribution P X (x),P Y (y) is stored in a storage device (HDD, SSD, etc.) and is output to the entropy conversion unit, mutual information conversion unit, and KLD conversion unit.

[0062] S13: The entropy conversion unit calculates the probability distribution P X (x),P Y Every time (y) is input, the entropy H is calculated for the sensor data of each of the sensors 12 and 13 at regular intervals.

[0063] The probability distribution P of the sensor 12 shown in FIG. X When the entropy H(X) is calculated for (x), the output shown in Figure 7(b) is obtained. In this case, the random variable X = {x1, x2, ..., x n} is a probability distribution P X (x), the entropy H(X) can be calculated from equation (1).

[0064]

number

[0065] Similarly, the probability distribution P of sensor 13 Y Calculate the entropy H(Y) for (y). Here, the random variable Y={y1, y2, ..., y n} is a probability distribution P Y (y), the entropy H(Y) can be calculated from equation (2).

[0066]

number

[0067] However, the entropy H is calculated using the same random variable in the sensor data of the sensors 12 and 13 (random variable X=random variable Y). In this respect, the unit system can be unified.

[0068] S14: The monitoring unit monitors the changes in the entropies H(X) and H(Y) calculated in S13, and diagnoses whether there is an abnormality or a sign of an abnormality each time the entropies H(X) and H(Y) are calculated. This diagnosis is performed for each entropy H(X) and H(Y) based on a preset threshold. This threshold is determined based on an empirically obtained range of change.

[0069] D3 (entropy H(X) after 5 seconds) in Figure 7(b) shows a state in which the entropy H(X) calculated from the probability distribution D2 exceeds the threshold. In this case, the entropy H(X) increases after 4 seconds, and after 5 seconds, it shows an abnormal value and exceeds the threshold. In this case, the rotating machine 11 is diagnosed as having an abnormality / sign of abnormality, and the diagnosis result is displayed on the display device. Note that the threshold can also be set in stages for each abnormality / sign of abnormality, and each abnormality / sign of abnormality can be displayed on the display device.

[0070] S15: The mutual information converter converts the probability distribution P X (x),P Y Each time (y) is input, the probability distribution P X (x),P Y (y) calculates the mutual information I for the same time. For example, the probability distribution P X (x),P Y If we calculate the mutual information I per second from (y), we get the output result shown in Figure 8(c).

[0071] where the random variable X={x1,x2,...,x n}, random variable Y={y1,y2,...,y n} is the probability distribution of P X (x),P Y (y), the mutual information I(X;Y) can be calculated from equation (3).

[0072]

number

[0073] P in equation (3) XY (x, y) represents the joint probability, which indicates the relative frequency when the sensor data from sensors 12 and 13 are included in each class at the same time.

[0074] For example, let X = Y, and the joint probability P XY Let's assume that (x1, y1) is to be calculated. In this case, if the two measured values ​​are "-0.39" and "-0.385" respectively at time "t=0.1" seconds, the frequency is "1", and the value divided by the number of sample data is the joint probability P XY The result is (x1,y1).

[0075] S16: The monitoring unit monitors the change in the value of the mutual information I(X;Y) calculated in S15, and diagnoses whether there is an abnormality / sign of an abnormality each time the mutual information I(X;Y) is calculated.

[0076] Here, if the change in the value of the mutual information I(X;Y) exceeds a preset threshold, it is determined that the correlation between the sensor data of the sensors 12 and 13 has collapsed and an abnormality has occurred in one of the sensor data. This threshold is also determined based on an empirically obtained change range.

[0077] For example, according to the output result in Figure 8(c), when the mutual information I(X;Y) is calculated with the random variable X = Y, the mutual information I does not change significantly up to 4 seconds, but the value starts to increase after 4 seconds.

[0078] In this case, as shown by D5 in Fig. 8(c), a change exceeding the threshold value was detected after 5 seconds, and therefore an abnormality was detected in one of the sensor data. Based on this detection, the rotating machine 11 is diagnosed as having an abnormality or a sign of abnormality, and the diagnosis result is displayed on the display device. Note that the threshold value can also be set in stages for each abnormality / sign of abnormality, and each abnormality / sign of abnormality can be displayed on the display device.

[0079] S17: The KLD conversion unit evaluates the sensor data of the sensors 12 and 13 using the KLDs of Non-Patent Documents 3 to 5. Specifically, the KLD conversion unit generates a reference probability distribution from the past probability distributions stored in advance in the data measurement device 14, records this, and uses it as an input. Here, as shown in FIG. 9(a), the reference probability distribution P X * The case where (x) is generated will be explained.

[0080] In this case, as shown in Figure 9(b), the probability distribution P X * Similar to (x), the probability distribution P Y (x) is obtained. At this time, the probability distribution P Y (x) is not only the probability distribution of different sensors, but also the probability distribution P X * The probability distribution of the same sensor as (x) is also used.

[0081] Then, the obtained probability distribution P Y (x) and the reference probability distribution P X * (x) and the KLD are calculated. Figure 9(c) shows the probability distribution P X * As a probability distribution of a sensor different from (x), the probability distribution P Y This represents the state in which (x) is compared.

[0082] At this time P Y (x) is P X * The same random variable as (x), that is, the random variable "X=Y={x1,x2,...x n}={y1,y2,...y n}" is the probability distribution of the sensor 13. Therefore, P Y (x) is P of S12 and S13 Y (y), and the probability distribution P of the sensor 13 output from the probability distribution conversion unit Y (x) can be used as it is as an input. Then, first, we use equation (4) to calculate "KLD(D KL(X||Y)".

[0083]

number

[0084] Next, the KLD has asymmetry and shows different values. This allows us to define the set of probability distributions P X * (x),P Y Two variables can be obtained from (x). The JSD of Non-Patent Document 1, which attempts to offset the asymmetry of KLD, can only obtain one numerical value from a set of probability distributions, and diagnosis can be made with relatively little information compared to KLD. In this embodiment, by utilizing the asymmetry of KLD for diagnosis, diagnosis based on abundant information becomes possible. Furthermore, diagnosis can be made based on abundant information compared to the "Wasserstein" distance of Non-Patent Document 2. Here, we use the "KLD(D KL (Y||X)" is calculated using equation (5).

[0085]

number

[0086] This results in the output of Figure 9(c), i.e., the bivariate "D KL (X||Y),D KL (Y||X)" is obtained. Also, P X * As the probability distribution of the same sensor as (x), the probability distribution P X (x) is compared with the probability distribution P Y Replace with (x) and perform the same process.

[0087] In this embodiment, the sensor 12 is used as an example and the KLD is calculated based on the probability distribution after one second shown in FIG. 9(a). However, the sensor 13 can also be used as the reference probability distribution. In this case, the reference probability distribution generated from the past probability distribution of the sensor 13 is used as P X *(x), and the probability distribution of the sensors 12 and 13 is compared with the probability distribution P Y Just read it as (x).

[0088] S18: The monitoring unit monitors each of the two variables calculated in S17, and diagnoses the rotating machine 11 based on deviations of each of the two variables from a normal distribution. The normal distribution is determined in advance from the values ​​of the two variables by executing S11, S12, and S17 while the rotating machine 11 is operating normally.

[0089] In FIG. 9(c), D6 indicates a normal distribution, and D7 indicates the bivariate. Here, the bivariate D7 is detected as an outlier from the normal distribution D6 and is therefore determined to be in an abnormal state. As a result, the rotating machine 11 is diagnosed as having an abnormality or a sign of an abnormality, and the diagnosis result is displayed on a display device. This outlier detection can be performed based on whether or not the deviation from the normal distribution is equal to or greater than a threshold, and the threshold can be set in stages for each abnormality or sign of an abnormality.

[0090] This type of equipment diagnosis enables diagnosis based on changes in entropy caused by the same random variable between sensor data collected by sensors 12 and 13, and further enables correlation evaluation by detecting outliers in the amount of change in mutual information and KLD. This makes it possible to facilitate diagnosis based on multiple sensor data.

[0091] The present invention is not limited to the above-described embodiment, and can be modified and implemented within the scope of the claims. For example, the sensors are not limited to the acceleration sensor 12 and the current sensor 13, and other sensors (level sensors, flow meters, etc.) may be used. Furthermore, it is not necessary to perform a diagnosis for each sensor using entropy H, and diagnosis without using the asymmetry of KLD is also possible. [Explanation of symbols]

[0092] 9...Facility diagnostic system 10...Facility diagnostic equipment 11...Rotating machine (equipment) 12...Acceleration sensor 13...Current sensor 14...Data measuring instrument

Claims

1. A device that installs a plurality of sensors in a facility to be diagnosed, and diagnoses the facility using time-series data of an arbitrary time width of each of the sensors as measurement data, a probability distribution conversion unit that sets a distribution interval for each fixed time period for the measurement data of each sensor, divides the set distribution interval into a number of classes that indicate a preset random variable, and converts the set distribution interval into a probability distribution according to the number of measurement values ​​of each class; a mutual information conversion unit that calculates mutual information for the same time from the probability distribution of each sensor converted by the probability distribution conversion unit using equation (3); [Equation 3] P Ｘ (x): Random variable of one sensor X = {x １ , x ２ ,... x ｎ } probability distribution P Ｙ (y): The random variable of the other sensor Y = {y １ , y ２ ,... .y ｎ } probability distribution However, random variable X = random variable Y I(X;Y):P Ｘ (x), P Ｙ Mutual information of (y) P ＸＹ (x, y): Joint probability (the relative frequency when data from one sensor and data from another sensor are included in each class at the same time) Equipped with an equipment diagnosis device that determines a correlation between data from one of the sensors and data from the other of the sensors based on a change in a value of mutual information calculated by the mutual information transform unit, and diagnoses the equipment in accordance with a result of the determination.

2. A device that installs a plurality of sensors in a facility to be diagnosed, and diagnoses the facility using time-series data of an arbitrary time width of each of the sensors as measurement data, a probability distribution conversion unit that sets a distribution interval for each fixed time period for the measurement data of each sensor, divides the set distribution interval into a number of classes that indicate a preset random variable, and converts the set distribution interval into a probability distribution according to the number of measurement values ​​of each class; From the probability distribution, a reference probability distribution P Ｘ ＊ (x), The probability distribution P Ｘ ＊ The probability distribution P of each sensor obtained at the same interval as (x) Ｘ (x), P Ｙ (y) to the probability distribution P Ｘ ＊ (x) is the same random variable "X = Y = {x １ , x ２ ,... x ｎ } is compared with the probability distribution P Ｙ (x) is obtained, The probability distribution P Ｘ ＊ (x), P Ｙ (x) to the KLD of equation (4) "Kullback-Leibler divergence / D ＫＬ a KLD conversion unit that converts the input data into a KLD data (X||Y) Equipped with [Equation 4] Depending on whether the transformed KLD is detected as an outlier away from the normal distribution in the scatter plot, An equipment diagnosis device for diagnosing the equipment.

3. The KLD conversion unit further Ｘ ＊ (x) and the P Ｙ (x) and replace it with the KLD "D" in equation (5). ＫＬ (Y||X)" while [Equation 5] Depending on whether the KLDs of equations (4) and (5) are both detected as outliers away from the normal distribution in the scatter plot, 3. An equipment diagnosis device according to claim 2, wherein said equipment is diagnosed.

4. an entropy conversion unit that converts the probability distribution of each sensor converted by the probability distribution conversion unit into an entropy for each arbitrary time period using equations (1) and (2); [Equation 1] [Equation 2] H(X): Entropy of one sensor H(Y): Entropy of the other sensor Either one of the correlation determination described in claim 1 and the outlier detection described in claim 2; a threshold determination as to whether the entropy of each sensor has changed by more than a threshold; 3. The equipment diagnosis device according to claim 1, wherein the equipment is diagnosed by:

5. A device that installs a plurality of sensors in a facility to be diagnosed, and diagnoses the facility using time-series data of an arbitrary time width of each of the sensors as measurement data, a probability distribution conversion unit that sets a distribution interval for each fixed time period for the measurement data of each sensor, divides the set distribution interval into a number of classes that indicate a preset random variable, and converts the set distribution interval into a probability distribution according to the number of measurement values ​​of each class; The system comprises a mutual information converter and a KLD converter according to claims 1 and 2, The correlation determination according to claim 1; The outlier detection according to claim 2; An equipment diagnosis device characterized by diagnosing the equipment by

6. A method executed by a device that installs a plurality of sensors in a facility to be diagnosed and diagnoses the facility using time-series data of an arbitrary time width of each of the sensors as measurement data, comprising: a probability distribution conversion step of setting a distribution interval for each fixed time period for the measurement data of each sensor, dividing the set distribution interval into a number of classes representing a preset random variable, and converting the measurement data into a probability distribution according to the number of measurement values ​​of each class; a mutual information conversion step of calculating mutual information for the same time from the probability distribution of each sensor converted in the probability distribution conversion step using equation (3); [Equation 3] P Ｘ (x): Random variable of one sensor X = {x １ , x ２ ,... x ｎ } probability distribution P Ｙ (y): The random variable of the other sensor Y = {y １ , y ２ ,... .y ｎ } probability distribution However, random variable X = random variable Y I(X;Y):P Ｘ (x), P Ｙ Mutual information of (y) P ＸＹ (x, y): Joint probability (the relative frequency when data from one sensor and data from another sensor are included in each class at the same time) a diagnosis step of determining a correlation between data from one of the sensors and data from the other sensor based on a change in the value of the mutual information calculated in the mutual information conversion step, and diagnosing the equipment in accordance with the result of the determination; An equipment diagnosis method comprising:

7. A method executed by a device that installs a plurality of sensors in a facility to be diagnosed and diagnoses the facility using time-series data of an arbitrary time width of each of the sensors as measurement data, comprising: a probability distribution conversion step of setting a distribution interval for each fixed time period for the measurement data of each sensor, dividing the set distribution interval into a number of classes representing a preset random variable, and converting the measurement data into a probability distribution according to the number of measurement values ​​of each class; From the probability distribution, a reference probability distribution P Ｘ ＊ (x), The probability distribution P Ｘ ＊ The probability distribution P of each sensor obtained at the same interval as (x) Ｘ (x), P Ｙ (y) to the probability distribution P Ｘ ＊ (x) is the same random variable "X = Y = {x １ , x ２ ,... x ｎ } is compared with the probability distribution P Ｙ (x) is obtained, The probability distribution P Ｘ ＊ (x), P Ｙ (x) to the KLD of equation (4) "Kullback-Leibler divergence / D ＫＬ a KLD transformation step of transforming (X||Y) into [Equation 4] a diagnosis step of diagnosing the equipment based on whether the KLD converted in the KLD conversion step is detected as an outlier away from a normal distribution in a scatter plot; An equipment diagnosis method comprising:

8. In the KLD conversion step, the probability distribution P Ｘ ＊ (x) and the P Ｙ (x) and replace it with the KLD "D" in equation (5). ＫＬ (Y||X)" while [Equation 5] In the diagnosis step, depending on whether or not both of the KLDs of the formulas (4) and (5) are detected as outliers away from the normal distribution in the scatter plot, 8. The equipment diagnosis method according to claim 7, further comprising diagnosing the equipment.

9. an entropy conversion step of converting the probability distribution of each sensor converted in the probability distribution conversion step into an entropy for each arbitrary time period using equations (1) and (2); Furthermore, [Equation 1] [Equation 2] H(X): Entropy of one sensor H(Y): Entropy of the other sensor The diagnostic step includes: Either one of the correlation determination described in claim 6 and the outlier detection described in claim 7; a threshold determination as to whether the entropy of each sensor has changed by more than a threshold; 8. The equipment diagnosis method according to claim 6, wherein the equipment is diagnosed by:

10. A method executed by a device that installs a plurality of sensors in a facility to be diagnosed and diagnoses the facility using time-series data of an arbitrary time width of each of the sensors as measurement data, comprising: a probability distribution conversion step of setting a distribution interval for each fixed time period for the measurement data of each sensor, dividing the set distribution interval into a number of classes representing a preset random variable, and converting the set distribution interval into a probability distribution according to the number of measurement values ​​of each class; Execute the mutual information conversion step and the KLD conversion step according to claims 6 and 7, The correlation determination according to claim 6. The outlier detection according to claim 7; An equipment diagnosis method comprising: diagnosing the equipment by

Images