JP2023068781A - Abnormality detection device, abnormality detection system, and abnormality detection method - Google Patents

Abstract

To improve rapidness, detection precision, and reliability of detection of abnormality in an object system based on a measurement result of physical quantity measured in the object system.SOLUTION: An abnormality detection device for detecting abnormality in an object system on the basis of physical quantity measured in the object system includes: a read part for reading time-series quantization data of physical quantity measured in the object system, and converting the read quantization data to time-series data that can be arithmetically processed; an analysis part for analyzing a predetermined number of most recent time-series data from among time-series data, and calculating a statistic based on the predetermined number of most recent time-series data and an error in the statistic; and a detection part for detecting abnormality in an object system on the basis of a statistic when the degree of dispersion of the statistic calculated by the analysis part satisfies a predetermined condition.SELECTED DRAWING: Figure 1

Description

The present invention relates to an abnormality detection device, an abnormality detection system, and an abnormality detection method.

2. Description of the Related Art Conventionally, an abnormality in a system is detected by measuring a physical quantity for evaluating the normality of the system with a measuring instrument and comparing the measured value with a predetermined reference value.

For example, in a high-voltage electric power system, an overload state of a system load or an abnormality such as a short circuit in an electric circuit is detected, and a circuit breaker or the like is operated to protect the system. For this purpose, the load current flowing through the circuit is measured by a current measuring instrument, the measured signal is input to the protection relay, and when the current based on the electrical signal output from the current measuring instrument exceeds a predetermined reference value, , to detect faults such as overloads and short circuits.

In order to protect the feeder system from overload currents that exceed the allowable load and excessive currents that occur during short circuits, the current flowing through the feeder system is constantly measured by a current measuring instrument in DC railway substation equipment, and the overload current and It detects system abnormalities such as short-circuit current and controls the DC circuit breaker.

The overload current in the DC feeding circuit is characterized by a gradual increase that exceeds the allowable current of the load. It is characterized in that the flowing current increases in a short time. In order to reduce the load on the system by detecting and breaking the short-circuit current early, the DC circuit breaker may use a DC selective breaking device that separates and breaks the short-circuit current and the overload current.

For example, there is a rate-of-current rate-of-change type direct-current selective breaker that sorts out a short-circuit current and an overload current based on the time rate of change (di/dt) of the feeding current. A current rate-of-change type direct-current selective cut-off device receives a measurement signal measured by a current measuring device. The current change rate type DC selective cutoff device operates at a current smaller than the specified current for overload current cutoff when di/dt based on the measurement signal is larger than the reference value, and when it is below the reference value operates at a current greater than the specified current for overload current interruption.

In Patent Document 1, a current I(t) flowing in a feeder line is measured by a current measuring instrument, and the measured signal is digitized and input to a microcomputer. Obtain the transfer function of the current I(t) and its variation ΔI, calculate ΔI from the digitized I(t) so as to satisfy the transfer function, and calculate the current I(t) if ΔI is greater than the reference value ΔIr block the Since the characteristic of ΔI corresponding to di/dt can be derived from the transfer function, it becomes possible to selectively cut off the overload current and the short-circuit current.

JP-A-63-53133

In the method of measuring a physical quantity for judging the normality of a system, such as a protective relay in a high-voltage electric power system, and detecting an abnormality when the measured value exceeds a predetermined reference value, When noise with a high level value is input to the abnormality detection circuit, there is a possibility of erroneously detecting an abnormality.

In the method of detecting an abnormality by measuring the time change of a physical quantity such as an electric current, as in Patent Document 1, the differentiated signal becomes large when high-frequency noise occurs, increasing the possibility of erroneously detecting an abnormality. In the anomaly detection system, it is required to detect noise and anomaly separately without erroneously detecting anomalies when noise occurs. At the same time, it is required to shorten the delay time from occurrence to detection of system anomalies that should be detected early, such as short-circuit currents in DC feeding systems.

The present invention has been made in consideration of the above, and aims to improve the promptness, detection accuracy, and reliability of abnormality detection in a target system based on the measurement results of physical quantities measured in the target system. aim.

In order to solve the above-described problems, according to one aspect of the present invention, there is provided an anomaly detection apparatus for detecting an anomaly in a target system based on a physical quantity measured in the target system, wherein the physical quantity measured in the target system is a reading unit that reads serial quantized data and converts the read quantized data into arithmetically processable time series data; an analysis unit that calculates a statistic based on the latest predetermined number of time-series data and an error in the statistic; and a detection unit that detects an abnormality in the target system based on the above.

According to one aspect of the present invention, for example, it is possible to improve the promptness, detection accuracy, and reliability of anomaly detection in a target system based on measurement results of physical quantities measured in the target system.

The figure which shows the structure of the abnormality detection apparatus which concerns on embodiment. 4 is a timing chart showing the timings of the output of a measuring instrument, AD conversion, digital signal reading, analysis, and the start and completion of anomaly detection in the anomaly detection device according to the embodiment; FIG. 4 is a diagram for explaining a method of calculating an average and standard deviation of time-series data by an analysis unit according to the embodiment; The figure which shows the calculation result of the average, standard deviation, and error of the time-series data of a physical quantity. FIG. 4 is a diagram for explaining an abnormality evaluation method (Part 1) based on the average and standard deviation of time-series data of physical quantities; The figure for demonstrating the abnormality evaluation method (part 2) based on the average and standard deviation of the time-series data of a physical quantity. The figure for demonstrating the abnormality evaluation method (the 3) based on the average and standard deviation of the time-series data of a physical quantity. FIG. 4 is a diagram for explaining a method of calculating a time rate of change of time-series data of a physical quantity and its standard deviation; The figure which shows the calculation result of the standard deviation of the time change rate of the time-series data of a physical quantity, and a time change rate. The figure for demonstrating the abnormality evaluation method based on the time change rate and standard deviation of the time-series data of a physical quantity. FIG. 5 is a diagram for explaining an abnormality evaluation method using the average and standard deviation of time-series data of physical quantities, and the time change rate and standard deviation of time-series data of physical quantities; 1 is a diagram showing the overall configuration of a DC feeding circuit according to Embodiment 1; FIG. 4 is a diagram showing the configuration of a fault current detection circuit according to the first embodiment; FIG. 4A and 4B are diagrams showing aspects of overload current and short-circuit current in the DC feeding circuit according to the first embodiment; FIG. 4A and 4B are diagrams for explaining a method of detecting an overload current and a short-circuit current according to the first embodiment; FIG. 4A and 4B are diagrams for explaining a method of detecting an overload current and a short-circuit current according to the first embodiment; FIG. The figure which shows the one-dimensional regression-analysis result of simulation of the noise by damping vibration. The figure which shows the one-dimensional regression-analysis result of simulation of a short circuit current. The figure which shows the comparison result of the simulation of the noise by damping vibration, and a short-circuit current. FIG. 8 is a diagram showing the configuration of a fault current detection circuit according to a second embodiment; FIG. 8 is a diagram showing the configuration of the DC feeding circuit according to the second embodiment and the state when a short-circuit accident occurs; The figure which shows the structure of the capacitor charging/discharging circuit using the capacitor capacity inspection system which concerns on Example 3. FIG.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments and examples of the present invention will be described with reference to the drawings. It should be noted that the embodiments and examples described below, including the drawings, are merely examples, and do not limit the invention according to the claims. Also, not all of the elements and their combinations described in the embodiments and examples are essential to the solution of the invention. Also, the illustration and description of the well-known configuration that is essential for the configuration of the invention may be omitted.

In the following description, in the embodiments or examples described later, the same reference numerals are given to the same configurations as in the embodiments or examples described above, and the description thereof is omitted, and the description will focus on the differences.

In the following description, a processor that executes a program or hardware logic that implements functions equivalent to the program performs predetermined processing while appropriately using a storage unit and/or an interface, etc., and implements various functional units.

A "processor" is one or more. The processor is typically a microprocessor such as a CPU (Central Processing Unit), but may be another type of processor such as a GPU (Graphics Processing Unit). Also, the processor may be single-core or multi-core. The processor may also be a broadly defined processor such as a hardware circuit (for example, FPGA (Field-Programmable Gate Array) or ASIC (Application Specific Integrated Circuit)) that performs part or all of the processing.

In the following description, for example, the notation "y" is equivalent to the notation in which "y" is written directly above y.

In the following description, the equal sign in determining the magnitude relationship between the determination target and the threshold may be included in any range of magnitude. For example, whether "B if A≧Th, C if A<Th" or "B if A>Th, and B if A≦Th" for the threshold value Th can be changed as appropriate. be.

[Embodiment 1]
FIG. 1 is a diagram showing the configuration of an abnormality detection device S according to an embodiment. An abnormality detection device S in this embodiment includes a measuring instrument 3 , an AD (Analog-Digital) conversion section 4 , a digital signal reading section 5 , an analysis section 6 and a detection section 7 .

The measuring instrument 3 measures a physical quantity 2 measured in the system 1 to be measured in order to evaluate the normality or abnormality of the system 1 to be detected. The AD converter 4 converts an analog signal, which is a measurement signal measured by the measuring instrument 3, into a digital signal (quantized data). The digital signal reading unit 5 reads the digital signal converted by the AD conversion unit 4 and converts it into data that can be processed.

The analysis unit 6 analyzes the time-series data including the latest data among the data converted from the digital signal read by the digital signal reading unit 5 based on a statistical method, and calculates the average, time change rate, etc. of the time-series data. Calculate the estimated value and the error between the estimated values and the degree of dispersion of time-series data (standard deviation, etc.). The detection unit 7 detects an abnormality in the system 1 based on the error between the estimated values calculated by the analysis unit 6 and the degree of variation in the time-series data.

The digital signal reading unit 5, analysis unit 6, and detection unit 7 may be configured by a processor such as a microcomputer or FPGA as an abnormality detection device. The anomaly detection device may be composed of one processor, or may be composed of a plurality of processors. When the abnormality detection device is composed of a plurality of processors, the digital signal reading section 5, the analysis section 6, and the detection section 7 are appropriately integrated and distributed and arranged in each processor. Alternatively, the abnormality detection system may be configured by including the abnormality detection device and the AD conversion section 4 . The anomaly detection system may further include a meter 3 .

FIG. 2 is a timing chart showing the timings of the output of the measuring instrument 3, AD conversion, digital signal reading, analysis, and the start and completion of the operation of abnormality detection in the abnormality detection device S according to the embodiment.

Short-circuit current and overload current detection will be described as an example when the system 1 to be detected for abnormality is an electric power system or a DC feeding system. In this case, the physical quantity 2 is, for example, the current flowing through the electric circuit. The measuring device 3 is a current transformer for current measurement, a Rogowski coil, a Hall CT (Current Transformer), or the like. The digital signal reading unit 5, the analysis unit 6, and the detection unit 7 are processors capable of arithmetic processing and other processing of digital data.

In FIG. 2, the rising edge of the pulse is the start of execution and the falling edge is the completion of the operation. Since the output 8 of the measuring instrument 3 is continuously input to the AD converter 4, it is always ON. Since the operation 9 of the AD conversion unit converts the analog signal into the digital signal at the predetermined period T that can ensure the real-time property, the operation start and operation completion of the conversion processing at the predetermined period T are repeated.

Each operation of the digital signal reading unit 5, the analysis unit 6, and the detection unit 7 (operations 10, 11, and 12, respectively) is performed until the AD conversion unit 4 starts the next operation (that is, the input of the next digital signal). to ). Operation 10 of the digital signal reading unit 5 is substantially synchronized with operation 9 of the AD conversion unit. Action 11 of analysis unit 6 begins substantially at the same time as action 10 is completed. The operation 12 of the detection unit 7 starts substantially at the same time when the operation 10 is completed, and is completed before the next operation 9 starts. The total operation time of the operations 10, 11, and 12 in one cycle is determined within the predetermined cycle T so that the digital signal reading unit 5, the analysis unit 6, and the detection unit 7 repeat the start and completion at the predetermined cycle T. It is In this way, anomalies are always evaluated in real time in consideration of the latest data, so early anomaly detection is possible.

Analysis of the abnormality detection device S in which the time-series data of the digital signal is analyzed based on a statistical method to calculate the estimated values such as the average and the rate of change over time, the error in the estimated values, and the degree of variation in the time-series data. The details of the unit 6 and the detection unit 7 that evaluates and detects an abnormality in the system 1 based on the calculation results of the analysis unit 6 will be described.

FIG. 3 is a diagram for explaining how the analysis unit 6 according to the embodiment calculates the average y − and the standard deviation σy of the time-series data. In FIG. 3 , the horizontal axis is time x and the vertical axis is data y of physical quantity 2 . For example, if the physical quantity 2 is current, the unit is amperes (A).

For the time-series data of the physical quantity 2, the average y of N data and the degree of variation, that is, the standard deviation σy are calculated. In FIG. 3, the number of data N = 3 is shown for the sake of simplicity. calculate. In the second calculation, x=[x2, x3, x4], y=[y2, y3, y4]. In the third calculation, x=[x3, x4, x5], y=[y3, y4, y5. ] to calculate the mean and standard deviation.

In this way, each time new data is input, the oldest data among the data used in the previous calculation is discarded by FIFO (First In First Out), and new data including the new data is generated. calculate. For example, in the second calculation where y4 is input, discard the oldest data x1, y1, include the latest data x4, y4, x=[x2, x3, x4], y=[y2, y3, y4 ] to find the average and standard deviation. Repeat the mean and standard deviation calculations for new x and y combinations each time data is entered. The mean y and standard deviation σy are calculated from equations (1) and (2).

where y is the average of y, j is the number of calculations, and σy is the standard deviation of y. Although the standard deviation is calculated using the unbiased variance in Equation (2), the normal variance obtained by replacing N−1 of 1/(N−1) with N may be used. Since j is the number of calculations, for example, in the explanation using FIG. 3, N=3. Calculate the mean and standard deviation of the data (combination of x=[x2, x3, x4], y=[y2, y3, y4]). By calculating the average y and the standard deviation σy, it is possible to evaluate how much the data varies from the average. If the standard deviation of the time-series data obtained is larger than 1% of the average, it is possible to estimate that noise has occurred. Detection accuracy can be improved by using the standard deviation as an index for abnormality detection in this way. Alternatively, the standard deviation may be substituted with the expression (3) as an index for abnormality detection.

In the above description, the standard deviation σy is used as an index of abnormality detection, but the standard deviation σy may be used as an index of error. The standard deviation σy can be calculated by Equation (4). In FIG. 4 and subsequent figures, the method of evaluating abnormality will be explained using the standard deviation σy, but the standard deviation σy may be substituted.

FIG. 4 is a diagram showing calculation results of the average y − , the standard deviation σy, and the error y error of the time-series data of the physical quantity 2. In FIG. The time x is incremented by 1, and the physical quantity data y changes between 2 and 4. As in the example of FIG. 3, the number of data is set to N=3. For example, as shown in the row of i=1 in the table of FIG. 3, the average y of x=[x1, x2, x3] and y=[y1, y2, y3] is 3.0, and the standard deviation σy is 0.0. 12, with an error of 4.1%. Also, as shown in the row of i=5, the average y of x=[x5, x6, x7] and y=[y5, y6, y7] is 3.2, the standard deviation σy is 0.41, and the error is 13. 1%, and the standard deviation σy and the error y error are larger than in the case of x=[x1, x2, x3] and y=[y1, y2, y3]. This is because the data range for y=[y5, y6, y7] is 2.6-3.6, compared to the data range 2.9-3.2 for y=[y1, y2, y3]. The reason for this is that the variation is large. In this way, the standard deviation σy and error y error increase in accordance with the variation in data.

FIG. 5 is a diagram for explaining an abnormality evaluation method (part 1) based on the average y and the standard deviation σy of the time-series data of the physical quantity 2. FIG. In FIG. 5, the horizontal axis is the average y and the vertical axis is the standard deviation σy. FIG. 5 exemplifies a system in which an abnormality is determined when the physical quantity 2 is greater than a predetermined reference value. The dotted line on the horizontal axis indicates the reference value Cty of the physical quantity 2. In FIG. 5, when the standard deviation σy is larger than expected, the reliability of the estimated value is low and it is not judged to be abnormal, so the vertical axis also shows the reference value Ctσy of the standard deviation σy, and the hatched part is judged to be abnormal. area.

The reference value Ctσy of the standard deviation σy may be, for example, a value assumed from the measurement error of the measuring instrument 3, or determined from the expected fluctuation width of the physical quantity 2 as a system. In any case, for example, when unexpected noise enters the abnormality detection device S, the standard deviation σy becomes large, so even if the estimation accuracy of the average y exceeds the reference value Cty, it is considered abnormal. If the standard deviation σy is small without evaluation, the estimation accuracy of the average y is high, so if the average y exceeds the reference value Cty, it is evaluated as abnormal. In this way, the accuracy and reliability of abnormality detection can be improved by evaluating the abnormality in consideration of the standard deviation.

FIG. 6 is a diagram for explaining the abnormality evaluation method (part 2) based on the average y and the standard deviation σy of the time-series data of the physical quantity 2. FIG. FIG. 6 exemplifies a system in which an abnormality is detected when the physical quantity 2 is smaller than a predetermined reference value Cty. In FIG. 6, as in the example shown in FIG. 5, when the standard deviation σy is larger than expected, the reliability of the estimated value is low and it is not judged to be abnormal. , and the shaded area was defined as an abnormality determination region.

FIG. 7 is a diagram for explaining an abnormality evaluation method (part 3) based on the average, standard deviation y, and standard deviation σy of the time-series data of the physical quantity 2. FIG. FIG. 7 exemplifies a system in which a reference value is determined in advance as to how much the standard deviation σy can fluctuate in a normal state. When the standard deviation σy exceeds the reference value Ctσy, an abnormality is evaluated and detected, so the shaded area is defined as an abnormality determination region.

As described above, in the abnormality detection device S according to the present embodiment, the physical quantity 2 is measured by the measuring instrument 3, and the measured analog signal is converted into a digital signal for processing. 5, 6, and 7, the abnormality detection conditions can be easily changed. That is, a characteristic curve of the standard deviation y and the standard deviation σy may be obtained, and the area above or below the characteristic curve may be used as the abnormality detection condition. Also, in FIGS. 5, 6 and 7, the vertical axis is the standard deviation σy, but instead of the standard deviation σy, the error y error of equation (3) or the standard deviation σy of equation (4) is used good too.

FIG. 8 is a diagram for explaining a method of calculating the time change rate a of the time-series data of the physical quantity 2 and its standard deviation σa. In FIG. 3 , the horizontal axis is time x and the vertical axis is data y of physical quantity 2 . For example, if the physical quantity 2 is current, the unit is ampere (A). For the time-series data of the physical quantity 2, the time change rate a of N data and its standard deviation σa are calculated. FIG. 8 shows a case where the number of data is N=3 for the sake of simplicity. Calculate its standard deviation σa. In the second calculation, x=[x2, x3, x4], y=[y2, y3, y4]. In the third calculation, x=[x3, x4, x5], y=[y3, y4, y5. ] and its standard deviation σa.

In this way, each time new data is input, the oldest data among the data used in the previous calculation is discarded by FIFO (First In First Out), and new data including the new data is generated. calculate. For example, in the second calculation where y4 is input, discard the oldest data x1, y1, include the latest data x4, y4, x=[x2, x3, x4], y=[y2, y3, y4 ] to find the time change rate a and its standard deviation σa. Each time data is input, the calculation of the time rate of change a and its standard deviation σa is repeated for a new combination of x and y. The time change rate a and the standard deviation σa are calculated using regression analysis, for example. Let y=ax+b be a regression equation in which a is the slope (that is, the rate of change over time) and b is the intercept. yi and the total error W of the regression equation are represented by Equation (5). N is the number of data and j is the number of calculations.

For example, referring to FIG. 7, N=3, and in the second calculation, i=j=2 and j+N-1=4. The deviation σa will be calculated. The slope a and The value of the intercept b can be calculated. a and b are represented by equations (6) and (7), respectively.

After deriving a, b may be obtained from the linear regression equation y=ax+b. In practice, these estimates a and b are estimates as well as the mean y, so we can calculate the standard deviation σa (error of a) and standard deviation σb (error of b). When the variation in data is large, that is, when noise occurs, the standard deviation σa and the standard deviation σb become large. In this case, as shown in equations (8), (9), and (10), σa and σb are derived after obtaining the residual sum of squares Sy ² of y.

The error aerror of the time rate of change a may be defined as shown in Equation (11).

In the above, the method of calculating the rate of change over time using the linear regression equation y=ax+b as the regression equation has been described. , α and β are coefficients, and t is time), etc., regression analysis may be performed using that function as a regression equation to calculate the rate of change over time and the standard deviation.

FIG. 9 is a diagram showing calculation results of the time-series data of the physical quantity 2 and the standard deviation σa of the time-series data. The time x increases by 1, and the physical quantity data y changes between 2 and 10. As in the example of FIG. 8, the number of data is set to N=3. For example, as shown in the row of i=1 in the table of FIG. error is 0%. This is because y always increases by 1 each time x increases by 1, and the time change rate a can be obtained without error.

On the other hand, as shown in the row of i = 4, the time change rate a of x = [x4, x5, x6] and y = [y4, y5, y6] is -0.05, the standard deviation σa is 0.144, The error aerror is 28.868%. This is negative when the time change rate a is obtained from two points of y4 = 3.5 and y5 = 3.2, but since y6 = 3.5, the time change rate a is obtained from y5 and y6. This is because the time change rate a obtained at the three points y4, y5, and y6 is both positive and negative, and the standard deviation σa and the error aerror increase. When an abnormality is detected using the rate of change over time a, the estimated accuracy of the rate of change over time a is verified based on the standard deviation σa and the error aerror to evaluate the abnormality.

FIG. 10 is a diagram for explaining an anomaly evaluation method based on the time change rate a and the standard deviation σa of the time-series data of the physical quantity 2. FIG. In FIG. 10, the horizontal axis is the time change rate a, and the vertical axis is the standard deviation σa. FIG. 10 exemplifies a system in which an abnormality is determined when the time rate of change a of the physical quantity 2 is greater than a predetermined reference value. A dotted line on the horizontal axis indicates the reference value Cta of the time rate of change a of the physical quantity 2 . In FIG. 10, when the standard deviation σa is larger than expected, the reliability of the estimated value is low and it is not determined to be abnormal. area.

The reference value Ctσa of the standard deviation σa may be obtained, for example, from the measurement error of the measuring instrument 3, or from the expected fluctuation range of the time change rate a of the physical quantity 2 in the system 1 You can decide. In any case, for example, when unexpected noise enters the anomaly detection device S, the standard deviation σa becomes larger than the reference value Ctσa, and the estimation accuracy of the time change rate a decreases. Even if the rate a exceeds the reference value Cta, it is not evaluated as abnormal. Evaluate as abnormal. In this way, the accuracy and reliability of abnormality evaluation can be enhanced.

Like the abnormality evaluation method using the mean y and standard deviation σy of physical quantity 2 shown in FIGS. The abnormality detection area may be determined according to the system. For example, when the time rate of change a of the physical quantity 2 is smaller than the predetermined reference value Cta, it is determined to be abnormal. Moreover, when the standard deviation σa is equal to or less than the reference value Ctσa, it is evaluated as abnormal. Alternatively, when the time rate of change a of the physical quantity 2 is greater than a predetermined reference value Cta, and the abnormality is determined to be abnormal, the time rate of change a of the physical quantity 2 is greater than the reference value Cta, and the standard deviation σa is the standard When it is equal to or less than the value Ctσa, it is evaluated as abnormal. Alternatively, in the system 1 in which the extent to which the standard deviation σa can fluctuate is known in advance, it is determined as abnormal when the standard deviation σa exceeds the reference value Ctσa.

As described above, in the abnormality detection device S according to the present embodiment, the physical quantity 2 is measured by the measuring instrument 3, and the measured analog signal is converted into a digital signal for statistical processing. In addition, the abnormality detection conditions can be easily changed. A characteristic curve of the time change rate a and the standard deviation σa may be obtained, and the area above or below the characteristic curve may be used as the abnormality detection condition. Also, the vertical axis may be the error of formula (10) aerror.

FIG. 11 is a diagram for explaining an abnormality evaluation method using the average y and standard deviation σy of time-series data of physical quantity 2, and the rate of change over time a and standard deviation σa of time-series data of physical quantity 2. . In FIG. 11, the horizontal axis is the average y, the vertical axis is the time change rate a, and the reference values are indicated by dotted lines. As indicated by diagonal lines in FIG. 11, the abnormality detection area is defined as an abnormality detection area in which the average y is equal to or greater than the reference value Cty and the temporal change a of the physical quantity 2 is equal to or greater than the reference value Cta. At this time, if at least one of the standard deviation σy of the time-series data and the standard deviation σa of the time change a is larger than the respective reference values Ctσy and Ctσa, it is not evaluated as abnormal. That is, the following are the abnormality detection conditions.
(y ~ > Cty ~) ∧ (σy < Ctσy) ∧ (a > Cta) ∧ (σa < Ctσa)
(12)

The approach of FIG. 11 is effective for systems 1 in which the abnormal mode depends on either the average y of the physical quantity 2 or the time rate of change a. Determine the abnormality detection area of the system 1 using the average y ~, standard deviation σy (or error y error), time change rate a, and standard deviation σa of time change rate a (or error of time change rate a) .

In FIG. 11, the abnormality detection area is set when the average y of the physical quantity 2 is equal to or greater than the reference value Cty and the time change a of the physical quantity 2 is equal to or greater than the reference value Cta. Since the abnormal evaluation area differs depending on the physical quantity 2, the abnormal area detection area may be changed according to the features of the system 1. FIG. For example, the abnormality detection region may be set such that the average y of the physical quantity 2 is equal to or less than the reference value Cty and the time change a of the physical quantity 2 is equal to or less than the reference value Cta. In the system 1 in which the variation range of the standard deviation σy of the time-series data and the standard deviation σa of the rate of change a over time is known in advance, the standard deviation σy and the standard deviation σa are different from the reference values Cty and Ctσa, respectively. An anomaly may be detected when exceeded.

In Example 1, an example in which the abnormality detection device S and the abnormality detection method described in the embodiment are applied to a railway DC feeding system will be described. In addition, "cl" in the drawing of the first embodiment indicates the closing state of the circuit breaker.

FIG. 12 is a diagram showing the overall configuration of the DC feeding circuit 1S according to the first embodiment. The DC feeding circuit 1S includes a substation 13 (substation A) as a first transformation facility and a substation 14 (substation B) as a second transformation facility adjacent to the substation 13. . A substation is also called a substation system. The substation 13 constitutes a substation system that receives AC power from a power grid and feeds DC power to a feeder line. An AC circuit breaker 15 , a transformer 16 , a rectifier 17 , a DC circuit breaker 18 , a DC circuit breaker 19 , a DC circuit breaker 20 and a DC circuit breaker 21 are arranged in the substation 13 . The substation 13 converts AC power received from the power grid into DC power via a transformer 16 and a rectifier 17, and feeds the DC power to a feeder line.

An AC circuit breaker 22 , a transformer 23 , a rectifier 24 , a DC circuit breaker 25 , a DC circuit breaker 26 , a DC circuit breaker 27 and a DC circuit breaker 28 are arranged in the substation 14 . The substation 14, like the substation 13, converts AC power received from the power grid into DC power via the transformer 23 and the rectifier 24, and feeds it to the feeder line.

During normal operation, DC circuit breaker 18, DC circuit breaker 19, DC circuit breaker 20, DC circuit breaker 21, DC circuit breaker 25, DC circuit breaker 26, DC circuit breaker 27, DC circuit breaker 28, AC circuit breaker 15, and , the AC circuit breaker 22 is closed. The DC circuit breaker 18 is connected to the section A of the overhead overhead line 29 . The DC circuit breaker 21 and the DC circuit breaker 25 are connected to the section B of the overhead overhead line 29 . The DC circuit breaker 28 is connected to the section C of the overhead overhead line 29 . The DC circuit breaker 19 is connected to the section F of the overhead overhead line 30 . The DC circuit breaker 20 and the DC circuit breaker 26 are connected to the section E of the overhead overhead line 30 . The DC circuit breaker 27 is connected to the section D of the overhead overhead line 30 .

The train 31 is supplied with DC power from the substation 13 and the substation 14 via the overhead overhead line 29 while the train 31 is running in the section B. The DC power supplied to train 31 returns to substation 13 and substation 14 via rail 32 . The paths of the current supplied to the train from the substation 13 and the substation 14 are shown as a current path 33 and a current path 34 .

In FIG. 12 , for example, in the substation 13 , the current measuring device 35 measures the current downstream of the DC circuit breaker 21 and inputs the analog signal to the fault current detection circuit 36 . The fault current detection circuit 36 is a DC selective cutoff device. The fault current detection circuit 36 evaluates abnormalities such as overload current and short-circuit current based on the input analog signal. When a fault current occurs, the fault current detection circuit 36 detects an abnormality and opens the DC circuit breaker 21 to interrupt the current. Although not shown, the DC circuit breaker 18, the DC circuit breaker 19, and the DC circuit breaker 20 are also provided with current measuring devices and fault current detection circuits, respectively. Each fault current detection circuit detects an abnormality when a fault current occurs, and controls the corresponding DC circuit breaker 18, DC circuit breaker 19, and DC circuit breaker 20, respectively.

Similarly in the substation 14 , the current measuring device 37 measures the current downstream of the DC circuit breaker 25 and inputs the analog signal to the fault current detection circuit 38 . The fault current detection circuit 38 is a DC selective cutoff device. The fault current detection circuit 38 evaluates abnormalities such as short-circuit current and load current based on the input analog signal. When a fault current occurs, the fault current detection circuit 38 detects an abnormality and opens the DC circuit breaker 25 to interrupt the current. Although not shown, the DC circuit breaker 26, the DC circuit breaker 27, and the DC circuit breaker 28 are also provided with current measuring devices and fault current detection circuits, respectively. Each fault current detection circuit detects an abnormality when a fault current occurs, and controls the corresponding DC circuit breaker 26, DC circuit breaker 27, and DC circuit breaker 28, respectively.

FIG. 13 is a diagram showing the configuration of the fault current detection circuit 36 according to the first embodiment. The fault current detection circuit 38 also has the same configuration as the fault current detection circuit 36 . An analog signal measured by the current measuring device 35 is input to the fault current detection circuit 36 . The fault current detection circuit 36 has the same configuration as the abnormality detection device S shown in FIG. be. Further, the fault current detection circuit 36 is configured including a circuit breaker control section 7a.

In the analysis unit 6 in this embodiment, the average I of the time-series data of the current I, the standard deviation σI of the time-series data of the current I, the time change rate dI/dt of the time-series data of the current I and its standard deviation σ _{dt /dI} . These calculations are based on the method described in FIGS. 3 and 8 of the embodiment. When the detection unit 7 detects an abnormality, the circuit breaker control unit 7 a transmits an opening command to the DC circuit breaker 21 to control the DC circuit breaker 21 .

The fault current detection circuit 36 may be configured by a processor such as a microcomputer or FPGA. The fault current detection circuit 36 may be composed of one processor or may be composed of a plurality of processors. When the fault current detection circuit 36 is composed of a plurality of processors, the AD conversion section 4, the digital signal reading section 5, the analysis section 6, the detection section 7, and the circuit breaker control section 7a are appropriately integrated and distributed. Located in each processor. For example, the AD conversion unit 4 may be arranged in one processor, and the digital signal reading unit 5, the analysis unit 6, the detection unit 7, and the circuit breaker control unit 7a may be arranged in one processor. Alternatively, the AD conversion section 4 may be arranged in one processor, the digital signal reading section 5, the analysis section 6, and the detection section 7 may be arranged in one processor, and the circuit breaker control section 7a may be arranged in one processor.

FIG. 14 is a diagram showing aspects of overload current and short-circuit current in the DC feeding circuit 1S according to the first embodiment. In FIG. 14, time t is plotted on the horizontal axis and current I is plotted on the vertical axis. In the DC feeding system, when the load current exceeds the allowable current or when a short circuit occurs, it is detected as the occurrence of a fault current, that is, as an abnormality in the feeding system, and the DC circuit breaker is controlled. As shown in FIG. 14, the overload current 39 is characterized by a gentle increase and becomes larger than the allowable current CtI of the load. , the current flowing through the circuit breaker installed at the substation increases in a short period of time (the time rate of change dI/dt is greater than a predetermined value), as shown by the time rate of change 40a. The fault current detection circuit 36 is required to selectively detect the overload current 39 and the short circuit current 40, not to operate unnecessarily when noise occurs, and to detect abnormalities such as overload and short circuit early. As described above, the abnormality detection device and the abnormality detection method described in the embodiments can be applied to the DC feeding system.

15A and 15B are diagrams for explaining a method of detecting overload current and short-circuit current according to the first embodiment. In FIG. 15A, the horizontal axis is the average I of the time-series data of the current I, and the vertical axis is the time change rate dI/dt of the time-series data of the current I. In FIG. In FIG. 15A, the reference values of the current I are I1 and I2, and the reference value of the time rate of change dI/dt of the current I is A1. In FIG. 15B, the horizontal axis represents the standard deviation σI of the time series data of the current I, and the vertical axis represents the standard deviation _σdI/dt of the time change rate dI/dt of the time series data of the current I. In FIG. 15B, the reference values of the current I are I1 and I2, and the reference value of the time change rate dI/dt of the current I is A1.

As indicated by diagonal lines in FIG. 15A, the fault current determination region is divided into region Re1 and region Re2. A region Re1 is a region used for detecting a short-circuit current, and is a region where failure is determined when the current is greater than the reference value I1 and the time rate of change of the current I is greater than the reference value A1. A region Re2 is a region used for detecting an overload current, and is a region in which failure is determined when the current is greater than the reference value I2 and the time rate of change of the current I is less than the reference value A1.

In the fault current detection circuit 36, the analysis unit 6 calculates the average I of the time-series data of the current I, the standard deviation σI of the time-series data of the current I, the time change rate dI/dt of the time-series data of the current I, and its standard A deviation σ _dI/dt is calculated by a statistical method, and overload current and short-circuit current are detected based on these calculation results. For example, the average I of the time-series data of the current I is from formula (1), the standard deviation σI of the time-series data of the current I is from formula (2), and the time rate of change dI/dt of the time-series data of the current I is From the equation (6), the standard deviation σ _dI/dt of the rate of change over time of the time-series data of the current I is obtained from the equation (9). In FIG. 15A, the reference value CtσI of the standard deviation σI in the time-series data of the current I and the reference value CtσdI _{/dt of the standard deviation σdI/dt} of the time change rate are determined, and the standard deviation _σI and the standard deviation _σdI/dt are determined. If at least one of them exceeds the reference value shown in FIG. 15B, even if the conditions shown in FIG. 15A correspond to regions Re1 and Re2, fault current (that is, overload current and short-circuit current) is not evaluated and not detected. shall be

That is, the condition for detecting the short-circuit current corresponds to the region Re1, and is given by Equation (13).
(I ~ > I1) ∧ (σI < Ct σI) ∧ (dI / dt > A1) ∧ (σ _{dI / dt} < Ctσ _{dI / dt} )
(13)

Moreover, the overload current detection condition corresponds to the region Re2, and is given by Equation (14).
(I ~ > I2) ∧ (σI < CtI) ∧ (dI/dt < A1) ∧ (σ _dI/dt < Ctσ _dI/dt )
(14)

In the above, the standard deviation σI and the standard deviation _σdI/dt were used as indicators of the error in the calculation results, but the error I error of the average I based on equation (3) and the time change rate dI/dt based on equation (11) The error dI/dterror of dt may be employed as an error indicator.

FIG. 16A is a diagram showing a one-dimensional regression analysis result of noise simulation due to damped oscillation. FIG. 16B is a diagram showing a simulated one-dimensional regression analysis result of short circuit current. FIG. 16C is a diagram showing comparison results of simulation of noise due to damped oscillation and short-circuit current. 16A and 16B, the effectiveness of the abnormality detection method in this embodiment will be described using the results of regression analysis by simulating noise and short-circuit current. FIG. 16A shows a waveform 16a of time-series data simulating noise with damped vibration. FIG. 16B shows a waveform 16b of time-series data simulating a short-circuit current rising at 3×10 ⁶ A/s.

In the waveform of the short-circuit current in FIG. 16B, it was assumed that the measuring instrument had an error of ±2%, and the error was calculated using pseudo-random numbers and obtained as y=3×10 ⁶ ×±2%. For comparison, it is necessary to perform regression analysis with the same number of data and time width, so as shown in FIGS. 16A and 16B, for data points with the same time width ΔT ((0-T) seconds) Regression analysis was performed. The number of data within ΔT in FIGS. 16A and 16B is also the same. The results are shown in FIG. 16C.

As shown in FIG. 16C, the time rate of change dI/dt of the current I was 3.12×10 ⁶ A/s for the damped vibration noise and 2.96×10 ⁶ A/s for the short circuit current. The standard deviation σ _dI /dt of the time rate of change dI/dt of the current I was 0.53×10 ⁶ A/s for the damped oscillation noise and 0.0059×10 ⁶ A/s for the short-circuit current. The error σ _dI/dt error of the time rate of change dI/dt of the current I was 17% for the damped oscillation noise and 0.2% for the short-circuit current. The time rate of change dI/dt of the damped oscillation noise and the short-circuit current are almost the same, but there is a large difference in the standard deviation _σdI/dt . This indicates that, according to this embodiment, the standard deviation _σdI/dt is used to prevent unnecessary abnormality detection when noise occurs.

Furthermore, when the time rate of change dI/dt is detected by an analog differential circuit, for example, if the measured current signal has a measurement error of ±2% as shown in FIG. It is considered that an error of about ±4%, which is twice the current measurement error, occurs. However, as in the present embodiment, by using regression analysis, which is a statistical method, the error σ _dI/dt error of the time rate of change dI/dt of the time rate of change is suppressed to 0.2% as shown in FIG. 16C. is made of. Therefore, according to the present embodiment, it can be seen that the abnormality detection accuracy is improved. The error σ _dI/dt error was calculated using Equation (11).

In a second embodiment, a method of estimating a fault point when a short-circuit current occurs between two adjacent substations using the fault current detection circuits 36 and 38 described in the first embodiment will be described.

FIG. 17 is a diagram showing the configuration of the fault current detection circuit 41 according to the second embodiment. In the fault current detection circuit 41 according to the second embodiment, an analog signal measured by the current measuring device 35 is input in the same way as the fault current detection circuit 36 described with reference to FIG. The fault current detection circuit 41 includes an AD converter 4 , a digital signal reader 5 , an analyzer 6 , a detector 7 and a fault point analyzer 42 . The failure point analysis unit 42 calculates the average I of the time-series data of the current I calculated by the analysis unit 6, the standard deviation σI of the time-series data of the current I, the time change rate dI/dt of the time-series data of the current I, and Using the standard deviation σ _dI/dt , the resistance R and inductance L of the electric path formed from the substation via the fault point 44 (FIG. 18) are calculated and transmitted to the expedition slave station 47 (FIG. 18). The expedition slave station 47 is a computer or the like.

The fault current detection circuit 41 may be configured by a processor such as a microcomputer or FPGA. The fault current detection circuit 41 may be composed of one processor or may be composed of a plurality of processors. When the fault current detection circuit 41 is composed of a plurality of processors, the AD conversion section 4, the digital signal reading section 5, the analysis section 6, the detection section 7, and the fault point analysis section 42 are appropriately integrated and distributed. Located in each processor.

FIG. 18 is a diagram showing a situation when a short-circuit accident occurs with the DC feeding circuit 2S according to the second embodiment. The DC feeding circuit 2S according to the second embodiment is the same as FIG. It includes a certain substation 14 (substation B). In FIG. 18, the case where a short-circuit accident occurs between the section B of the up overhead wire 29 and the rail 32 is considered. 20, DC circuit breaker 26, DC circuit breaker 27 and DC circuit breaker 28 are not shown.

In FIG. 18, in the substation 13, the current measuring device 35 measures the current downstream of the DC circuit breaker 21 and inputs the analog signal to the fault current detection circuit 41 described in FIG. Similarly, in the substation 14 , the current measuring device 37 measures the current downstream of the DC circuit breaker 25 and inputs the analog signal to the fault current detection circuit 43 . The configuration of the fault current detection circuit 43 is the same as that of the fault current detection circuit 41 .

If a short-circuit accident occurs between the section B of the up overhead wire 29 and the rail 32, the resistance at the failure point 44 becomes 0, so the AC input of the substation 13 flows to the failure point 44 via the DC circuit breaker 21. A current path 45 and a current path 46 that flows from the AC input of the substation 14 to the fault point 44 via the DC circuit breaker 25 .

As shown in FIG. 18, the resistance in the overhead wire 29 from the substation 13 to the fault point 44 is _R11 , the inductance is _L11 , the resistance in the rail 32 from the fault point 44 to the substation A13 is _R12 , the inductance is L ₁₂ . The resistance R ₁ of the resistor in current path 45 is R ₁ =R ₁₁ +R ₁₂ and the inductance L ₁ in current path 45 is L ₁ =L ₁₁ +L ₁₂ . Similarly, the resistance of the overhead wire 29 from the substation 14 to the fault point 44 is R ₂₁ and the inductance is L ₂₁ , the resistance of the rail 32 from the fault point 44 to the substation 14 is R ₂₂ and the inductance is L ₂₂ . The resistance _R2 in the current path 46 becomes _R2 = _R21 + _R22 , and the inductance _L2 in the current path 46 becomes _L2 = _L21 + _L22 .

When the short-circuit current is detected, the fault current detection circuit 41 calculates the resistance _R1 and the inductance L1, the fault current detection circuit ₄₃ calculates the resistance _R2 and the inductance _L2 , and the remote monitoring and control equipment, etc. to the expedition slave station 47. The expedition slave station 47 calculates the position of the fault point based on the received resistance _R1 and inductance _L1 , and resistance _R2 and inductance _L2 . Details of the calculation method are described below.

As shown in FIG. 18, the current path 45 is a series LR circuit. Assuming that the output voltage of the substation 13 is V0 (V) and the current flowing through the current path 45 is I, the following formula (15) holds. For example, it may be a constant such as V0=1500 (V), and the sending voltage may be measured by a voltage measuring device (not shown).

As described with reference to FIG. 17, in the fault current detection circuit 41, the fault point analysis unit 42 calculates the average I of the time-series data of the current I calculated by the analysis unit 6, The standard deviation σI, the time change rate dI/dt of the time-series data of the current I, and the standard deviation σdI/dt are recorded. In addition, the sending voltage V0 measured by a voltage measuring device (not shown) is also recorded. Among the average I of the time-series data of the current I and the time change rate dI/dt of the time-series data of the current I, the failure point analysis unit 42 determines the time at which the short-circuit current is detected as t0, and the time after ΔT seconds as Assuming that t1=t0+ΔT, equations (16) and (17) are established from equation (15).

A resistance R1 and an inductance L1 are obtained from the simultaneous equations of the equations (13) and (14).

At this time, the average I of the time-series data of the current I and the time-change rate dI/dt of the time-series data of the current I are the standard deviation σI of the time-series data of the current I and the time-change rate of the time-series data of the current I The calculation result of the standard deviation σ _dI/dt of is smaller than the reference value. By doing so, R ₁ and L ₁ can be obtained with high accuracy.

In this way, the result calculated by the analysis unit 6 of the fault current detection circuit 41 is recorded by the fault point analysis unit 42, and by using the time-series data of the calculation result and the equations (18) and (19), the substation The resistance and inductance of the circuit at the location and fault point can be derived. Using equations (18) and (19), the resistance _R2 and inductance _L2 of current path 46 can also be calculated. Since resistance and inductance are proportional to the length of the electrical path, expedition slave station 47 uses R ₁ , R ₂ if the distance D _AB of substation 13 and substation 14 is known to determine the fault from substation 13 . The distance D _A to the point 44 and the distance D _B from the substation 14 to the fault point 44 can be derived as shown in equations (20) and (21). For example, R ₁ :R ₂ =D _A :D _B =D _A :(D _AB -D _A )=(D _AB -D _B ):D _B.

Also, using L ₁ and L ₂ , the distance D _A and the distance D _B can be derived as shown in Equations (22) and (23).

In addition, in a DC feeding system consisting of a series circuit of an inductance L and a resistance R, the resistance and inductance may be obtained from the equation (24) representing the current I at the short circuit occurrence time t when the voltage E is sent from the substation. .

As can be seen from the equation (24), in the DC LR circuit, the current I approaches E/R as time elapses. Regression analysis is performed using equation (25) as a regression curve to calculate regression coefficients c and d. Calculate the resistance and inductance from the value and voltage E, and calculate the distance from the substation to the fault point from equations (20) and (21), or from equations (21) and (22). good.

Note that it is not always necessary for the expedition slave station 47 to calculate the failure point 44 . For example, when a short-circuit current occurs, the fault point analysis unit 42 of the fault current detection circuit 41 or 43 that detected the short-circuit current communicates with the fault point analysis unit 42 of the fault current detection circuit 41 or 43 of the adjacent substation. to receive the resistance and inductance values. The fault point analysis unit 42 of the fault current detection circuit 41 or 43 that has detected the short-circuit current receives the resistance and inductance from the fault point analysis unit 42 of the fault current detection circuit 41 or 43 of the adjacent substation, and calculates The failure point may be calculated using the resistance and inductance obtained.

In Example 3, a capacitor capacity inspection system and a capacitor capacity inspection method to which the abnormality detection device S and the abnormality detection method described in the embodiment are applied will be described. Example 3 is an example of a case in which a plurality of physical quantities are measured in the target system.

FIG. 19 is a diagram showing the configuration of a capacitor charging/discharging circuit 3S using the capacitor capacity inspection system according to the third embodiment. A capacitor charging/discharging circuit 3S shown in FIG. 19 includes a DC power supply 48, a charging switch 49, a charging resistor 50, a capacitor 51, a discharging resistor 52, a discharging switch 53, a charging/discharging control device 54, a diode 55, a current measuring device 56, and a voltage measuring device. It is composed of a device 57 and a capacitor capacity inspection unit 58 .

For the charging switch 49 and the discharging switch 53, a contact relay or a non-contact semiconductor switch or the like is used. The charge/discharge control device 54 turns off the discharge switch 53 during charging and then turns on the charge switch 49 to start charging. to start discharging.

A diode 55 is arranged to control current flow from the capacitor 51 to the discharge resistor 52 during discharge. A current measuring device 56 measures the current flowing through the capacitor 51 . A voltage measuring device 57 measures the voltage of the capacitor 51 . The current measuring device 56 inputs an analog signal of the measured current to the capacitor capacity checking section 58 . The voltage measuring instrument 57 inputs an analog signal of the measured voltage to the capacitor capacity checking section 58 .

The capacitor capacity inspection unit 58 includes an AD conversion unit 4a, a digital signal reading unit 5a, and an analysis unit 6a. The AD converter 4a converts the analog signal of the measured current into a digital signal. The digital signal reading unit 5a reads the digital signal converted by the AD conversion unit 4a and converts it into data so that it can be processed. The analysis unit 6a analyzes the time-series data including the latest data among the data converted from the digital signal read by the digital signal reading unit 5a based on a statistical method, and calculates the average, time change rate, etc. of the time-series data. Calculate the estimated value and the error between the estimated values and the degree of dispersion of time-series data (standard deviation, etc.).

Also, the capacitor capacity inspection unit 58 includes an AD conversion unit 4b, a digital signal reading unit 5b, and an analysis unit 6b. The AD converter 4b converts the analog signal of the measured voltage into a digital signal. The digital signal reading unit 5b reads the digital signal converted by the AD conversion unit 4b and converts it into data so that it can be processed. The analysis unit 6b analyzes the time-series data including the latest data among the data converted from the digital signal read by the digital signal reading unit 5a based on a statistical method, and calculates the average, time change rate, etc. of the time-series data. Calculate the estimated value and the error between the estimated values and the degree of dispersion of time-series data (standard deviation, etc.).

The capacitor capacity inspection unit 58 includes a capacitor capacity calculation unit (detection unit) 59 that estimates the capacity of the capacitor 51 based on the calculation results of the analysis units 6a and 6b. The capacitor capacity calculation unit 59 transmits the calculated estimated value of the capacitor capacity to the management unit 60 .

The management unit 60 compares the managed value of the capacitor capacity with the estimated value of the capacitor capacity calculated by the capacitor capacity calculation unit 59, and issues an alarm as an abnormality when it deviates from the managed value. The management unit 60 also records the estimated value of the capacitance received from the capacitance calculation unit 59 .

In FIG. 19 , the managing section 60 is shown outside the capacitor capacity checking section 58 , but it may be arranged inside the capacitor capacity checking section 58 . The evaluation and inspection of the capacity of the capacitor 51 are performed when the capacitor 51 is charged or discharged, and are performed by a command from the charge/discharge control device 54 .

The analysis unit 6a calculates the average I of the time-series data of the current I and the standard deviation σI of the time-series data of the current I. The analysis unit 6b calculates the time change rate dV/dt of the time-series data of the voltage V and its standard deviation _σdV/dt . The average I and standard deviation σI are calculated by the method shown in FIG. 3, and the time change rate dV/dt and standard deviation _σdV/dt are calculated by the method shown in FIG.

Alternatively, a model function assumed in the charging circuit may be created, and regression analysis may be performed using the model function as a regression equation. In both cases, calculations are made using statistical methods.

A method of calculating the capacitor capacity in the capacitor capacity calculator 59 will be described below. Assuming that C is the capacity of the capacitor 51, Q is the charge stored in the capacitor, and V is the capacitor voltage, equation (26) holds.

Differentiating equation (26) with time yields equation (27).

Since the change in charge over time is current I, equation (27) is transformed into equation (28).

Substituting the average I of the time-series data of the current I calculated by the analysis unit 6a and the time change rate dV/dt of the time-series data of the voltage V calculated by the analysis unit 6b into the equation (28), the capacitor capacitance can be calculated. At this time, by using data when the standard deviation σI of the time-series data of the current I and the standard deviation _σdV/dt of the rate of change over time of the time-series data of the voltage V are smaller than a predetermined reference value, the capacitance of the capacitor can improve the calculation accuracy of

Note that the capacitor capacity inspection unit 58 and management unit 60 may be configured by a processor such as a microcomputer or FPGA. The capacitor capacity inspection unit 58 may be composed of one processor, or may be composed of a plurality of processors. When the capacitor capacity inspection unit 58 is composed of a plurality of processors, the AD conversion units 4a and 4b, the digital signal reading units 5a and 5b, the analysis units 6a and 6b, the capacitor capacity calculation unit 59, and the management unit 60 It is appropriately integrated and distributed and placed in each processor.

(Other embodiments)
In conjunction with the above embodiments and examples, the following embodiments are further disclosed.

(1) An anomaly detection device for detecting an anomaly in a target system based on a physical quantity measured in the target system, reading time-series quantized data of the physical quantity measured in the target system, a reading unit that converts quantized data into time series data that can be processed; and a statistic based on the latest predetermined number of time series data that analyzes the latest predetermined number of the time series data. and an analysis unit that calculates an error of the statistic, and a detection that detects an abnormality in the target system based on the statistic when the degree of variation in the statistic calculated by the analysis unit satisfies a predetermined condition. An abnormality detection device, comprising:

In (1) above, the physical quantity to be measured for evaluating the normality of the target system is measured with a measuring instrument, and the measured analog signal is converted to a digital signal and converted into data so that it can be processed. Estimated values (statistics) such as the average and time change rate, their errors, and the degree of dispersion of time-series data are calculated based on statistical methods for time-series data including past multiple points and the latest data. Detect system anomalies based on the calculation results. Therefore, if the error in the calculated average or rate of change over time or the degree of variation in the time-series data is small, the reliability of the estimated value will be high. do. If there is a large amount of error in the calculated average, time rate of change, etc., or the degree of variation in the time-series data, the reliability of the estimated value will be low, so even if the estimated value reaches the standard for abnormal values, it will not be evaluated as abnormal. Therefore, the accuracy and reliability of abnormality detection can be improved.

(2) In the anomaly detection device according to (1) above, the reading unit reads the quantized data at a predetermined cycle, converts the read quantized data into the time-series data, and After the reading unit reads the quantized data of the current cycle of the predetermined cycle until the quantized data of the next cycle of the predetermined cycle is read, the analysis unit reads the quantized data of the current cycle. Analyze the latest predetermined number of time-series data including time-series data that has been converted to be arithmetically processable, and calculate the statistical amount of the current period based on the latest predetermined number of time-series data and the error of the statistical amount of the current period. When the calculation process is completed, and the statistic of the current cycle calculated by the analysis unit and the error of the statistic of the current cycle satisfy a predetermined condition, the statistic of the current cycle is An anomaly detection device characterized by completing a process of detecting an anomaly in the target system based on the above.

In (2) above, the calculation of estimated values such as the average value and time rate of change of time-series data, the calculation of errors, and the detection of anomalies are completed between the reception of the latest digital signal and the input of the next digital signal. We will always consider the latest data and evaluate and detect anomalies in real time. Therefore, since the estimated value, the error, and the degree of variation of the time-series data are calculated in consideration of the latest data at all times to evaluate the abnormality, it is possible to detect the abnormality at an early stage without delay.

(3) The anomaly detection device according to (1) above, wherein the statistic is the average of the latest predetermined number of time-series data, and the error of the statistic is the average or the latest predetermined is the standard deviation of the time-series data of the numbers, and the detection unit detects an anomaly in the target system based on the average when the standard deviation calculated by the analysis unit satisfies a predetermined condition. An anomaly detection device characterized by:

In (3) above, when the average of the latest predetermined number of time-series data or the standard deviation of the latest predetermined number of time-series data satisfies a predetermined condition, the reliability of the average increases. By detecting anomalies in the target system, the accuracy and reliability of anomaly detection can be improved.

(4) The anomaly detection device according to (1) above, wherein the statistic is the rate of change over time of the latest predetermined number of time-series data, and the error in the statistic is the rate of change over time. standard deviation, and the detection unit detects an abnormality in the target system based on the time rate of change when the standard deviation calculated by the analysis unit satisfies a predetermined condition. detection device.

In (4) above, when the standard deviation of the rate of change over time of the latest predetermined number of time-series data satisfies a predetermined condition, the reliability of the rate of change over time increases. By detecting anomalies, the accuracy and reliability of anomaly detection can be improved.

(5) The anomaly detection device according to (1) above, wherein the statistic is the average and time change rate of the latest predetermined number of time-series data, and the error of the statistic is the average or The standard deviation of the latest predetermined number of time-series data and the standard deviation of the rate of change over time, and the detection unit detects that the standard deviation calculated by the analysis unit satisfies a predetermined condition, the average An anomaly detection device that detects an anomaly of the target system based on the rate of change over time.

In (5) above, the average of the latest predetermined number of time-series data, the standard deviation of the latest predetermined number of time-series data, and the standard deviation of the rate of change over time of the latest predetermined number of time-series data satisfy a predetermined condition. In this case, the reliability of the average and the rate of change with time increases, so by detecting anomalies in the target system based on the average and the rate of change with time, the accuracy and reliability of anomaly detection can be improved.

(6) In the anomaly detection device according to (4) or (5) above, the analysis unit performs a regression analysis of the latest predetermined number of time-series data to determine the time rate of change and the standard of the time rate of change. An anomaly detection device characterized by calculating a deviation.

In (6) above, by calculating the rate of change with time and the standard deviation of the rate of change with time by regression analysis, the accuracy of abnormality detection can be improved.

(7) In the anomaly detection device according to (5) above, the target system is a substation system that converts the voltage of power received from a power network and supplies it to a feeder line via a circuit breaker, and the physical quantity is It is the current flowing through the substation system, and when the detection unit detects an abnormality in the substation system, the circuit breaker control unit performs control to open the circuit breaker and cut off the supply of power to the feeder line. An abnormality detection device comprising:

According to the above (7), abnormal currents such as overload currents and short-circuit currents can be accurately discriminated from noise in the feeding system, and the circuit breaker can be opened quickly and appropriately.

(8) In the abnormality detection device according to (7) above, a plurality of the substation systems are present along the feeder line, and when an abnormality of the substation system is detected by the detection unit, the current from the substation system using the average of the time-series data of the current, the time-series data of the current or the standard deviation of the average, the rate of change of the time-series data of the current, and the standard deviation of the time-series rate of change a failure point analysis unit that calculates the resistance and inductance of an electric line formed through the failure point of the electric wire, and the resistance and the inductance of the substation system where an abnormality calculated by the failure point analysis unit is detected; , the failure point is calculated based on the resistance and the inductance of a substation system adjacent to the substation system calculated by the failure point analysis unit.

In (8) above, based on the resistance and inductance of the substation system calculated quickly and accurately, the failure point in the electrical circuit with the substation system as the starting point and the end point is estimated, and the failure point can be quickly and accurately estimated.

(9) In the anomaly detection device according to (1) or (2) above, there are a plurality of physical quantities, and the reading unit reads the time-series quantized data for each of the plurality of physical quantities. , the read quantized data is converted into time-series data that can be processed, the analysis unit calculates the statistic and an error of the statistic for each of the plurality of physical quantities, and the detection unit detecting an abnormality in the target system based on the statistic for each of the plurality of physical quantities when the degree of variation in the statistic for each of the plurality of physical quantities calculated by the analysis unit satisfies a predetermined condition; An anomaly detection device characterized by:

In (9) above, when the degree of variation in the statistical quantity for each of the plurality of physical quantities satisfies a predetermined condition, an abnormality in the target system is detected based on the statistical quantity for each of the plurality of physical quantities. Multiple physical quantities can be used to improve the accuracy of anomaly detection and to calculate highly accurate estimates of various indices.

(10) In the abnormality detection device according to (9) above, the target system is a capacitor charging/discharging system having a capacitor that receives a DC voltage from a DC power supply, and the physical quantity flows through the capacitor. The reading unit reads the time-series quantized data for each of the current flowing through the capacitor and the voltage of the capacitor, and the time when the read quantized data can be processed. converted to series data, the analysis unit calculates the average of the current flowing through the capacitor and the error in the average, and the time rate of change of the voltage of the capacitor and the error in the time rate of change, and the detection unit Abnormality characterized by calculating the capacitance of the capacitor based on the average and the time change rate when the average error and the time change rate error calculated by the analysis unit satisfy a predetermined condition. detection device.

In the above (10), when the average error of the current flowing through the capacitor and the error of the time rate of change of the voltage of the capacitor satisfy predetermined conditions, the capacitor By calculating the capacity of the capacitor, it is possible to check the capacity of the capacitor with high accuracy.

(11) An abnormality detection device according to any one of (1) to (10) above; An anomaly detection system characterized by:

According to the above (11), the time-series quantized data obtained by converting the physical quantity measured in the target system is used as a starting point to detect anomalies in the target system, etc., so that the noise peculiar to the analog signal is transferred to the digital signal. Since it can be identified, it is possible to accurately detect anomalies in the target system.

(12) An anomaly detection method executed by an anomaly detection device for detecting an anomaly in a target system based on a physical quantity measured in the target system, wherein time-series quantized data of the physical quantity measured in the target system is a reading step of reading, converting the read quantized data into arithmetically processable time series data; analyzing a latest predetermined number of time series data out of the time series data, an analysis step of calculating a statistic based on series data and an error in the statistic; and a detection step of detecting an abnormality in the anomaly detection method.

The present invention is not limited to the above embodiments, and includes various modifications. For example, the above-described embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. Also, as long as there is no contradiction, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, and to add the configuration of another embodiment to the configuration of one embodiment. Moreover, it is possible to add, delete, replace, integrate, or distribute a part of the configuration of each embodiment. Also, the configurations and processes shown in the embodiments can be appropriately distributed, integrated, or replaced based on processing efficiency or implementation efficiency.

1: system, 1S, 2S: DC feeding circuit, 2: physical quantity, 3: measuring instrument, 4, 4a, 4b: AD conversion section, 5, 5a, 5b: digital signal reading section, 6, 6a, 6b: analysis Part, 7: Detector, 13, 14: Substation, 15: AC Circuit Breaker, 16: Transformer, 17: Rectifier, 18, 19, 20, 21, 25, 26, 27, 28: DC Circuit Breaker, 22 : AC circuit breaker, 23: Transformer, 24: Rectifier, 29: Up catenary, 30: Down catenary, 31: Train, 32: Rail, 35, 37: Current measuring device, 36, 38: Fault current detection circuit, 41 : fault current detection circuit, 42: fault point analysis unit, 43: fault current detection circuit, 47: expedition slave station, 48: DC power supply, 49: charging switch, 50: charging resistor, 51: capacitor, 52: discharging resistor, 53: Discharge switch, 55: Diode, 56: Current measuring instrument, 57: Voltage measuring instrument, 59: Capacitance calculation unit (detection unit), 60: Management unit

Claims (12)

An anomaly detection device that detects an anomaly in the target system based on a physical quantity measured in the target system,
a reading unit that reads time-series quantized data of a physical quantity measured in the target system and converts the read quantized data into arithmetically processable time-series data;
an analysis unit that analyzes a predetermined number of the latest time-series data out of the time-series data, and calculates a statistic based on the predetermined number of the latest time-series data and an error in the statistic;
and a detection unit that detects an abnormality in the target system based on the statistic when the degree of variation in the statistic calculated by the analysis unit satisfies a predetermined condition. The abnormality detection device according to claim 1,
The reading unit reads the quantized data at a predetermined cycle, converts the read quantized data into the time-series data,
After the reading unit reads the quantized data of the current cycle of the predetermined cycle until the quantized data of the next cycle of the predetermined cycle is read,
The analysis unit analyzes the latest predetermined number of time-series data including time-series data in which the quantized data of the current cycle is converted to be arithmetically processable, and analyzes the current cycle based on the latest predetermined number of time-series data Complete the process of calculating the statistics of and the error of the statistics of the current cycle,
If the statistic of the current cycle calculated by the analysis unit and the error of the statistic of the current cycle satisfy a predetermined condition, the detection unit detects an abnormality of the target system based on the statistic of the current cycle. An anomaly detection device characterized by completing the process of detecting the The abnormality detection device according to claim 1,
The statistic is the average of the latest predetermined number of time-series data,
The statistical error is the average or the standard deviation of the latest predetermined number of time-series data,
The abnormality detection device, wherein the detection unit detects an abnormality in the target system based on the average when the standard deviation calculated by the analysis unit satisfies a predetermined condition. The abnormality detection device according to claim 1,
The statistic is the rate of change over time of the latest predetermined number of time-series data,
The error of the statistic is the standard deviation of the rate of change over time,
The anomaly detection device, wherein the detection unit detects an anomaly of the target system based on the time rate of change when the standard deviation calculated by the analysis unit satisfies a predetermined condition. The abnormality detection device according to claim 1,
The statistic is the average and time rate of change of the latest predetermined number of time-series data,
The statistical error is the standard deviation of the average or the latest predetermined number of time-series data and the standard deviation of the time rate of change,
The abnormality detection device, wherein the detection unit detects an abnormality in the target system based on the average and the time rate of change when the standard deviation calculated by the analysis unit satisfies a predetermined condition. . The abnormality detection device according to claim 4 or 5,
The anomaly detection device, wherein the analysis unit calculates the time rate of change and the standard deviation of the time rate of change by regression analysis of the latest predetermined number of time-series data. The abnormality detection device according to claim 5,
The target system is a substation system that converts the voltage of the power received from the power grid and supplies it to the feeder line via the circuit breaker,
The physical quantity is a current flowing through the substation system,
Abnormality detection, comprising: a circuit breaker control unit that performs control to open the circuit breaker and cut off power supply to the feeder line when an abnormality in the substation system is detected by the detection unit. Device. The abnormality detection device according to claim 7,
A plurality of the substation systems are present along the feeder line,
When an abnormality in the substation system is detected by the detection unit, the average of the time-series data of the current, the time-series data of the current or the standard deviation of the average, the time change rate of the time-series data of the current, and , a failure point analysis unit that calculates the resistance and inductance of the electric circuit formed from the substation system through the failure point of the feeder line using the standard deviation of the time rate of change;
the resistance and the inductance of the substation system in which an abnormality is detected calculated by the fault point analysis unit; the resistance and the inductance of a substation system adjacent to the substation system calculated by the fault point analysis unit; An anomaly detection device, wherein the failure point is calculated based on. The abnormality detection device according to claim 1 or 2,
the physical quantity is plural,
The reading unit reads the time-series quantized data for each of the plurality of physical quantities, converts the read quantized data into time-series data that can be processed,
The analysis unit calculates the statistic and the error of the statistic for each of the plurality of physical quantities,
The detection unit detects an abnormality of the target system based on the statistic for each of the plurality of physical quantities when the degree of variation in the statistic for each of the plurality of physical quantities calculated by the analysis unit satisfies a predetermined condition. An anomaly detection device characterized by detecting The abnormality detection device according to claim 9,
The target system is a capacitor charging/discharging system having a capacitor that receives a DC voltage from a DC power supply,
the physical quantity is the current flowing through the capacitor and the voltage of the capacitor;
The reading unit reads the time-series quantized data for each of the current flowing through the capacitor and the voltage of the capacitor, converts the read quantized data into time-series data that can be processed,
The analysis unit calculates the average of the current flowing through the capacitor and the error of the average, and the time rate of change of the voltage of the capacitor and the error of the time rate of change,
The detection unit calculates the capacitance of the capacitor based on the average and the time rate of change calculated by the analysis unit when the average error and the time rate of change error satisfy a predetermined condition. An anomaly detection device characterized by: The abnormality detection device according to any one of claims 1 to 10;
and a conversion unit that converts the physical quantity measured in the target system into the time-series quantized data. An anomaly detection method executed by an anomaly detection device that detects an anomaly in a target system based on a physical quantity measured in the target system,
a reading step of reading time-series quantized data of a physical quantity measured in the target system, and converting the read quantized data into arithmetically processable time-series data;
an analysis step of analyzing the latest predetermined number of time-series data among the time-series data, and calculating a statistic based on the latest predetermined number of time-series data and an error of the statistic;
and a detection step of detecting an anomaly in the target system based on the statistic when the degree of variation in the statistic calculated by the analyzing step satisfies a predetermined condition.

Images