Power distribution network fault diagnosis method and system based on wide-area synchronization measurement data

Abstract

A method and system for fault diagnosis of distribution networks based on wide-area synchronous measurement data first analyzes the similarity of power signal interference from random factors in the distribution network, aggregating microgrids with similar random interference factors to obtain various power grid monitoring random interference clusters. Second, based on the degree of power signal interference from random factors, a fault diagnosis benchmark point for each power grid monitoring random interference cluster is obtained. Then, the differences between each monitoring point and the corresponding power grid monitoring random interference cluster's fault diagnosis benchmark point are analyzed, and a fault diagnosis verification factor for each monitoring point in the current monitoring time period is calculated. Finally, based on the fault diagnosis verification factor for each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, pre-diagnosis processing is performed on each monitoring point to obtain the fault diagnosis pre-processing result for each monitoring point. This invention improves the real-time performance and reliability of fault diagnosis in distribution networks.

Description

Technical Field

[0001] This invention belongs to the field of smart distribution network fault diagnosis technology, specifically, it relates to a method and system for fault diagnosis of distribution networks based on wide-area synchronous measurement data. Background Technology

[0002] Most of the power distribution network is exposed to the elements and is highly susceptible to factors such as severe weather, natural disasters, environmental pollution, external damage, and construction quality. This often leads to accidents and faults in the power distribution network. The most direct consequence of power distribution network faults is local or large-scale power outages, which can damage equipment in the distribution lines, cause inconvenience to users, and affect industrial production and commercial operations. Therefore, timely diagnosis of faults in the power distribution network is of great practical significance.

[0003] In existing technologies, wide-area synchronous phasor measurement technology can synchronously sample analog voltage and current signals using GPS information to obtain the amplitude and phase angle of the voltage and current signals. It features high-precision synchronous phasor measurement across different locations, high-speed communication, and rapid response, making it highly suitable for fault diagnosis in distribution networks. However, because wide-area synchronous phasor measurement technology generates a large amount of data, efficiently processing this data and extracting useful information to provide powerful data analysis capabilities to support real-time or near-real-time fault diagnosis remains a challenge. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for fault diagnosis of distribution networks based on wide-area synchronous measurement data, which improves the efficiency of fault diagnosis of distribution networks and enhances the real-time performance and reliability of fault diagnosis of distribution networks.

[0005] The present invention adopts the following technical solution.

[0006] This invention proposes a method for fault diagnosis of distribution networks based on wide-area synchronous measurement data, comprising:

[0007] A synchronous phasor measurement device is used to collect each analog power signal at each monitoring point during each monitoring time period. Each analog power signal is then processed by frequency domain transformation to obtain the power time spectrum diagram of each monitoring point.

[0008] Based on each power time spectrum diagram, the fundamental signal and each harmonic signal of the corresponding simulated power signal are obtained. Based on the regularity characteristics of the fluctuation of each harmonic signal, the corresponding random factor interference index is calculated. Based on the random factor interference index of all harmonic signals of each simulated power signal, the random factor interference harmonics are obtained. Based on the difference characteristics between the frequency of each fundamental signal and the frequency of each random factor interference harmonic, and the signal strength of each random factor interference harmonic of each fundamental signal, the diagnostic strong interference index of each random factor interference harmonic is calculated.

[0009] Based on all accidental interference harmonics and corresponding diagnostic strong interference indices, construct each interference diagnostic dimension reduction signal of each power time spectrum diagram, calculate the corresponding homogeneous interference index between monitoring points based on the similarity characteristics between the interference diagnostic dimension reduction signals of monitoring points, and obtain each power grid monitoring accidental interference cluster based on the homogeneous interference index between all monitoring points.

[0010] The fault diagnosis confidence factor for each monitoring point is calculated based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics at each monitoring point. The fault diagnosis benchmark point for each power grid monitoring random factor interference cluster is obtained based on the fault diagnosis confidence factor for each monitoring point and the homogeneous interference index between each monitoring point and the other monitoring points.

[0011] Based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis benchmark point of the random interference cluster of power grid monitoring within the current monitoring time period, the fault diagnosis verification factor of each monitoring point in the current monitoring time period is calculated. Based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, pre-diagnosis processing is performed on each monitoring point to obtain the fault diagnosis pre-processing result of each monitoring point.

[0012] Preferably, a synchronous phasor measurement device is set up at each monitoring point, and the synchronous phasor measurement device is used to collect each analog power signal at each monitoring point during each monitoring time period;

[0013] Each analog power signal at each monitoring point is filtered and converted from analog to digital to obtain a digital power signal at each monitoring point. Wavelet transform is then used to perform frequency domain transformation on each digital power signal to obtain a power time spectrum diagram for each monitoring point.

[0014] Preferably, the fundamental signal and each harmonic signal of the corresponding analog power signal are obtained based on each power time-frequency spectrum diagram, and the corresponding random factor interference index is calculated based on the regularity characteristics of the fluctuation of each harmonic signal, including:

[0015] Based on each power time spectrum diagram, the corresponding fundamental signal and each harmonic signal are extracted from the corresponding analog power signal, and the Hearst exponent of each harmonic signal is calculated. The Petitt mutation point detection algorithm is used to extract all mutation points in each harmonic signal. The standard deviation of the time interval between all mutation points in each harmonic signal and the next adjacent mutation point is used as the mutation interval dispersion index of each harmonic signal. The Hearst exponent of each harmonic signal is added to a preset parameter tuning factor to obtain the adjusted autocorrelation index of each harmonic signal. The ratio of the mutation interval dispersion index of each harmonic signal to the corresponding adjusted autocorrelation index is used as the random factor interference index of each harmonic signal.

[0016] The preset parameter tuning factor is set to 0.001.

[0017] Preferably, the random factor interference harmonics are obtained based on the random factor interference index of all harmonic signals of each analog power signal, including:

[0018] The random factor interference index of all harmonic signals is normalized to obtain the normalized random factor interference index of each harmonic signal. Harmonic signals with a normalized random factor interference index greater than a preset random factor threshold are regarded as random factor interference harmonics.

[0019] Preferably, based on the difference between the frequency of each fundamental signal and the frequencies of the corresponding random factor interference harmonics, and the signal strength of each random factor interference harmonic of each fundamental signal, the diagnostic strong interference index of each random factor interference harmonic is calculated, including:

[0020] The absolute value of the difference between the frequency of each random factor interference harmonic and the frequency of the corresponding fundamental signal is used as the near-frequency interference index of each random factor interference harmonic. The sum of the signal strengths of each random factor interference harmonic at all sampling times is used as the random factor interference intensity of each random factor interference harmonic. The product of the near-frequency interference index of each random factor interference harmonic and the corresponding random factor interference intensity is used as the diagnostic strong interference index of each random factor interference harmonic.

[0021] Preferably, the interference diagnosis dimension reduction signals for each power time-frequency spectrum are constructed based on all accidental interference harmonics and the corresponding diagnostic strong interference index, including:

[0022] The normalized value of the sum of the diagnostic strong interference indices of all random factor interference harmonics of each analog power signal is used as the diagnostic interference attention factor of each analog power signal, and the floor value of the product of the number of random factor interference harmonics of each analog power signal and the corresponding diagnostic interference attention factor is used as the number of random factor interference retention terms of each analog power signal.

[0023] For each random factor interference harmonic of each simulated power signal, each random factor interference harmonic is treated as a row to construct an interference diagnosis matrix for each simulated power signal. The singular value decomposition algorithm is used to decompose each interference diagnosis matrix into a left singular vector, a singular value matrix, and a right singular vector.

[0024] Based on the number of accidental interference retention terms for each simulated power signal, key singular values ​​are extracted from the singular value matrix of each simulated power signal to obtain the key singular values ​​of each simulated power signal. The left singular vector, key singular values, and right singular vector of each simulated power signal are then processed to obtain the interference diagnosis dimensionality reduction signals for each simulated power signal.

[0025] Preferably, the isomorphic interference index between monitoring points is calculated based on the similarity features between the interference diagnosis dimensionality-reduced signals of the monitoring points, including:

[0026] Based on the similarity characteristics of the reduced-dimensional interference diagnostic signals corresponding to the same type of simulated power signals between monitoring points, the isomorphic interference index between monitoring points is expressed as follows:

[0027]

[0028] in, Let be the isomorphic interference exponent between the i1th and i2th monitoring points, where exp() represents an exponential function with the natural constant as the base, nw is the number of analog power signal types, and Mah(,) is the Manhattan distance between the two signals. The number of interference diagnosis downsampling signals for the j-th simulated power signal at the i1-th monitoring point. The number of interference diagnosis downsampling signals for the j-th simulated power signal at the i2-th monitoring point. This is the k1th interference diagnosis dimension reduction signal of the jth simulated power signal at the i1th monitoring point. It is the k2th interference diagnosis dimension reduction signal of the jth simulated power signal at the i2th monitoring point.

[0029] Preferably, the interference clusters of random factors in power grid monitoring are obtained based on the homogeneous interference index among all monitoring points, including:

[0030] All monitoring points are clustered based on the same type of interference index among all monitoring points to obtain each power grid monitoring accidental factor interference cluster. The clustering process is designed as follows: the same type of interference index among monitoring points is used as the similarity criterion for the clustering algorithm, and the clustering algorithm is used to cluster all monitoring points.

[0031] Preferably, the fault diagnosis confidence factor for each monitoring point is calculated based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics at each monitoring point, including:

[0032] The product of the random factor interference index and the corresponding diagnostic strong interference index of each random factor interference harmonic is taken as the corresponding random factor strong interference factor. The sum of the random factor strong interference factors of all random factor interference harmonics at each monitoring point is taken as the distribution fault factor of each monitoring point. The calculation result of the exponential function with the natural constant as the base and the negative number of the distribution fault factor of each monitoring point as the exponent is taken as the fault diagnosis confidence factor of each monitoring point.

[0033] Preferably, the fault diagnosis benchmark point for each power grid monitoring random factor interference cluster is obtained based on the fault diagnosis confidence factor for each monitoring point and the homogeneous interference index between each monitoring point and the remaining monitoring points, including:

[0034] For each monitoring point within a cluster of random interference factors in power grid monitoring, the average value of the same type interference index between each monitoring point and the other monitoring points is used as the same type random factor benchmark factor for each monitoring point. The ratio of the same type random factor benchmark factor to the corresponding fault diagnosis confidence factor for each monitoring point is used as the corresponding same type flawless index. The monitoring point with the largest same type flawless index within each cluster of random interference factors in power grid monitoring is used as the fault diagnosis benchmark point for each cluster of random interference factors in power grid monitoring.

[0035] Preferably, based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis benchmark point of the random interference cluster in the current monitoring time period, the fault diagnosis verification factor of each monitoring point in the current monitoring time period is calculated, including:

[0036] The difference between the fault diagnosis confidence factor of each monitoring point and the corresponding fault diagnosis benchmark point of the random interference cluster of power grid monitoring within the current monitoring period is used as the fault diagnosis verification factor of each monitoring point in the current monitoring period.

[0037] Preferably, pre-diagnosis processing is performed on each monitoring point based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, to obtain the fault diagnosis pre-processing result of each monitoring point, including:

[0038] Based on all the simulated power signals of each monitoring point, signal prediction processing is performed on the simulated power signals of each monitoring point to obtain the predicted simulated power signals of each monitoring point in the next monitoring time period.

[0039] The signal prediction processing is designed as follows: the normalized value of the fault diagnosis verification factor of each monitoring point is multiplied by the preset number of autoregressive terms as the number of autoregressive terms in the differential autoregressive moving average model, and the differential autoregressive moving average model is used to predict each simulated power signal of each monitoring point in the next monitoring period.

[0040] Based on the predicted simulated power signals of each monitoring point in the next monitoring period, the fault diagnosis verification factor of each monitoring point in the next monitoring period is calculated. The fault diagnosis verification factor is normalized. When the fault diagnosis verification factor is greater than a preset threshold, the fault diagnosis preprocessing result of the monitoring point is determined to be faulty; otherwise, the fault diagnosis preprocessing result of the monitoring point is determined to be faultless.

[0041] This invention also proposes a distribution network fault diagnosis system based on wide-area synchronous measurement data, comprising:

[0042] The data processing module is used to acquire each analog power signal of each monitoring point in each monitoring time period using a synchronous phasor measurement device, and to perform frequency domain conversion processing on each analog power signal to obtain the power time spectrum of each monitoring point.

[0043] The random factor interference classification module is used to obtain the fundamental signal and various harmonic signals of the corresponding simulated power signal based on each power time-frequency spectrum diagram. It calculates the corresponding random factor interference index based on the regularity characteristics of the fluctuations of each harmonic signal, obtains the random factor interference harmonics based on the random factor interference indices of all harmonic signals of each simulated power signal, calculates the diagnostic strong interference index of each random factor interference harmonic based on the difference characteristics between the frequency of each fundamental signal and the frequencies of the corresponding random factor interference harmonics, and the signal strength of each random factor interference harmonic of each fundamental signal; constructs each interference diagnostic dimension-reduced signal of each power time-frequency spectrum diagram based on all random factor interference harmonics and the corresponding diagnostic strong interference indices; calculates the corresponding homogeneous interference index between monitoring points based on the similarity characteristics between the interference diagnostic dimension-reduced signals of monitoring points; and obtains each power grid monitoring random factor interference cluster based on the homogeneous interference index between all monitoring points.

[0044] The fault diagnosis module is used to calculate the fault diagnosis confidence factor for each monitoring point based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics at each monitoring point. Based on the fault diagnosis confidence factor of each monitoring point and the homogeneous interference index between each monitoring point and the other monitoring points, it obtains the fault diagnosis benchmark point for each power grid monitoring random factor interference cluster. Based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding power grid monitoring random factor interference cluster fault diagnosis benchmark point within the current monitoring time period, it calculates the fault diagnosis verification factor for each monitoring point in the current monitoring time period. Based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, it performs pre-diagnosis processing on each monitoring point to obtain the fault diagnosis pre-processing result for each monitoring point.

[0045] The present invention is also a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to perform operations according to the instructions to execute the steps of the method.

[0046] The present invention is also a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method.

[0047] The beneficial effects of this invention are that, compared with the prior art, it at least includes:

[0048] (1) First, the degree of interference of the power signal of the distribution network to the environment is analyzed. Based on each power time spectrum diagram, the fundamental signal and each harmonic signal of the corresponding simulated power signal are obtained. Based on the regularity of the fluctuation of each harmonic signal, the corresponding random factor interference index is calculated. Based on the random factor interference index of all harmonic signals of each simulated power signal, the random factor interference harmonics are obtained. Based on the difference between the frequency of each fundamental signal and the frequency of each random factor interference harmonic, and the signal strength of each random factor interference harmonic of each fundamental signal, the diagnostic strong interference index of each random factor interference harmonic is calculated. Based on all random factor interference harmonics and the corresponding diagnostic strong interference index, the interference diagnosis dimension reduction signal of each power time spectrum diagram is constructed. Based on the similarity between the interference diagnosis dimension reduction signals of the monitoring points, the corresponding homogeneous interference index between the monitoring points is calculated. Based on the homogeneous interference index between all monitoring points, the random factor interference clusters of each power grid monitoring are obtained. The microgrids with similar random interference factors are aggregated, which can avoid the impact of random factors such as the environment on the fault diagnosis of the microgrid. The subsequent analysis of each power grid monitoring random factor interference cluster improves the real-time performance of the microgrid fault diagnosis.

[0049] (2) The fault diagnosis confidence factor of each monitoring point is calculated based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics of each monitoring point. The fault diagnosis benchmark point of each power grid monitoring random factor interference cluster is obtained based on the fault diagnosis confidence factor of each monitoring point and the same type interference index between each monitoring point and the other monitoring points. The fault diagnosis verification factor of each monitoring point in the current monitoring period is calculated based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding power grid monitoring random factor interference cluster fault diagnosis benchmark point in the current monitoring period. By comparing the monitoring point with the benchmark point with a lower degree of interference in the power grid monitoring random factor interference cluster, the reliability of fault diagnosis of microgrid is improved.

[0050] (3) Based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, pre-diagnosis processing is performed on each monitoring point to obtain the fault diagnosis pre-processing result of each monitoring point. By adaptively determining the number of autoregressive terms of the differential autoregressive moving average model according to the fault diagnosis verification factor of the monitoring point, the real-time performance of the simulated power signal prediction of the monitoring point is improved while ensuring the accuracy of the prediction of the simulated power signal of the monitoring point, thereby improving the real-time performance of fault diagnosis of the microgrid.

[0051] (4) By predicting the simulated power signal of the microgrid in the next monitoring period and performing fault pre-diagnosis of the microgrid based on the predicted simulated power signal, the abnormality of the microgrid can be detected before or at the beginning of the fault, which further improves the real-time performance of fault diagnosis of the microgrid. Attached Figure Description

[0052] Figure 1 This is a flowchart of a power distribution network fault diagnosis method based on wide-area synchronous measurement data proposed in this invention;

[0053] Figure 2 This is a schematic diagram illustrating the acquisition of random interference clusters in power grid monitoring in an embodiment of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0055] This invention proposes a method for fault diagnosis of distribution networks based on wide-area synchronous measurement data, such as... Figure 1 and Figure 2 As shown, it includes:

[0056] Step 1: Use a synchronous phasor measurement device to collect each analog power signal at each monitoring point during each monitoring time period, and perform frequency domain conversion processing on each analog power signal to obtain the power time spectrum diagram of each monitoring point.

[0057] The implementation of wide-area synchronous phasor measurement technology mainly relies on the coordinated operation of multiple key components and technologies, such as synchronous phasor measurement devices, satellite timing systems, high-speed communication networks, and data processing and analysis centers. Therefore, before using wide-area synchronous phasor measurement technology for fault diagnosis in distribution networks, it is necessary to first install synchronous phasor measurement devices at each monitoring point to collect relevant power data from the distribution network.

[0058] Specifically, step 1 includes:

[0059] Step 1.1: Set up a synchronous phasor measurement device at each monitoring point, and use the synchronous phasor measurement device to collect each analog power signal at each monitoring point during each monitoring time period;

[0060] Step 1.2: Filter and convert each analog power signal at each monitoring point to analog-to-digital to obtain each digital power signal at each monitoring point. Use wavelet transform to perform frequency domain conversion on each digital power signal to obtain the power time-frequency spectrum of each monitoring point.

[0061] The simulated power signal at each monitoring point includes simulated voltage and simulated current signals, and the power time-frequency spectrum at each monitoring point includes voltage and current spectra. Wavelet transform can perform multi-scale frequency decomposition of the signal, providing both time-domain and frequency-domain information.

[0062] Step 2: Based on each power time spectrum diagram, obtain the fundamental signal and each harmonic signal of the corresponding simulated power signal. Calculate the corresponding random factor interference index based on the regularity characteristics of the fluctuations of each harmonic signal. Obtain the random factor interference harmonics based on the random factor interference index of all harmonic signals of each simulated power signal. Calculate the diagnostic strong interference index of each random factor interference harmonic based on the difference between the frequency of each fundamental signal and the frequency of each corresponding random factor interference harmonic, and the signal strength of each random factor interference harmonic of each fundamental signal.

[0063] The fundamental signal extracted from the simulated power signal at each monitoring point represents the original signal of the corresponding simulated power signal, while the harmonic signals extracted from the simulated power signal at each monitoring point represent the interference signals that affect the simulated power signal. The changes in interference signals caused by random environmental factors are often irregular and difficult to predict; however, the changes in interference signals caused by distribution network faults often exhibit regular characteristics and are predictable. By analyzing the fluctuation regularity characteristics of the interference signals that affect the simulated power signal, it is possible to distinguish between interference signals caused by random environmental factors and interference signals caused by distribution network faults.

[0064] Specifically, step 2 includes:

[0065] Step 2.1: Extract the corresponding fundamental signal and each harmonic signal from the corresponding analog power signal based on each power time spectrum diagram, and calculate the Hearst exponent of each harmonic signal; use the Petitt mutation point detection algorithm to extract all mutation points in each harmonic signal, and use the standard deviation of the time interval between all mutation points in each harmonic signal and the next adjacent mutation point as the mutation interval dispersion index of each harmonic signal; add a preset parameter adjustment factor to the Hearst exponent of each harmonic signal as the adjusted autocorrelation index of each harmonic signal; and use the ratio of the mutation interval dispersion index of each harmonic signal to the corresponding adjusted autocorrelation index as the random factor interference index of each harmonic signal.

[0066] The preset tuning factor is a value preset by the user. In this embodiment, the preset tuning factor is 0.001. As for the value of the preset tuning factor, in other implementation methods, the implementer can choose it at his own discretion. This embodiment does not impose any special restrictions on this.

[0067] The Hearst exponent reflects the autocorrelation of a time series, while the abrupt change interval dispersion index represents the dispersion of the time interval between each abrupt change point and adjacent abrupt change points in a harmonic signal. The stronger the autocorrelation of the harmonic signal and the closer the time interval between each abrupt change point and adjacent abrupt change points, the more regular the fluctuation of the harmonic signal is, and the more likely the harmonic signal is an interference signal caused by a power distribution network fault. The smaller the random factor interference index value is. Conversely, the weaker the autocorrelation of the harmonic signal and the greater the difference in the time interval between each abrupt change point and adjacent abrupt change points, the more likely the harmonic signal is an interference signal corresponding to random factors such as environmental factors, and the larger the random factor interference index value is.

[0068] Step 2.2: Normalize the random factor interference index of all harmonic signals to obtain the normalized random factor interference index of each harmonic signal. Harmonic signals with a normalized random factor interference index greater than a preset random factor threshold are regarded as random factor interference harmonics.

[0069] When diagnosing faults in microgrids, aggregating microgrids with similar incidental interference factors can avoid the impact of incidental factors such as the environment on the fault diagnosis of microgrids, while improving the real-time performance of microgrid fault diagnosis.

[0070] Step 2.3: The absolute value of the difference between the frequency of each random factor interference harmonic and the frequency of the corresponding fundamental signal is taken as the near-frequency interference index of each random factor interference harmonic. The sum of the signal strengths of each random factor interference harmonic at all sampling times is taken as the random factor interference intensity of each random factor interference harmonic. The product of the near-frequency interference index of each random factor interference harmonic and the corresponding random factor interference intensity is taken as the diagnostic strong interference index of each random factor interference harmonic.

[0071] When the difference between the frequency of the random interference harmonic and the frequency of the corresponding fundamental signal is smaller, and the signal strength of the random interference harmonic is stronger, it indicates that the random interference harmonic is more likely to cause greater interference to the fundamental signal, and the diagnostic strong interference index value is larger.

[0072] Step 3: Construct the interference diagnosis dimension reduction signal of each power time spectrum diagram based on all random factor interference harmonics and the corresponding diagnostic strong interference index. Calculate the corresponding homogeneous interference index between monitoring points based on the similarity characteristics between the interference diagnosis dimension reduction signals of the monitoring points. Obtain the random factor interference clusters of each power grid monitoring point based on the homogeneous interference index between all monitoring points.

[0073] Since most of the distribution network is exposed to the outdoors, it is highly susceptible to factors such as severe weather, natural disasters, environmental pollution, external damage, and construction quality. There are many accidental factors that affect the microgrid. In order to improve the real-time performance of subsequent fault diagnosis of the microgrid, it is necessary to perform dimensionality reduction analysis on the interference harmonics of each accidental factor in each simulated power signal.

[0074] Specifically, step 3 includes:

[0075] Step 3.1: The normalized value of the sum of the diagnostic strong interference indices of all random factor interference harmonics of each analog power signal is used as the diagnostic interference attention factor of each analog power signal. The floor value of the product of the number of random factor interference harmonics of each analog power signal and the corresponding diagnostic interference attention factor is used as the number of random factor interference retained terms of each analog power signal.

[0076] Step 3.2: For each random factor interference harmonic of each simulated power signal, each random factor interference harmonic is treated as a row to construct the interference diagnosis matrix of each simulated power signal. The singular value decomposition algorithm is used to decompose each interference diagnosis matrix into a left singular vector, a singular value matrix and a right singular vector.

[0077] Singular value decomposition (SVD) algorithms can reduce the dimensionality of data through eigenvalue decomposition. In a singular value matrix, all values ​​except those on the diagonal are singular and all other positions are 0.

[0078] Step 3.3: Based on the number of accidental interference retention terms for each simulated power signal, perform key singular value extraction processing on the singular value matrix of each simulated power signal to obtain the key singular values ​​of each simulated power signal. Process the left singular vector, key singular values ​​and right singular vector of each simulated power signal to obtain the interference diagnosis dimension reduction signals of each simulated power signal.

[0079] Step 3.4: Based on the similarity characteristics of the reduced-dimensional interference diagnostic signals corresponding to the same type of simulated power signals between monitoring points, the homogeneous interference index between monitoring points is expressed as follows:

[0080]

[0081] in, Let be the isomorphic interference exponent between the i1th and i2th monitoring points, where exp() represents an exponential function with the natural constant as the base, nw is the number of analog power signal types, and Mah(,) is the Manhattan distance between the two signals. The number of interference diagnosis downsampling signals for the j-th simulated power signal at the i1-th monitoring point. The number of interference diagnosis downsampling signals for the j-th simulated power signal at the i2-th monitoring point. This is the k1th interference diagnosis dimension reduction signal of the jth simulated power signal at the i1th monitoring point. It is the k2th interference diagnosis dimension reduction signal of the jth simulated power signal at the i2th monitoring point.

[0082] The smaller the Manhattan distance between the various interference diagnostic dimensionality reduction signals corresponding to the same type of analog power signals between monitoring points, the more likely the monitoring points are to be affected by the same accidental factors, and the larger the same type interference index value is.

[0083] Step 3.5: Cluster all monitoring points based on the same type interference index among all monitoring points to obtain each power grid monitoring accidental factor interference cluster. The clustering process is designed as follows: use the same type interference index among monitoring points as the similarity criterion of the clustering algorithm, and use the clustering algorithm to cluster all monitoring points.

[0084] Step 4: Calculate the fault diagnosis confidence factor for each monitoring point based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics at each monitoring point. Based on the fault diagnosis confidence factor for each monitoring point and the homogeneous interference index between each monitoring point and the other monitoring points, obtain the fault diagnosis benchmark point for each power grid monitoring random factor interference cluster.

[0085] Since the power grid monitoring random factor interference cluster contains multiple monitoring points affected by the same random factor, fault diagnosis of the distribution network can be achieved by comparing the monitoring point with the benchmark point with a lower degree of interference in the power grid monitoring random factor interference cluster.

[0086] Specifically, step 4 includes:

[0087] Step 4.1: The product of the random factor interference index and the corresponding diagnostic strong interference index of each random factor interference harmonic is taken as the corresponding random factor strong interference factor. The sum of the random factor strong interference factors of all random factor interference harmonics at each monitoring point is taken as the distribution fault factor of each monitoring point. The calculation result of the exponential function with the natural constant as the base and the negative number of the distribution fault factor of each monitoring point as the exponent is taken as the fault diagnosis confidence factor of each monitoring point.

[0088] Step 4.2: For each monitoring point within each cluster of random interference factors in power grid monitoring, the average value of the same type interference index between each monitoring point and the other monitoring points is taken as the same type random factor benchmark factor for each monitoring point. The ratio of the same type random factor benchmark factor to the corresponding fault diagnosis confidence factor for each monitoring point is taken as the corresponding same type flawless index. The monitoring point with the largest same type flawless index within each cluster of random interference factors in power grid monitoring is taken as the fault diagnosis benchmark point for each cluster of random interference factors in power grid monitoring.

[0089] The less likely a monitoring point is to fail, and the more similar the accidental factors affecting the monitoring point and other monitoring points are, the more suitable it is to use the monitoring point as the benchmark for fault diagnosis of the distribution network.

[0090] Step 5: Calculate the fault diagnosis verification factor for each monitoring point in the current monitoring period based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis benchmark point of the random interference cluster of power grid monitoring within the current monitoring period. Perform pre-diagnosis processing on each monitoring point based on the fault diagnosis verification factor of each monitoring point in the current monitoring period and all simulated power signals of each monitoring point to obtain the fault diagnosis pre-processing result of each monitoring point.

[0091] Since wide-area synchronous phasor measurement technology generates a large amount of data, it is difficult to guarantee the real-time performance of microgrid fault diagnosis if the microgrid is directly diagnosed based on the simulated power signal of the microgrid in the current monitoring period. By predicting the simulated power signal of the microgrid in the next monitoring period and performing pre-diagnosis of the microgrid based on the predicted simulated power signal, it is possible to detect microgrid anomalies before or in the early stage of a fault, thereby improving the real-time performance of microgrid fault diagnosis.

[0092] Specifically, step 5 includes:

[0093] Step 5.1: Use the difference between the fault diagnosis confidence factor of each monitoring point and the corresponding fault diagnosis benchmark point of the random interference cluster of power grid monitoring during the current monitoring period as the fault diagnosis verification factor of each monitoring point during the current monitoring period.

[0094] Step 5.2: Based on all the simulated power signals of each monitoring point, perform signal prediction processing on the simulated power signals of each monitoring point to obtain the predicted simulated power signals of each monitoring point in the next monitoring time period;

[0095] The signal prediction processing is designed as follows: the normalized value of the fault diagnosis verification factor of each monitoring point is multiplied by the preset number of autoregressive terms as the number of autoregressive terms in the differential autoregressive moving average model, and the differential autoregressive moving average model is used to predict each simulated power signal of each monitoring point in the next monitoring period.

[0096] It should be noted that by adaptively determining the number of autoregressive terms in the differential autoregressive moving average model based on the fault diagnosis verification factor of the monitoring point, the accuracy of the simulated power signal prediction for the monitoring point is ensured while the real-time performance of the simulated power signal prediction for the monitoring point is improved.

[0097] Step 5.3: Calculate the fault diagnosis verification factor for each monitoring point in the next monitoring time period based on the predicted simulated power signals of each monitoring point in the next monitoring time period. Normalize the fault diagnosis verification factor. When the fault diagnosis verification factor is greater than a preset threshold, determine that the fault diagnosis preprocessing result of the monitoring point is that a fault has occurred; otherwise, determine that the fault diagnosis preprocessing result of the monitoring point is that there is no fault.

[0098] This invention also proposes a distribution network fault diagnosis system based on wide-area synchronous measurement data, comprising:

[0099] The data processing module is used to acquire each analog power signal of each monitoring point in each monitoring time period using a synchronous phasor measurement device, and to perform frequency domain conversion processing on each analog power signal to obtain the power time spectrum of each monitoring point.

[0100] The random factor interference classification module is used to obtain the fundamental signal and various harmonic signals of the corresponding simulated power signal based on each power time-frequency spectrum diagram. It calculates the corresponding random factor interference index based on the regularity characteristics of the fluctuations of each harmonic signal, obtains the random factor interference harmonics based on the random factor interference indices of all harmonic signals of each simulated power signal, calculates the diagnostic strong interference index of each random factor interference harmonic based on the difference characteristics between the frequency of each fundamental signal and the frequencies of the corresponding random factor interference harmonics, and the signal strength of each random factor interference harmonic of each fundamental signal; constructs each interference diagnostic dimension-reduced signal of each power time-frequency spectrum diagram based on all random factor interference harmonics and the corresponding diagnostic strong interference indices; calculates the corresponding homogeneous interference index between monitoring points based on the similarity characteristics between the interference diagnostic dimension-reduced signals of monitoring points; and obtains each power grid monitoring random factor interference cluster based on the homogeneous interference index between all monitoring points.

[0101] The fault diagnosis module is used to calculate the fault diagnosis confidence factor for each monitoring point based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics at each monitoring point. Based on the fault diagnosis confidence factor of each monitoring point and the homogeneous interference index between each monitoring point and the other monitoring points, it obtains the fault diagnosis benchmark point for each power grid monitoring random factor interference cluster. Based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding power grid monitoring random factor interference cluster fault diagnosis benchmark point within the current monitoring time period, it calculates the fault diagnosis verification factor for each monitoring point in the current monitoring time period. Based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, it performs pre-diagnosis processing on each monitoring point to obtain the fault diagnosis pre-processing result for each monitoring point.

[0102] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0103] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0104] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0105] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for fault diagnosis of distribution networks based on wide-area synchronous measurement data, characterized in that, include: A synchronous phasor measurement device is used to collect each analog power signal at each monitoring point during each monitoring time period. Each analog power signal is then processed by frequency domain transformation to obtain the power time spectrum diagram of each monitoring point. Based on each power time spectrum diagram, the fundamental signal and each harmonic signal of the corresponding simulated power signal are obtained. Based on the regularity characteristics of the fluctuation of each harmonic signal, the corresponding random factor interference index is calculated. Based on the random factor interference index of all harmonic signals of each simulated power signal, the random factor interference harmonics are obtained. Based on the difference characteristics between the frequency of each fundamental signal and the frequency of each random factor interference harmonic, and the signal strength of each random factor interference harmonic of each fundamental signal, the diagnostic strong interference index of each random factor interference harmonic is calculated. Based on all accidental interference harmonics and corresponding diagnostic strong interference indices, construct each interference diagnostic dimension reduction signal of each power time spectrum diagram, calculate the corresponding homogeneous interference index between monitoring points based on the similarity characteristics between the interference diagnostic dimension reduction signals of monitoring points, and obtain each power grid monitoring accidental interference cluster based on the homogeneous interference index between all monitoring points. The fault diagnosis confidence factor for each monitoring point is calculated based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics at each monitoring point. The fault diagnosis benchmark point for each power grid monitoring random factor interference cluster is obtained based on the fault diagnosis confidence factor for each monitoring point and the homogeneous interference index between each monitoring point and the other monitoring points. Based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis benchmark point of the random interference cluster of power grid monitoring within the current monitoring time period, the fault diagnosis verification factor of each monitoring point in the current monitoring time period is calculated. Based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, pre-diagnosis processing is performed on each monitoring point to obtain the fault diagnosis pre-processing result of each monitoring point.

2. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 1, characterized in that, A synchronous phasor measurement device is set up at each monitoring point, and the synchronous phasor measurement device is used to collect each analog power signal at each monitoring point during each monitoring time period. Each analog power signal at each monitoring point is filtered and converted from analog to digital to obtain a digital power signal at each monitoring point. Wavelet transform is then used to perform frequency domain transformation on each digital power signal to obtain a power time spectrum diagram for each monitoring point.

3. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 1, characterized in that, Based on each power time-frequency spectrum diagram, the fundamental signal and each harmonic signal of the corresponding analog power signal are obtained. Based on the regularity characteristics of the fluctuation of each harmonic signal, the corresponding random factor interference index is calculated, including: Based on each power time spectrum diagram, the corresponding fundamental signal and each harmonic signal are extracted from the corresponding analog power signal, and the Hearst exponent of each harmonic signal is calculated. The Petitt mutation point detection algorithm is used to extract all mutation points in each harmonic signal. The standard deviation of the time interval between all mutation points in each harmonic signal and the next adjacent mutation point is used as the mutation interval dispersion index of each harmonic signal. The Hearst exponent of each harmonic signal is added to a preset parameter tuning factor to obtain the adjusted autocorrelation index of each harmonic signal. The ratio of the mutation interval dispersion index of each harmonic signal to the corresponding adjusted autocorrelation index is used as the random factor interference index of each harmonic signal. The preset parameter tuning factor is set to 0.

001.

4. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 3, characterized in that, Random factor interference harmonics are obtained based on the random factor interference index of all harmonic signals for each analog power signal, including: The random factor interference index of all harmonic signals is normalized to obtain the normalized random factor interference index of each harmonic signal. Harmonic signals with a normalized random factor interference index greater than a preset random factor threshold are regarded as random factor interference harmonics.

5. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 4, characterized in that, Based on the differences between the frequencies of each fundamental signal and the corresponding frequencies of various random interference harmonics, and the signal strength of each random interference harmonic of each fundamental signal, the diagnostic strong interference index of each random interference harmonic is calculated, including: The absolute value of the difference between the frequency of each random factor interference harmonic and the frequency of the corresponding fundamental signal is used as the near-frequency interference index of each random factor interference harmonic. The sum of the signal strengths of each random factor interference harmonic at all sampling times is used as the random factor interference intensity of each random factor interference harmonic. The product of the near-frequency interference index of each random factor interference harmonic and the corresponding random factor interference intensity is used as the diagnostic strong interference index of each random factor interference harmonic.

6. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 1, characterized in that, Based on all accidental interference harmonics and their corresponding diagnostic strong interference indices, the following interference diagnostic dimensionality-reduced signals are constructed for each power time-frequency spectrum, including: The normalized value of the sum of the diagnostic strong interference indices of all random factor interference harmonics of each analog power signal is used as the diagnostic interference attention factor of each analog power signal, and the floor value of the product of the number of random factor interference harmonics of each analog power signal and the corresponding diagnostic interference attention factor is used as the number of random factor interference retention terms of each analog power signal. For each random factor interference harmonic of each simulated power signal, each random factor interference harmonic is treated as a row to construct an interference diagnosis matrix for each simulated power signal. The singular value decomposition algorithm is used to decompose each interference diagnosis matrix into a left singular vector, a singular value matrix, and a right singular vector. Based on the number of accidental interference retention terms for each simulated power signal, key singular values ​​are extracted from the singular value matrix of each simulated power signal to obtain the key singular values ​​of each simulated power signal. The left singular vector, key singular values, and right singular vector of each simulated power signal are then processed to obtain the interference diagnosis dimensionality reduction signals for each simulated power signal.

7. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 6, characterized in that, Based on the similarity features between the reduced-dimensional signals of interference diagnosis at monitoring points, the isomorphic interference index between the corresponding monitoring points is calculated, including: Based on the similarity characteristics of the reduced-dimensional interference diagnostic signals corresponding to the same type of simulated power signals between monitoring points, the isomorphic interference index between monitoring points is expressed as follows: in, Let be the isomorphic interference exponent between the i1th and i2th monitoring points, where exp() represents an exponential function with the natural constant as the base, nw is the number of analog power signal types, and Mah(,) is the Manhattan distance between the two signals. The number of interference diagnosis downsampling signals for the j-th simulated power signal at the i1-th monitoring point. The number of interference diagnosis downsampling signals for the j-th simulated power signal at the i2-th monitoring point. This is the k1th interference diagnosis dimension reduction signal of the jth simulated power signal at the i1th monitoring point. It is the k2th interference diagnosis dimension reduction signal of the jth simulated power signal at the i2th monitoring point.

8. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 7, characterized in that, Based on the homogeneous interference index among all monitoring points, the random factor interference clusters for each power grid monitoring point are obtained, including: All monitoring points are clustered based on the same type of interference index among all monitoring points to obtain each power grid monitoring accidental factor interference cluster. The clustering process is designed as follows: the same type of interference index among monitoring points is used as the similarity criterion for the clustering algorithm, and the clustering algorithm is used to cluster all monitoring points.

9. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 1, characterized in that, The fault diagnosis confidence factor for each monitoring point is calculated based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics at each monitoring point, including: The product of the random factor interference index and the corresponding diagnostic strong interference index of each random factor interference harmonic is taken as the corresponding random factor strong interference factor. The sum of the random factor strong interference factors of all random factor interference harmonics at each monitoring point is taken as the distribution fault factor of each monitoring point. The calculation result of the exponential function with the natural constant as the base and the negative number of the distribution fault factor of each monitoring point as the exponent is taken as the fault diagnosis confidence factor of each monitoring point.

10. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 9, characterized in that, Based on the fault diagnosis confidence factor for each monitoring point and the similar interference index between each monitoring point and the remaining monitoring points, the fault diagnosis benchmark point for each power grid monitoring random factor interference cluster is obtained, including: For each monitoring point within a cluster of random interference factors in power grid monitoring, the average value of the same type interference index between each monitoring point and the other monitoring points is used as the same type random factor benchmark factor for each monitoring point. The ratio of the same type random factor benchmark factor to the corresponding fault diagnosis confidence factor for each monitoring point is used as the corresponding same type flawless index. The monitoring point with the largest same type flawless index within each cluster of random interference factors in power grid monitoring is used as the fault diagnosis benchmark point for each cluster of random interference factors in power grid monitoring.

11. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 1, characterized in that, Based on the differences in fault diagnosis confidence factors between each monitoring point and the corresponding fault diagnosis benchmark point of the random interference cluster in power grid monitoring within the current monitoring period, the fault diagnosis verification factor for each monitoring point in the current monitoring period is calculated, including: The difference between the fault diagnosis confidence factor of each monitoring point and the corresponding fault diagnosis benchmark point of the random interference cluster of power grid monitoring within the current monitoring period is used as the fault diagnosis verification factor of each monitoring point in the current monitoring period.

12. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 11, characterized in that, Based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, pre-diagnosis processing is performed on each monitoring point to obtain the fault diagnosis pre-processing results for each monitoring point, including: Based on all the simulated power signals of each monitoring point, signal prediction processing is performed on the simulated power signals of each monitoring point to obtain the predicted simulated power signals of each monitoring point in the next monitoring time period. The signal prediction processing is designed as follows: the normalized value of the fault diagnosis verification factor of each monitoring point is multiplied by the preset number of autoregressive terms as the number of autoregressive terms in the differential autoregressive moving average model, and the differential autoregressive moving average model is used to predict each simulated power signal of each monitoring point in the next monitoring period. Based on the predicted simulated power signals of each monitoring point in the next monitoring period, the fault diagnosis verification factor of each monitoring point in the next monitoring period is calculated. The fault diagnosis verification factor is normalized. When the fault diagnosis verification factor is greater than a preset threshold, the fault diagnosis preprocessing result of the monitoring point is determined to be faulty; otherwise, the fault diagnosis preprocessing result of the monitoring point is determined to be faultless.

13. A distribution network fault diagnosis system based on wide-area synchronous measurement data, characterized in that, include: The data processing module is used to acquire each analog power signal of each monitoring point in each monitoring time period using a synchronous phasor measurement device, and to perform frequency domain conversion processing on each analog power signal to obtain the power time spectrum of each monitoring point. The random factor interference classification module is used to obtain the fundamental signal and various harmonic signals of the corresponding simulated power signal based on each power time-frequency spectrum diagram. It calculates the corresponding random factor interference index based on the regularity characteristics of the fluctuations of each harmonic signal, obtains the random factor interference harmonics based on the random factor interference indices of all harmonic signals of each simulated power signal, calculates the diagnostic strong interference index of each random factor interference harmonic based on the difference characteristics between the frequency of each fundamental signal and the frequencies of the corresponding random factor interference harmonics, and the signal strength of each random factor interference harmonic of each fundamental signal; constructs each interference diagnostic dimension-reduced signal of each power time-frequency spectrum diagram based on all random factor interference harmonics and the corresponding diagnostic strong interference indices; calculates the corresponding homogeneous interference index between monitoring points based on the similarity characteristics between the interference diagnostic dimension-reduced signals of monitoring points; and obtains each power grid monitoring random factor interference cluster based on the homogeneous interference index between all monitoring points. The fault diagnosis module is used to calculate the fault diagnosis confidence factor for each monitoring point based on the random factor interference index and the diagnostic strong interference index of all random factor interference harmonics at each monitoring point. Based on the fault diagnosis confidence factor of each monitoring point and the homogeneous interference index between each monitoring point and the other monitoring points, it obtains the fault diagnosis benchmark point for each power grid monitoring random factor interference cluster. Based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding power grid monitoring random factor interference cluster fault diagnosis benchmark point within the current monitoring time period, it calculates the fault diagnosis verification factor for each monitoring point in the current monitoring time period. Based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all simulated power signals of each monitoring point, it performs pre-diagnosis processing on each monitoring point to obtain the fault diagnosis pre-processing result for each monitoring point.

14. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-12.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-12.

Images