Method for monitoring abnormal operation state of electric equipment on expressway

Abstract

The invention relates to the technical field of equipment anomaly detection, in particular to a method for monitoring operation state anomaly of highway electric equipment. Comprising the following steps: collecting bus voltage and current effective value waveform data and voltage and current effective value waveform data of each feeder line; performing data detection on the voltage and the current of each node, and obtaining a power curve of each feeder line; de-noising the load data; detecting power change caused by load switching, and combining load curves in the same operation working state; solving the change of the mean value of the power curve in the same working state, and segmenting the power curve according to the change of the mean value; obtaining a probability density function curve of the power under the normal operation condition of the equipment; and measuring the difference degree between the feeder load power curve and the probability distribution under the normal operation condition of the equipment in the same operation state, and judging the operation state of the electric equipment, so that the accuracy of monitoring the abnormal operation state of the electric equipment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment anomaly detection, and in particular to a method for monitoring the abnormal operation status of highway electrical equipment. Background Art

[0002] In recent years, with the continuous increase in the mileage of expressways, expressways have played an increasingly important role in promoting economic development, easing traffic pressure, and facilitating people's lives. The normal operation of expressway power equipment is a prerequisite for ensuring driving safety. The expressway power supply and distribution system adopts a relatively centralized power supply mode, which draws 10kV lines from the distribution network in the nearby area to the substation of the expressway. In addition, the expressway power supply and distribution system has many types and quantities of loads, and high energy consumption, especially the many types of control equipment in the tunnel, which has certain requirements for power quality. However, the power supply quality of the distribution network near the expressway is unstable, the power supply range is large, and the environment is complex (mountainous areas / rural areas, tunnels, etc.), which makes operation and maintenance management difficult and the maintenance risk is high. Therefore, it is of great significance to monitor the abnormal operation status of expressway power equipment.

[0003] When a highway load is short-circuited, has poor contact, or has equipment aging, or other abnormalities or failures occur, the load power will change. Currently, the most common method for detecting abnormalities in electrical equipment is to analyze electrical parameters and determine whether the equipment is abnormal based on change point detection or threshold analysis.

[0004] However, there are many loads on highways, and the common loads have low operating power. In addition, grid voltage fluctuations will affect the power curve, among which damaged or aging loads account for a low proportion, making it difficult to distinguish between grid voltage fluctuations and abnormal small power loads. Therefore, conventional abnormal monitoring methods for power equipment will result in the inability to identify abnormal operating conditions and aging of small power equipment, or mistakenly identify grid voltage fluctuations as abnormal power equipment, resulting in inaccurate monitoring results. Summary of the invention

[0005] The purpose of the present invention is to provide a method for monitoring the abnormal operating status of electrical equipment on highways, aiming to solve the technical problem that the abnormal operating status and aging of small-power equipment cannot be identified in the existing abnormal monitoring method of electrical equipment, or the grid voltage fluctuation is mistakenly judged as abnormality of electrical equipment, resulting in inaccurate monitoring results.

[0006] To achieve the above object, the present invention adopts a method for monitoring abnormal operation status of highway electrical equipment, comprising the following steps: The bus voltage and current RMS waveform data and each feeder voltage and current RMS waveform data are acquired by collecting the voltage and current transformers installed on the bus and each feeder in the highway distribution network; According to the system topology, the voltage and current of each node are detected, and the power curve of each feeder is obtained; De-noising the load data; Detect power changes caused by load switching and merge load curves under the same operating state; Solve the change of the mean value of the power curve under the same working state, and segment the power curve according to the change of the mean value; Obtain the probability density function curve of power under normal operation of the equipment; Measures the degree of difference between the feeder load power curve and the probability distribution under normal equipment operation under the same operating state.

[0007] Among them, in the step of acquiring bus voltage and current effective value waveform data and each feeder voltage and current effective value waveform data through voltage and current transformers installed on the bus and each feeder in the highway distribution network: Define voltage RMS waveform data , Current RMS waveform data , waveform data of effective voltage of each feeder , Current RMS waveform data , the sampling frequency is .

[0008] Among them, in the step of denoising the load data: Use ADF to test whether the power load series is a non-stationary time series; By using empirical wavelet transform, the original signal is decomposed into components with different frequencies and scales according to its Fourier spectrum, and the trend term and residual term are obtained; Find and delete abnormal data.

[0009] Among them, in the step of using empirical wavelet transform to decompose the original signal into components with different frequencies and scales according to its Fourier spectrum to obtain the trend term and the residual term: Perform Fourier transform on the voltage and current signals respectively to obtain their frequency spectra; Adaptively segment the Fourier spectrum; Construct each segment of the empirical wavelet according to the spectrum segmentation, and determine the empirical wavelet function and the empirical scaling function; Construct an empirical wavelet transform, perform inner product operations on the original signal with the empirical wavelet function and the empirical scaling function to obtain detail coefficients and approximate coefficients; The original signal is reconstructed based on the detail coefficient and the approximate coefficient, and two sequences are obtained, namely the residual term and the trend term. The spectrum with a higher amplitude corresponds to the trend term.

[0010] Among them, in the step of finding abnormal data and deleting it: Apply 3 to the residual Three-sigma rule ,in is the residual term sequence, is the standard deviation, is the mean; when If this inequality is satisfied, the voltage and current data at that time point are considered normal; when If the inequality is not satisfied, the voltage and current data at that time point are considered abnormal and the abnormal data are deleted.

[0011] Among them, in the step of detecting the voltage and current of each node according to the system topology and obtaining the power curve of each feeder: Using Kirchhoff's voltage and current laws, if the moment satisfies and A situation in which is the voltage of each feeder, , it is considered that the data at that moment is abnormal, and the voltage and current data at the abnormal moment are eliminated. , Substitution , get the power curve of each feeder .

[0012] Among them, in the step of detecting the power change caused by load switching and merging the load curves under the same operating state: Select a one-dimensional convolution kernel ,application G Perform convolution operation on the feeder power to reduce the amount of data, enhance the change edge, and calculate ,when ,in k is an empirical constant, then it is considered that the load is switched at this moment; The load curves under the same operating state are combined to calculate the mean of the power curves in each window; the calculation formula is: ; In the formula, T is the window length; The clustering algorithm is used to calculate the sum of squared errors; the calculation formula is: ; in , is the number of cluster centers; Using the elbow rule, calculate the k The loss function of the value isSSE The smallest hour k As the center number of the cluster, each cluster of data obtained is the load curve under the same working condition; The windows of the same cluster are reorganized in time order to form a continuous power curve under each working state .

[0013] Wherein, in the step of segmenting the power curve obtained under each working state: The segmentation basis is to calculate the residual of each sequence separately to minimize the total residual of the sequence, so as to obtain the average change of the load in each period of time under the same working state. The average load of each sequence is ,in K The number of segments into which the load sequence is divided.

[0014] Wherein, in the step of obtaining the probability density function curve under normal operation of the equipment: Calculate the power curve of a feeder load under the same operating conditions The probability density distribution function is obtained to obtain the probability density function curve under normal operation of the equipment. P .

[0015] Among them, in the step of measuring the difference between two probability distributions of power curves of a feeder load under the same working condition and judging the operating state of the electrical equipment: When the probability density curve of different time periods under the same working condition P , Q The mean Changes occur, i.e. , If the rated power is the same as that of the load, it is considered that the electrical equipment is abnormal.

[0016] The present invention provides a method for monitoring the abnormal operation status of highway power equipment. The method collects bus voltage and current effective value waveform data and each feeder voltage and current effective value waveform data through voltage and current transformers installed on the bus and each feeder in the highway distribution network; performs data detection on the voltage and current of each node according to the system topology structure, and obtains the power curve of each feeder; performs denoising on the load data; detects the power change caused by load switching, and merges the load curves under the same operating working state; solves the change of the mean value of the power curve under the same working state, and segments the power curve according to the change of the mean value; obtains the probability density function curve of the power under the normal operation of the equipment; measures the degree of difference between the probability distribution of the feeder load power curve and the normal operation of the equipment under the same operating state; and improves the accuracy of monitoring the abnormal operation status of the electronic equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flow chart of the steps of the method for monitoring abnormal operation status of highway electrical equipment of the present invention. DETAILED DESCRIPTION

[0019] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0020] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0021] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0022] See also Figure 1 The present invention provides a method for monitoring the abnormal operation status of highway electrical equipment, comprising the following steps: Step 1: The bus voltage and current RMS waveform data and each feeder voltage and current RMS waveform data are collected through the voltage and current transformers installed on the bus and each feeder in the highway distribution network.

[0023] In this embodiment, the voltage effective value waveform data is defined as , Current RMS waveform data , waveform data of effective voltage of each feeder , Current RMS waveform data , the sampling frequency is .

[0024] Step 2: According to the system topology, perform data detection on the voltage and current of each node and obtain the power curve of each feeder.

[0025] In this embodiment, according to the system topology, the voltage and current of each node are detected, and Kirchhoff's voltage and current law is used. If the moment satisfies and A situation in which is the voltage of each feeder, , it is considered that the data at that moment is abnormal, and the voltage and current data at the abnormal moment are eliminated. , Substitution , get the power curve of each feeder .

[0026] Step 3: De-noise the load data.

[0027] In this implementation, the load data is denoised, and the specific process of denoising is as follows: Use ADF to test whether the power load series is a non-stationary time series; for example, use the adfuller package related to statsmodel in python to test the load series; Use empirical wavelet transform, for example, use the ewt function in MATLAB 2023b version to decompose the original signal into components with different frequencies and scales according to its Fourier spectrum, and obtain the trend term and the residual term; Find and delete abnormal data.

[0028] The empirical wavelet transform is used to decompose the original signal into components with different frequencies and scales according to its Fourier spectrum, and the trend term and residual term are obtained. The specific steps are as follows: Perform Fourier transform on the voltage and current signals respectively to obtain their frequency spectra; Adaptively segment the Fourier spectrum; Construct each segment of the empirical wavelet according to the spectrum segmentation, and determine the empirical wavelet function and the empirical scaling function; Construct an empirical wavelet transform, perform inner product operations on the original signal with the empirical wavelet function and the empirical scaling function to obtain detail coefficients and approximate coefficients; The original signal is reconstructed based on the detail coefficient and the approximate coefficient, and two sequences are obtained, namely the residual term and the trend term. The spectrum with a higher amplitude corresponds to the trend term.

[0029] The specific steps for finding abnormal data and deleting it are as follows: Apply 3 to the residual law ,in is the residual term sequence, is the standard deviation, is the mean; when If this inequality is satisfied, the voltage and current data at that time point are considered normal; when If the inequality is not satisfied, the voltage and current data at that time point are considered abnormal and the abnormal data are deleted.

[0030] Step 4: Detect the power change caused by load switching and merge the load curves under the same operating state.

[0031] In this embodiment, a one-dimensional Sobel filter is used to detect the power change caused by load switching, and a one-dimensional convolution kernel is selected. ,application G Perform convolution operation on the feeder power to reduce the amount of data, enhance the change edge, and calculate ,when ,in k If is an empirical constant, it is considered that the load is switched at this moment; therefore, each window is cut from one change point to the next change point, and the data in the window reflects the transition from a stable working state before load switching to the next stable working state after load switching; The load curves under the same operating state are combined to calculate the mean of the power curves in each window; the calculation formula is: ; In the formula, T is the window length; based on The clustering algorithm is used to calculate the sum of squared errors; the calculation formula is: ; in , is the number of cluster centers; Using the elbow rule, calculate the k The loss function of the value is SSE The smallest hour k As the center number of the cluster, each cluster of data obtained is the load curve under the same working condition; The windows of the same cluster are reorganized in time order to form a continuous power curve under each working state .

[0032] Step 5: Calculate the change in the mean value of the power curve under the same working state, and segment the power curve according to the change in the mean value.

[0033] In this embodiment, the change of the mean value of the power curve under the same working state is solved, and the power curve under each working state is obtained by using a greedy principle dynamic programming algorithm, such as the findchangepts function in MATLAB2023b version, to perform data segmentation. The segmentation is based on calculating the residual of each sequence separately to minimize the total residual of the sequence, thereby obtaining the average change of the load in each time period under the same working state. The average load of each sequence is ,in K The number of segments into which the load sequence is divided.

[0034] Step 6: Obtain the probability density function curve of power under normal operation of the equipment.

[0035] In this embodiment, the power curve of a feeder load under the same working condition is calculated. The probability density distribution function is obtained to obtain the probability density function curve under normal operation of the equipment. P .

[0036] Step 7: Measure the difference between the feeder load power curve and the probability distribution under normal equipment operation under the same operating state.

[0037] In this embodiment, the difference between two probability distributions of the power curve of a feeder load under the same working condition is measured, and the operating state of the electrical equipment is determined.

[0038] When the probability density curve of different time periods under the same working condition P , Q The mean Changes occur, i.e. , If the rated power is the same as that of the load, it is considered that the electrical equipment is abnormal.

[0039] In the present invention, firstly, bus voltage and current effective value waveform data and each feeder voltage and current effective value waveform data are collected by voltage and current transformers installed on the bus and each feeder in the highway distribution network; according to the system topology, the voltage and current of each node are detected, and the power curve of each feeder is obtained; then the load data is denoised; then the power change caused by load switching is detected, and the load curves under the same operating working state are merged; then the change in the mean value of the power curve under the same working state is solved, and the power curve is segmented according to the change in the mean value; the probability density function curve of the power under normal operation of the equipment is obtained; finally, the degree of difference between the probability distribution of the feeder load power curve and the normal operation of the equipment under the same operating state is measured; through the above method, the accuracy of monitoring the abnormal operating state of electronic equipment is improved.

[0040] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the contents disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application.

[0041] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A method for monitoring abnormal operation status of highway electrical equipment, characterized in that: The steps include: The bus voltage and current RMS waveform data and each feeder voltage and current RMS waveform data are acquired by collecting the voltage and current transformers installed on the bus and each feeder in the highway distribution network; According to the system topology, the voltage and current of each node are detected, and the power curve of each feeder is obtained; De-noising the load data; Detect power changes caused by load switching and merge load curves under the same operating state; Solve the change of the mean value of the power curve under the same working state, and segment the power curve according to the change of the mean value; Obtain the probability density function curve of power under normal operation of the equipment; Measures the degree of difference between the feeder load power curve and the probability distribution under normal equipment operation under the same operating state.

2. The method for monitoring abnormal operation status of highway electrical equipment according to claim 1, characterized in that: In the step of acquiring bus voltage and current effective value waveform data and each feeder voltage and current effective value waveform data through voltage and current transformers installed on the bus and each feeder in the highway distribution network: Define voltage RMS waveform data , Current RMS waveform data , waveform data of effective voltage of each feeder , Current RMS waveform data , the sampling frequency is .

3. The method for monitoring abnormal operation status of highway electrical equipment according to claim 2, characterized in that: In the step of denoising the load data: Use ADF to test whether the power load series is a non-stationary time series; By using empirical wavelet transform, the original signal is decomposed into components with different frequencies and scales according to its Fourier spectrum, and the trend term and residual term are obtained; Find and delete abnormal data.

4. The method for monitoring abnormal operation status of highway electrical equipment according to claim 3, characterized in that: In the step of using empirical wavelet transform to decompose the original signal into components with different frequencies and scales according to its Fourier spectrum, and obtaining trend terms and residual terms: Perform Fourier transform on the voltage and current signals respectively to obtain their frequency spectra; Adaptively segment the Fourier spectrum; Construct each segment of the empirical wavelet according to the spectrum segmentation, and determine the empirical wavelet function and the empirical scaling function; Construct an empirical wavelet transform, perform inner product operations on the original signal with the empirical wavelet function and the empirical scaling function to obtain detail coefficients and approximate coefficients; The original signal is reconstructed based on the detail coefficient and the approximate coefficient, and two sequences are obtained, namely the residual term and the trend term. The spectrum with a higher amplitude corresponds to the trend term.

5. The method for monitoring abnormal operation status of highway electrical equipment according to claim 4, characterized in that: In the steps of finding and deleting abnormal data: Apply 3 to the residual law ,in is the residual term sequence, is the standard deviation, is the mean; when If this inequality is satisfied, the voltage and current data at that time point are considered normal; when If the inequality is not satisfied, the voltage and current data at that time point are considered abnormal and the abnormal data are deleted.

6. The method for monitoring abnormal operation status of highway electrical equipment according to claim 5, characterized in that: In the step of detecting the voltage and current of each node according to the system topology and obtaining the power curve of each feeder: Using Kirchhoff's voltage and current laws, if the moment satisfies and A situation in which is the voltage of each feeder, , it is considered that the data at that moment is abnormal, and the voltage and current data at the abnormal moment are eliminated. , Substitution , get the power curve of each feeder .

7. The method for monitoring abnormal operation status of highway electrical equipment according to claim 6, characterized in that: In the step of detecting the power change caused by load switching and merging the load curves under the same operating state: Select a one-dimensional convolution kernel ,application G Perform convolution operation on the feeder power to reduce the amount of data, enhance the change edge, and calculate ,when ,in k is an empirical constant, then it is considered that the load is switched at this moment; The load curves under the same operating state are combined to calculate the mean of the power curves in each window; the calculation formula is: ； In the formula, T is the window length; For the Feeder in The power of time; The clustering algorithm is used to calculate the sum of squared errors; the calculation formula is: ； in is the number of cluster centers, Indicates Clusters, It is Cluster centers, i.e. cluster The center point of the inner sample, Representation Cluster A sample point within Using the elbow rule, calculate the k The loss function of the value is SSE The smallest hour k As the center number of the cluster, each cluster of data obtained is the load curve under the same working condition; The windows of the same cluster are reorganized in time order to form a continuous power curve under each working state .

8. The method for monitoring abnormal operation status of highway electrical equipment according to claim 7, characterized in that: In the step of segmenting the power curve obtained under each working state: The segmentation basis is to calculate the residual of each sequence separately to minimize the total residual of the sequence, so as to obtain the average change of the load in each period of time under the same working state. The average load of each sequence is ,in K The number of segments into which the load sequence is divided.

9. The method for monitoring abnormal operation status of highway electrical equipment according to claim 8, characterized in that: In the step of obtaining the probability density function curve under normal operation of the equipment: Calculate the power curve of a feeder load under the same operating conditions The probability density distribution function is obtained to obtain the probability density function curve under normal operation of the equipment. P .

10. The method for monitoring abnormal operation status of highway electrical equipment according to claim 9, characterized in that: In the step of measuring the difference between two probability distributions of power curves of a feeder load under the same working condition and judging the operating status of the electrical equipment: When the probability density curves of different time periods under the same working condition P , Q The mean Changes occur, i.e. , If the rated power is the same as that of the load, it is considered that the electrical equipment is abnormal.

Images