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

Abstract

The invention discloses a power distribution network fault diagnosis method and system based on wide-area synchronous measurement data, and the method comprises the steps: firstly analyzing the similarity degree of the power signals of a power distribution network interfered by accidental factors, carrying out the aggregation of micro-grids with similar accidental interference factors, and obtaining a monitoring even interference cluster of each power grid; secondly, obtaining a fault diagnosis reference point of each power grid monitoring even factor interference cluster by combining the interference degree of accidental factors on the power signals of the power distribution network; analyzing the difference between each monitoring point and the fault diagnosis reference point of the corresponding power grid monitoring even cause interference cluster, and calculating the fault diagnosis check factor of each monitoring point in the current monitoring time period; and finally, performing pre-diagnosis processing on each monitoring point based on the fault diagnosis check factor of each monitoring point in the current monitoring time period and all the analog power signals of each monitoring point to obtain a fault diagnosis preprocessing result of each monitoring point. According to the invention, the real-time performance and reliability of fault diagnosis of the power distribution network are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent distribution network fault diagnosis, and in particular, relates to a distribution network fault diagnosis method and system based on wide-area synchronous measurement data. Background Art

[0002] Most distribution networks are exposed outdoors and are extremely vulnerable to factors such as severe weather, natural disasters, environmental pollution, external damage, and construction quality. As a result, accidents and failures often occur in distribution network lines. The most direct consequence of distribution network failures is local or large-scale power outages, which are likely to damage equipment in the distribution lines, causing inconvenience to users' lives and affecting industrial production and commercial operations. Therefore, timely diagnosis of faults in distribution networks is of extremely important practical significance.

[0003] Existing technologies, such as wide-area synchronized phasor measurement, utilize GPS information to synchronously sample analog voltage and current signals, thereby determining their amplitude and phase angle. This technology offers high-precision, remote synchronized phasor measurement, high-speed communication, and rapid response, making it ideal for fault diagnosis in distribution networks. However, since wide-area synchronized phasor measurement generates a large amount of data, efficiently processing this data and extracting useful information to provide robust data analysis capabilities to support real-time or near-real-time fault diagnosis presents a challenge. Summary of the Invention

[0004] To address the deficiencies in the prior art, the present invention provides a distribution network fault diagnosis method and system based on wide-area synchronous measurement data, which improves the efficiency of distribution network fault diagnosis and enhances the real-time performance and reliability of distribution network fault diagnosis.

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

[0006] The present invention proposes a distribution network fault diagnosis method based on wide-area synchronous measurement data, comprising:

[0007] A synchronized phasor measurement device is used to collect the analog power signals of each monitoring point in each monitoring time period, and each analog power signal is converted into a frequency domain to obtain the power time-frequency spectrum of each monitoring point.

[0008] Based on each power time-frequency spectrum, 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 accidental factor interference index is calculated; based on the accidental factor interference index of all harmonic signals of each analog power signal, the accidental factor interference harmonic is obtained; based on the difference characteristics between the frequency of each fundamental signal and the frequency of each corresponding accidental factor interference harmonic, and the signal strength of each accidental factor interference harmonic of each fundamental signal, the diagnostic strong interference index of each accidental factor interference harmonic is calculated;

[0009] Based on all the accidental interference harmonics and the corresponding diagnostic strong interference index, each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum is constructed. Based on the similarity characteristics between the interference diagnosis dimensionality reduction signals of the monitoring points, the corresponding same-type interference index between the monitoring points is calculated. Based on the same-type interference index between all monitoring points, the accidental interference cluster of each power grid monitoring is obtained;

[0010] The fault diagnosis confidence factor of each monitoring point is calculated based on the accidental factor interference index and the diagnostic strong interference index of all accidental factor interference harmonics of each monitoring point. The fault diagnosis reference point of each power grid monitoring accidental 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.

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

[0012] Preferably, a synchronized phasor measurement device is provided at each monitoring point, and each analog power signal of each monitoring point is collected by the synchronized phasor measurement device during each monitoring time period;

[0013] Each analog power signal of each monitoring point is filtered and analog / digital converted to obtain each digital power signal of each monitoring point. Each digital power signal is converted into the frequency domain using wavelet transform to obtain each power time-frequency spectrum of each monitoring point.

[0014] Preferably, obtaining the fundamental signal and each harmonic signal of the corresponding analog power signal based on each power time-frequency spectrum, and calculating the corresponding accidental factor interference index based on the regularity characteristics of the fluctuation of each harmonic signal, includes:

[0015] According to each power time-frequency spectrum, the corresponding fundamental signal and each harmonic signal are extracted from the corresponding analog power signal, and the Hurst index of each harmonic signal is calculated; the Pettitt mutation point detection algorithm is used to extract all mutation points in each harmonic signal, and the standard deviation of the time intervals between all mutation points and the adjacent subsequent mutation points in each harmonic signal is used as the mutation interval dispersion index of each harmonic signal, the Hurst index of each harmonic signal plus a preset parameter adjustment factor is used as the adjusted autocorrelation index of each harmonic signal, and the ratio of the mutation interval dispersion index of each harmonic signal to the corresponding adjusted autocorrelation index is used as the accidental factor interference index of each harmonic signal;

[0016] Among them, the preset parameter adjustment factor value is 0.001.

[0017] Preferably, obtaining the accidental interference harmonics based on the accidental interference index of all harmonic signals of each simulated power signal includes:

[0018] The accidental factor interference indexes of all harmonic signals are normalized to obtain the normalized accidental factor interference index of each harmonic signal, and the harmonic signals whose normalized accidental factor interference index is greater than a preset accidental factor threshold are regarded as accidental factor interference harmonics.

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

[0020] The absolute value of the difference between the frequency of each accidental factor interference harmonic and the frequency of the corresponding fundamental signal is taken as the close-frequency interference index of each accidental factor interference harmonic, the sum of the signal strength of each accidental factor interference harmonic at all sampling moments is taken as the accidental factor interference intensity of each accidental factor interference harmonic, and the product of the close-frequency interference index of each accidental factor interference harmonic and the corresponding accidental factor interference intensity is taken as the diagnostic strong interference index of each accidental factor interference harmonic.

[0021] Preferably, constructing each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum based on all accidental interference harmonics and corresponding diagnostic strong interference indices includes:

[0022] The normalized value of the sum of the diagnostic strong interference indices of all accidental interference harmonics of each simulated power signal is used as the diagnostic interference attention factor of each simulated power signal, and the upward rounded value of the product of the number of accidental interference harmonics of each simulated power signal and the corresponding diagnostic interference attention factor is used as the number of accidental interference retained items of each simulated power signal;

[0023] For each accidental interference harmonic of each simulated power signal, each accidental interference harmonic is taken as a row to construct an interference diagnosis matrix for each simulated power signal, and a 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 retained items of accidental interference factors of each analog power signal, the singular value matrix of each analog power signal is subjected to key singular value extraction processing to obtain the key singular value of each analog power signal, and the left singular vector, key singular value and right singular vector of each analog power signal are processed to obtain the interference diagnosis dimensionality reduction signals of each analog power signal.

[0025] Preferably, the similarity characteristics between the interference diagnosis dimensionality reduction signals of the monitoring points are used to calculate the corresponding isotype interference index between the monitoring points, including:

[0026] Based on the similarity characteristics of the interference diagnosis dimensionality reduction signals corresponding to the same type of simulated power signals between monitoring points, the same type of interference index between monitoring points is expressed as follows:

[0027]

[0028] in, is the same type interference index between the i1th monitoring point and the i2th monitoring point, exp() represents the exponential function with the natural constant as the base, nw is the number of types of simulated power signals, Mah(,) is the Manhattan distance between the two signals, is the number of interference diagnosis dimensionality reduction signals of the j-th simulated power signal at the i1-th monitoring point, is the number of interference diagnosis dimensionality reduction signals of the j-th simulated power signal at the i2-th monitoring point, is the k1th interference diagnosis dimension reduction signal of the jth analog power signal at the i1th monitoring point, It is the k2th interference diagnosis dimensionality reduction signal of the jth analog power signal at the i2th monitoring point.

[0029] Preferably, obtaining the interference cluster of accidental factors of each power grid monitoring based on the same type interference index between all monitoring points includes:

[0030] All monitoring points are clustered based on the homogeneous interference index between all monitoring points to obtain the interference clusters of accidental factors in each power grid monitoring. The clustering process is designed as follows: the homogeneous interference index between monitoring points is used as the similarity measurement criterion of the clustering algorithm, and the clustering algorithm is used to cluster all monitoring points.

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

[0032] The product of the accidental factor interference index of each accidental factor interference harmonic and the corresponding diagnostic strong interference index is taken as the corresponding accidental factor strong interference factor, the sum of the accidental factor strong interference factors of all accidental factor interference harmonics of each monitoring point is taken as the distribution fault factor of each monitoring point, and the calculation result of the exponential function with a natural constant as the base and the opposite 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, obtaining the fault diagnosis reference point of each grid monitoring accidental factor interference cluster based on the fault diagnosis confidence factor of each monitoring point and the same type interference index between each monitoring point and the remaining monitoring points includes:

[0034] For each monitoring point in each power grid monitoring accidental factor interference cluster, the average value of the same-type interference index between each monitoring point and the other monitoring points is used as the same-type accidental factor benchmark factor of each monitoring point, the ratio of the same-type accidental factor benchmark factor of each monitoring point to the corresponding fault diagnosis confidence factor is used as the corresponding same-type completeness index, and the monitoring point with the largest same-type completeness index in each power grid monitoring accidental factor interference cluster is used as the fault diagnosis benchmark point of each power grid monitoring accidental factor interference cluster.

[0035] Preferably, the fault diagnosis calibration factor of each monitoring point in the current monitoring time period is calculated based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the fault diagnosis reference point of the corresponding power grid monitoring accidental factor interference cluster in the current monitoring time period, including:

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

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

[0038] Performing signal prediction processing on the analog power signals of each monitoring point based on all analog power signals of each monitoring point to obtain each predicted analog power signal of each monitoring point in the next monitoring time period;

[0039] The signal prediction process is designed as follows: the normalized value of the fault diagnosis calibration 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 time period;

[0040] Based on the predicted simulated power signals of each monitoring point in the next monitoring time period, the fault diagnosis calibration factor of each monitoring point in the next monitoring time period is calculated, and the fault diagnosis calibration factor is normalized. When the fault diagnosis calibration factor is greater than a preset threshold, the fault diagnosis preprocessing result of the monitoring point is determined to be a fault; otherwise, the fault diagnosis preprocessing result of the monitoring point is determined to be no fault.

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

[0042] A data processing module is used to collect each analog power signal of each monitoring point in each monitoring time period using a synchronized phasor measurement device, perform frequency domain conversion processing on each analog power signal, and obtain each power time-frequency spectrum of each monitoring point;

[0043] An accidental factor interference classification module is used to obtain the fundamental signal and each harmonic signal of the corresponding analog power signal based on each power time-frequency spectrum, calculate the corresponding accidental factor interference index based on the regularity characteristics of the fluctuation of each harmonic signal, obtain the accidental factor interference harmonic based on the accidental factor interference index of all harmonic signals of each analog power signal, calculate the diagnostic strong interference index of each accidental factor interference harmonic based on the difference characteristics between the frequency of each fundamental signal and the frequency of the corresponding each accidental factor interference harmonic, and the signal strength of each accidental factor interference harmonic of each fundamental signal; construct each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum based on all accidental factor interference harmonics and the corresponding diagnostic strong interference index, calculate the corresponding same-type interference index between monitoring points based on the similarity characteristics between the interference diagnosis dimensionality reduction signals of the monitoring points, and obtain each power grid monitoring accidental factor interference cluster based on the same-type interference index between all monitoring points;

[0044] A fault diagnosis module is used to calculate the fault diagnosis confidence factor of each monitoring point based on the accidental factor interference index and the diagnostic strong interference index of all accidental factor interference harmonics of each monitoring point, obtain the fault diagnosis reference point of each power grid monitoring accidental factor interference cluster based on the fault diagnosis confidence factor of each monitoring point and the same-type interference index between each monitoring point and the remaining monitoring points; calculate the fault diagnosis calibration factor of each monitoring point in the current monitoring time period based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the fault diagnosis benchmark point of the corresponding power grid monitoring accidental factor interference cluster in the current monitoring time period, perform pre-diagnosis processing on each monitoring point based on the fault diagnosis calibration factor of each monitoring point in the current monitoring time period and all analog power signals of each monitoring point, and obtain the fault diagnosis pre-processing result of each monitoring point.

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

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

[0047] The beneficial effects of the present invention are as follows:

[0048] (1) First, the degree of interference of the power signal of the distribution network by the environment is analyzed. Based on each power time-frequency spectrum, 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 accidental factor interference index is calculated. Based on the accidental factor interference index of all harmonic signals of each analog power signal, the accidental factor interference harmonic is obtained. Based on the difference characteristics between the frequency of each fundamental signal and the frequency of the corresponding accidental factor interference harmonic and the signal strength of each accidental factor interference harmonic of each fundamental signal, the diagnostic strong interference index of each accidental factor interference harmonic is calculated. Based on all the accidental factor interference harmonics and the corresponding diagnostic strong interference index, each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum is constructed; based on the similarity characteristics between the interference diagnosis dimensionality reduction signals of the monitoring points, the corresponding same-type interference index between the monitoring points is calculated. Based on the same-type interference index between all monitoring points, each grid monitoring accidental factor interference cluster is obtained. Microgrids with similar accidental interference factors are aggregated to avoid the influence of accidental factors such as the environment on the fault diagnosis of the microgrid. Subsequently, each grid monitoring accidental factor interference cluster is analyzed to improve 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 accidental factor interference index and the diagnostic strong interference index of all accidental factor interference harmonics of each monitoring point. The fault diagnosis reference point of each grid monitoring accidental 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 calibration factor of each monitoring point in the current monitoring time period is calculated based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the fault diagnosis reference point of the corresponding grid monitoring accidental factor interference cluster in the current monitoring time period. By comparing the monitoring point with the reference point with a lower degree of interference in the grid monitoring accidental factor interference cluster to which it belongs, the reliability of fault diagnosis of the microgrid is improved.

[0050] (3) Pre-diagnostic processing is performed on each monitoring point based on the fault diagnosis calibration factor of each monitoring point in the current monitoring time period and all analog power signals of each monitoring point to obtain the fault diagnosis pre-processing result of each monitoring point. The number of autoregressive terms of the differential autoregressive moving average model is adaptively determined according to the fault diagnosis calibration factor of the monitoring point. While ensuring the accuracy of the prediction of the analog power signal of the monitoring point, the real-time performance of the prediction of the analog power signal of the monitoring point is improved, thereby improving the real-time performance of the fault diagnosis of the microgrid;

[0051] (4) By predicting the analog power signal of the microgrid in the next monitoring time period and performing pre-fault diagnosis of the microgrid based on the predicted analog power signal, it is possible to pay attention to the abnormality of the microgrid before the microgrid fault occurs or in the early stage of the fault, further improving the real-time performance of fault diagnosis of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flow chart of a distribution network fault diagnosis method based on wide-area synchronous measurement data proposed by the present invention;

[0053] Figure 2 Schematic diagram of obtaining accidental interference clusters for power grid monitoring according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] The present invention proposes a distribution network fault diagnosis method based on wide-area synchronous measurement data. Figure 1 and Figure 2 As shown, including:

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

[0057] The implementation of wide-area synchronized phasor measurement technology relies primarily on the collaborative work of multiple key components and technologies, including synchronized phasor measurement devices, satellite timing systems, high-speed communication networks, and data processing and analysis centers. Therefore, before using wide-area synchronized phasor measurement technology to diagnose distribution network faults, synchronized phasor measurement devices must be deployed at each monitoring point to collect relevant power data from the distribution network.

[0058] Specifically, step 1 includes:

[0059] Step 1.1: Setting a synchronized phasor measurement device at each monitoring point, and using the synchronized phasor measurement device to collect analog power signals at each monitoring point during each monitoring period;

[0060] In step 1.2, each analog power signal of each monitoring point is filtered and analog / digital converted to obtain each digital power signal of each monitoring point, and each digital power signal is converted into the frequency domain using wavelet transform to obtain each power time-frequency spectrum of each monitoring point.

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

[0062] Step 2: Based on each power time-frequency spectrum, obtain the fundamental signal and each harmonic signal of the corresponding analog power signal, calculate the corresponding accidental factor interference index based on the regularity of the fluctuation of each harmonic signal, obtain the accidental factor interference harmonic based on the accidental factor interference index of all harmonic signals of each analog power signal, and calculate the diagnostic strong interference index of each accidental factor interference harmonic based on the difference characteristics between the frequency of each fundamental signal and the frequency of the corresponding accidental factor interference harmonic and the signal strength of each accidental factor interference harmonic of each fundamental signal.

[0063] The fundamental signal extracted from the analog power signal at each monitoring point represents the original analog power signal, while the harmonic signal extracted from the analog power signal at each monitoring point represents the interference signal affecting the analog power signal. Interference signals caused by accidental factors such as the environment often exhibit irregular fluctuations and are difficult to predict; however, interference signals caused by distribution network faults often exhibit regular fluctuations and are predictable. By analyzing the regularity of the fluctuations of the interference signals affecting the analog power signal, it is possible to distinguish interference signals caused by accidental factors such as the environment from 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 according to each power time-frequency spectrum, and calculate the Hurst index of each harmonic signal; use the Pettitt mutation point detection algorithm to extract all mutation points in each harmonic signal, and use the standard deviation of the time intervals between all mutation points and the adjacent subsequent mutation points in each harmonic signal as the mutation interval discrete index of each harmonic signal; add the preset parameter adjustment factor to the Hurst index of each harmonic signal as the adjusted autocorrelation index of each harmonic signal; and use the ratio of the mutation interval discrete index of each harmonic signal to the corresponding adjusted autocorrelation index as the accidental factor interference index of each harmonic signal.

[0066] The preset parameter adjustment factor is a manually preset value. In this embodiment, the preset parameter adjustment factor is 0.001. As for the value of the preset parameter adjustment factor, as other implementation methods, the implementer can select it at will, and this embodiment does not impose any special restrictions on this.

[0067] The Hurst exponent can reflect the autocorrelation of the time series, and the mutation interval dispersion index represents the degree of dispersion of the time intervals between each mutation point and adjacent mutation points in the harmonic signal. When the autocorrelation of the harmonic signal is stronger and the time intervals between each mutation point and adjacent mutation points in the harmonic signal are closer, it means that the fluctuation of the harmonic signal is more regular, the harmonic signal is more likely to be an interference signal caused by a distribution network fault, and the value of the accidental factor interference index is smaller; when the autocorrelation of the harmonic signal is weaker and the difference in the time intervals between each mutation point and adjacent mutation points in the harmonic signal is greater, the harmonic signal is more likely to be an interference signal corresponding to accidental factors such as the environment, and the value of the accidental factor interference index is larger.

[0068] Step 2.2, normalize the accidental factor interference index of all harmonic signals to obtain the normalized accidental factor interference index of each harmonic signal, and take the harmonic signal whose normalized accidental factor interference index is greater than the preset accidental factor threshold as the accidental factor interference harmonic.

[0069] When diagnosing microgrid faults, by aggregating microgrids with similar accidental interference factors, the impact of accidental factors such as the environment on the fault diagnosis of the microgrid can be avoided, and the real-time performance of the microgrid fault diagnosis can be improved.

[0070] In step 2.3, the absolute value of the difference between the frequency of each accidental factor interference harmonic and the frequency of the corresponding fundamental signal is used as the close-frequency interference index of each accidental factor interference harmonic, the sum of the signal strength of each accidental factor interference harmonic at all sampling moments is used as the accidental factor interference intensity of each accidental factor interference harmonic, and the product of the close-frequency interference index of each accidental factor interference harmonic and the corresponding accidental factor interference intensity is used as the diagnostic strong interference index of each accidental factor interference harmonic.

[0071] When the difference between the frequency of the accidental interference harmonic and the frequency of the corresponding fundamental signal is smaller and the signal strength of the accidental interference harmonic is stronger, it means that the accidental 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: Based on all the accidental interference harmonics and the corresponding diagnostic strong interference index, each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum is constructed; based on the similarity characteristics between the interference diagnosis dimensionality reduction signals of the monitoring points, the corresponding same-type interference index between the monitoring points is calculated; based on the same-type interference index between all monitoring points, the accidental interference cluster of each power grid monitoring is obtained.

[0073] Since most of the distribution network is exposed in the wild, it is extremely vulnerable to factors such as bad 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 harmonics of each accidental factor interference in each simulated power signal.

[0074] Specifically, step 3 includes:

[0075] In step 3.1, the normalized value of the sum of the diagnostic strong interference indices of all accidental interference harmonics of each simulated power signal is used as the diagnostic interference attention factor of each simulated power signal, and the rounded-up value of the product of the number of accidental interference harmonics of each simulated power signal and the corresponding diagnostic interference attention factor is used as the number of retained accidental interference items of each simulated power signal.

[0076] In step 3.2, for each accidental interference harmonic of each simulated power signal, each accidental interference harmonic is taken as a row, and an interference diagnosis matrix of each simulated power signal is constructed. 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] The singular value decomposition algorithm can reduce the dimension of data through eigenvalue decomposition. The singular value matrix has singular values ​​on the diagonal and all other positions are 0.

[0078] Step 3.3, based on the number of retained items of accidental interference factors of each analog power signal, the singular value matrix of each analog power signal is subjected to key singular value extraction processing to obtain the key singular value of each analog power signal, and the left singular vector, key singular value and right singular vector of each analog power signal are processed to obtain the interference diagnosis dimensionality reduction signals of each analog power signal.

[0079] In step 3.4, based on the similarity characteristics of the interference diagnosis dimensionality reduction signals corresponding to the same type of simulated power signals between the monitoring points, the same type of interference index between the monitoring points is expressed as follows:

[0080]

[0081] in, is the same type interference index between the i1th monitoring point and the i2th monitoring point, exp() represents the exponential function with the natural constant as the base, nw is the number of types of simulated power signals, Mah(,) is the Manhattan distance between the two signals, is the number of interference diagnosis dimensionality reduction signals of the j-th simulated power signal at the i1-th monitoring point, is the number of interference diagnosis dimensionality reduction signals of the j-th simulated power signal at the i2-th monitoring point, is the k1th interference diagnosis dimension reduction signal of the jth analog power signal at the i1th monitoring point, It is the k2th interference diagnosis dimensionality reduction signal of the jth analog power signal at the i2th monitoring point.

[0082] When the Manhattan distance of each interference diagnosis dimensionality reduction signal corresponding to the same type of simulated power signal between monitoring points is smaller, it means that the monitoring point is more likely to be affected by the same accidental factors, and the same type of interference index value is larger.

[0083] In step 3.5, all monitoring points are clustered based on the homologous interference index between all monitoring points to obtain the interference clusters of accidental factors of each power grid monitoring. The clustering process is designed as follows: the homologous interference index between the monitoring points is used as the similarity measurement criterion of the clustering algorithm, and the clustering algorithm is used to cluster all monitoring points.

[0084] Step 4: Calculate the fault diagnosis confidence factor of each monitoring point based on the accidental factor interference index and diagnostic strong interference index of all accidental factor interference harmonics of each monitoring point, and obtain the fault diagnosis benchmark point of each power grid monitoring accidental factor interference cluster based on the fault diagnosis confidence factor of each monitoring point and the same-type interference index between each monitoring point and the remaining monitoring points.

[0085] Since the grid monitoring accidental factor interference cluster contains multiple monitoring points affected by the same accidental factors, the fault diagnosis of the distribution network can be achieved by comparing the monitoring points with the benchmark points with lower interference levels in the grid monitoring accidental factor interference cluster to which they belong.

[0086] Specifically, step 4 includes:

[0087] In step 4.1, the product of the accidental factor interference index of each accidental factor interference harmonic and the corresponding diagnostic strong interference index is used as the corresponding accidental factor strong interference factor, the sum of the accidental factor strong interference factors of all accidental factor interference harmonics of each monitoring point is used as the distribution fault factor of each monitoring point, and the calculation result of the exponential function with a natural constant as the base and the opposite number of the distribution fault factor of each monitoring point as the exponent is used as the fault diagnosis confidence factor of each monitoring point.

[0088] Step 4.2, for each monitoring point in each power grid monitoring accidental factor interference cluster, the average value of the same-type interference index between each monitoring point and the other monitoring points is used as the same-type accidental factor benchmark factor of each monitoring point, the ratio of the same-type accidental factor benchmark factor of each monitoring point to the corresponding fault diagnosis confidence factor is used as the corresponding same-type completeness index, and the monitoring point with the largest same-type completeness index in each power grid monitoring accidental factor interference cluster is used as the fault diagnosis benchmark point of each power grid monitoring accidental factor interference cluster.

[0089] The smaller the possibility of failure at a monitoring point, and the closer the impact of accidental factors on the monitoring point and other monitoring points are, the more suitable it is to use the monitoring point as a benchmark point for distribution network fault diagnosis.

[0090] Step 5: Calculate the fault diagnosis calibration factor of each monitoring point in the current monitoring time period based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis reference point of the grid monitoring accidental factor interference cluster in the current monitoring time period, perform pre-diagnosis processing on each monitoring point based on the fault diagnosis calibration factor of each monitoring point in the current monitoring time period and all analog power signals of each monitoring point, and obtain the fault diagnosis pre-processing result of each monitoring point.

[0091] Since wide-area synchronized phasor measurement technology generates a large amount of data, it is difficult to ensure the real-time performance of microgrid fault diagnosis if the fault diagnosis is performed directly on the microgrid based on the analog power signal of the microgrid in the current monitoring time period. By predicting the analog power signal of the microgrid in the next monitoring time period and performing pre-fault diagnosis on the microgrid based on the predicted analog power signal, it is possible to pay attention to the abnormality of the microgrid before the fault occurs or in the early stage of the fault, thereby improving the real-time performance of microgrid fault diagnosis.

[0092] Specifically, step 5 includes:

[0093] Step 5.1: The difference in fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis reference point of the grid monitoring accidental factor interference cluster in the current monitoring period is used as the fault diagnosis verification factor of each monitoring point in the current monitoring period.

[0094] Step 5.2: performing signal prediction processing on the analog power signal of each monitoring point based on all analog power signals of each monitoring point to obtain each predicted analog power signal of each monitoring point in the next monitoring time period;

[0095] Among them, the signal prediction processing is designed as follows: the normalized value of the fault diagnosis calibration factor of each monitoring point is multiplied by the preset number of autoregressive terms as the number of autoregressive terms of 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 time period.

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

[0097] Step 5.3, calculate the fault diagnosis calibration factor of each monitoring point in the next monitoring time period based on each predicted simulated power signal of each monitoring point in the next monitoring time period, normalize the fault diagnosis calibration factor, and when the fault diagnosis calibration factor is greater than the preset threshold, determine that the fault diagnosis preprocessing result of the monitoring point is a fault; otherwise, determine that the fault diagnosis preprocessing result of the monitoring point is no fault.

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

[0099] A data processing module is used to collect each analog power signal of each monitoring point in each monitoring time period using a synchronized phasor measurement device, perform frequency domain conversion processing on each analog power signal, and obtain each power time-frequency spectrum of each monitoring point;

[0100] An accidental factor interference classification module is used to obtain the fundamental signal and each harmonic signal of the corresponding analog power signal based on each power time-frequency spectrum, calculate the corresponding accidental factor interference index based on the regularity characteristics of the fluctuation of each harmonic signal, obtain the accidental factor interference harmonic based on the accidental factor interference index of all harmonic signals of each analog power signal, calculate the diagnostic strong interference index of each accidental factor interference harmonic based on the difference characteristics between the frequency of each fundamental signal and the frequency of the corresponding each accidental factor interference harmonic, and the signal strength of each accidental factor interference harmonic of each fundamental signal; construct each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum based on all accidental factor interference harmonics and the corresponding diagnostic strong interference index, calculate the corresponding same-type interference index between monitoring points based on the similarity characteristics between the interference diagnosis dimensionality reduction signals of the monitoring points, and obtain each power grid monitoring accidental factor interference cluster based on the same-type interference index between all monitoring points;

[0101] A fault diagnosis module is used to calculate the fault diagnosis confidence factor of each monitoring point based on the accidental factor interference index and the diagnostic strong interference index of all accidental factor interference harmonics of each monitoring point, obtain the fault diagnosis reference point of each power grid monitoring accidental factor interference cluster based on the fault diagnosis confidence factor of each monitoring point and the same-type interference index between each monitoring point and the remaining monitoring points; calculate the fault diagnosis calibration factor of each monitoring point in the current monitoring time period based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the fault diagnosis benchmark point of the corresponding power grid monitoring accidental factor interference cluster in the current monitoring time period, perform pre-diagnosis processing on each monitoring point based on the fault diagnosis calibration factor of each monitoring point in the current monitoring time period and all analog power signals of each monitoring point, and obtain the fault diagnosis pre-processing result of each monitoring point.

[0102] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0103] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

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

[0105] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state 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++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of 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., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A distribution network fault diagnosis method based on wide-area synchronous measurement data, characterized in that: include: A synchronous phasor measurement device is used to collect each analog power signal of each monitoring point in each monitoring time period, and each analog power signal is converted into a frequency domain to obtain each power time-frequency spectrum of each monitoring point; Based on each power time-frequency spectrum, 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 accidental factor interference index is calculated; based on the accidental factor interference index of all harmonic signals of each analog power signal, the accidental factor interference harmonic is obtained; based on the difference characteristics between the frequency of each fundamental signal and the frequency of each corresponding accidental factor interference harmonic, and the signal strength of each accidental factor interference harmonic of each fundamental signal, the diagnostic strong interference index of each accidental factor interference harmonic is calculated; Based on all the accidental interference harmonics and the corresponding diagnostic strong interference index, each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum is constructed. Based on the similarity characteristics between the interference diagnosis dimensionality reduction signals of the monitoring points, the corresponding same-type interference index between the monitoring points is calculated. Based on the same-type interference index between all the monitoring points, the accidental interference clusters of each power grid monitoring are obtained. The fault diagnosis confidence factor of each monitoring point is calculated based on the accidental factor interference index and the diagnostic strong interference index of all accidental factor interference harmonics of each monitoring point, and the fault diagnosis reference point of each power grid monitoring accidental 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 other monitoring points; Based on the difference characteristics of the fault diagnosis confidence factors between each monitoring point and the corresponding fault diagnosis benchmark point of the grid monitoring accidental factor interference cluster in the current monitoring time period, the fault diagnosis calibration factor of each monitoring point in the current monitoring time period is calculated, and pre-diagnosis processing is performed on each monitoring point based on the fault diagnosis calibration 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 preprocessing result of each monitoring point.

2. The distribution network fault diagnosis method based on wide-area synchronous measurement data according to claim 1 is characterized in that: A synchronized phasor measurement device is set at each monitoring point, and each analog power signal of each monitoring point is collected by the synchronized phasor measurement device in each monitoring time period; Each analog power signal of each monitoring point is filtered and analog / digital converted to obtain each digital power signal of each monitoring point. Each digital power signal is converted into frequency domain by wavelet transform to obtain each power time-frequency spectrum of each monitoring point.

3. The method for diagnosing distribution network faults based on wide-area synchronous measurement data according to claim 1, characterized in that: Based on each power time-frequency spectrum, the fundamental signal and each harmonic signal of the corresponding analog power signal are obtained, and based on the regular characteristics of the fluctuation of each harmonic signal, the corresponding accidental factor interference index is calculated, including: According to each power time-frequency spectrum diagram, the corresponding fundamental signal and each harmonic signal are extracted from the corresponding analog power signal, and the Hurst index of each harmonic signal is calculated; the Pettitt mutation point detection algorithm is used to extract all mutation points in each harmonic signal, and the standard deviation of the time intervals between all mutation points and the adjacent subsequent mutation points in each harmonic signal is used as the mutation interval discrete index of each harmonic signal, the Hurst index of each harmonic signal plus the preset parameter adjustment factor is used as the adjusted autocorrelation index of each harmonic signal, and the ratio of the mutation interval discrete index of each harmonic signal to the corresponding adjusted autocorrelation index is used as the accidental factor interference index of each harmonic signal; Among them, the preset parameter adjustment factor value is 0.

001.

4. The method for diagnosing distribution network faults based on wide-area synchronous measurement data according to claim 3, characterized in that: Obtaining accidental interference harmonics based on the accidental interference index of all harmonic signals of each simulated power signal, including: The accidental factor interference indexes of all harmonic signals are normalized to obtain the normalized accidental factor interference index of each harmonic signal, and the harmonic signal whose normalized accidental factor interference index is greater than a preset accidental factor threshold is taken as an accidental factor interference harmonic.

5. The method for diagnosing distribution network faults based on wide-area synchronous measurement data according to claim 4, characterized in that: Based on the difference characteristics between the frequency of each fundamental wave signal and the frequency of each corresponding harmonic interference of accidental factors, and the signal strength of each harmonic interference of accidental factors of each fundamental wave signal, the diagnostic strong interference index of each harmonic interference of accidental factors is calculated, including: The absolute value of the difference between the frequency of each accidental factor interference harmonic and the frequency of the corresponding fundamental signal is taken as the close-frequency interference index of each accidental factor interference harmonic, the sum of the signal strengths of each accidental factor interference harmonic at all sampling moments is taken as the accidental factor interference intensity of each accidental factor interference harmonic, and the product of the close-frequency interference index of each accidental factor interference harmonic and the corresponding accidental factor interference intensity is taken as the diagnostic strong interference index of each accidental factor interference harmonic.

6. The method for diagnosing distribution network faults based on wide-area synchronous measurement data according to claim 1, characterized in that: Based on all the accidental interference harmonics and the corresponding diagnostic strong interference index, each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum is constructed, including: The normalized value of the sum of the diagnostic strong interference indexes of all the accidental interference harmonics of each simulated power signal is used as the diagnostic interference attention factor of each simulated power signal, and the rounded-up value of the product of the number of accidental interference harmonics of each simulated power signal and the corresponding diagnostic interference attention factor is used as the number of accidental interference retained items of each simulated power signal; For each accidental interference harmonic of each simulated power signal, each accidental interference harmonic is taken as a row to construct an interference diagnosis matrix of each simulated power signal, and a 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 retained items of accidental interference factors of each simulated power signal, the singular value matrix of each simulated power signal is processed with key singular values ​​to obtain the key singular values ​​of each simulated power signal, and the left singular vector, key singular value and right singular vector of each simulated power signal are processed to obtain the interference diagnosis dimensionality reduction signals of each simulated power signal.

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

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

9. The method for diagnosing distribution network faults based on wide-area synchronous measurement data according to claim 1, characterized in that: The fault diagnosis confidence factor of each monitoring point is calculated based on the accidental factor interference index and the diagnostic strong interference index of all accidental factor interference harmonics of each monitoring point, including: The product of the accidental factor interference index of each accidental factor interference harmonic and the corresponding diagnostic strong interference index is taken as the corresponding accidental factor strong interference factor, the sum of the accidental factor strong interference factors of all accidental factor interference harmonics of each monitoring point is taken as the distribution fault factor of each monitoring point, and the calculation result of the exponential function with a natural constant as the base and the opposite 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 method for diagnosing distribution network faults based on wide-area synchronous measurement data according to claim 9, characterized in that: Based on the fault diagnosis confidence factor of each monitoring point and the same type interference index between each monitoring point and other monitoring points, the fault diagnosis benchmark point of each grid monitoring accidental factor interference cluster is obtained, including: For each monitoring point in each power grid monitoring accidental factor interference cluster, the average value of the same-type interference index between each monitoring point and the other monitoring points is taken as the same-type accidental factor benchmark factor of each monitoring point, the ratio of the same-type accidental factor benchmark factor of each monitoring point to the corresponding fault diagnosis confidence factor is taken as the corresponding same-type completeness index, and the monitoring point with the largest same-type completeness index in each power grid monitoring accidental factor interference cluster is taken as the fault diagnosis benchmark point of each power grid monitoring accidental factor interference cluster.

11. The method for diagnosing distribution network faults based on wide-area synchronous measurement data according to claim 1, characterized in that: Based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis reference point of the grid monitoring accidental factor 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: The difference in fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis reference point of the grid monitoring accidental factor interference cluster in the current monitoring time period is used as the fault diagnosis verification factor of each monitoring point in the current monitoring time period.

12. The method for diagnosing distribution network faults 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 the simulated power signals of each monitoring point, each monitoring point is pre-diagnosed and processed to obtain the fault diagnosis pre-processing result of each monitoring point, including: Based on all the analog power signals of each monitoring point, signal prediction processing is performed on the analog power signal of each monitoring point to obtain each predicted analog power signal 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 of 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 time period; The fault diagnosis verification factor of each monitoring point in the next monitoring time period is calculated based on each predicted simulated power signal of each monitoring point in the next monitoring time period, and 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 a fault; otherwise, the fault diagnosis preprocessing result of the monitoring point is determined to be no fault.

13. A distribution network fault diagnosis system based on wide-area synchronous measurement data, characterized in that: include: A data processing module is used to collect each analog power signal of each monitoring point in each monitoring time period by using a synchronous phasor measurement device, perform frequency domain conversion processing on each analog power signal, and obtain each power time-frequency spectrum diagram of each monitoring point; An accidental factor interference classification module is used to obtain the fundamental signal and each harmonic signal of the corresponding analog power signal based on each power time-frequency spectrum diagram, calculate the corresponding accidental factor interference index based on the regularity characteristics of the fluctuation of each harmonic signal, obtain the accidental factor interference harmonic based on the accidental factor interference index of all harmonic signals of each analog power signal, calculate the diagnostic strong interference index of each accidental factor interference harmonic based on the difference characteristics between the frequency of each fundamental signal and the frequency of the corresponding accidental factor interference harmonic, and the signal strength of each accidental factor interference harmonic of each fundamental signal; construct each interference diagnosis dimensionality reduction signal of each power time-frequency spectrum diagram based on all accidental factor interference harmonics and the corresponding diagnostic strong interference index, calculate the same type interference index between the corresponding monitoring points based on the similarity characteristics between the interference diagnosis dimensionality reduction signals of the monitoring points, and obtain each power grid monitoring accidental factor interference cluster based on the same type interference index between all monitoring points; A fault diagnosis module is used to calculate the fault diagnosis confidence factor of each monitoring point based on the accidental factor interference index and the diagnostic strong interference index of all accidental factor interference harmonics of each monitoring point, obtain the fault diagnosis reference point of each power grid monitoring accidental factor interference cluster based on the fault diagnosis confidence factor of each monitoring point and the same type interference index between each monitoring point and the remaining monitoring points; calculate the fault diagnosis verification factor of each monitoring point in the current monitoring time period based on the difference characteristics of the fault diagnosis confidence factor between each monitoring point and the corresponding fault diagnosis reference point of the power grid monitoring accidental factor interference cluster in the current monitoring time period, perform pre-diagnosis processing on each monitoring point based on the fault diagnosis verification factor of each monitoring point in the current monitoring time period and all analog power signals of each monitoring point, and obtain the fault diagnosis pre-processing result of 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 execute the steps of the method according to any one of claims 1 to 12.

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

Images