Multi-source hybrid partial discharge separation method, system and equipment for power equipment and medium

Abstract

The invention discloses a power equipment multi-source hybrid partial discharge separation method, system and equipment and a medium, and belongs to the technical field of power equipment partial discharge detection. The method comprises the following steps: extracting a characteristic parameter sequence of a multi-source hybrid partial discharge ultrahigh frequency single pulse signal; forming a characteristic parameter matrix by the plurality of characteristic parameter sequences, extracting key characteristic parameters, and obtaining a key characteristic parameter matrix of a plurality of multi-source mixed partial discharge ultrahigh frequency single pulse signals; quantifying the clustering degree of the key characteristic parameter matrix into a height score; forming a three-dimensional characteristic parameter matrix of the multi-source mixed partial discharge ultrahigh frequency single pulse signal by using the height fraction and the key characteristic parameter matrix; and performing clustering analysis on the three-dimensional characteristic parameter matrixes of the plurality of multi-source mixed partial discharge ultrahigh frequency single pulse signals to obtain a separation result. The technical problem that the difficulty of analysis and diagnosis of partial discharge signals is increased due to coexistence of multiple defects of power equipment is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of partial discharge detection of electric power equipment, and in particular relates to a method, system, equipment and medium for separating multi-source mixed partial discharge of electric power equipment. Background Art

[0002] Partial discharge (PD) refers to the non-penetrating discharge that occurs in the insulation of the equipment. It is the main manifestation of insulation failure in the early stage. It is not only the main cause of insulation degradation, but also a characteristic quantity that characterizes the insulation condition. By conducting offline or online partial discharge detection, the insulation defects inside the transformer can be found to a certain extent, so as to further accurately diagnose the insulation condition of the equipment. In different insulation systems, partial discharge has different discharge types, including tip discharge, suspended discharge, air gap discharge, surface discharge, oil discharge, internal discharge, etc.

[0003] When there are defects in multiple locations in power equipment such as transformers and circuit breakers, the partial discharge sensor will receive signals from non-homologous sources. At this time, the partial discharge waveform and partial discharge pulse phase distribution (PRPD) spectrum will show the mixing and overlapping of multi-source signals. If the partial discharge detection and diagnosis theory based on a single defect is still used, the number, type and location of the defects cannot be accurately determined. Therefore, it is necessary to propose a signal separation method to separate multi-source mixed partial discharge signals. Summary of the invention

[0004] The object of the present invention is to provide a method, system, device and medium for separating multi-source mixed partial discharge of power equipment, so as to separate the mixed partial discharge signal of transformer and solve the technical problem that the coexistence of multiple defects of power equipment increases the difficulty of analyzing and diagnosing partial discharge signals.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] In a first aspect, the present invention provides a method for separating multi-source mixed partial discharge of electric power equipment, comprising:

[0007] Extract characteristic parameter sequence of multi-source mixed partial discharge ultra-high frequency single pulse signal;

[0008] Combining characteristic parameter sequences of several multi-source mixed partial discharge ultra-high frequency single pulse signals into a characteristic parameter matrix of a multi-source mixed partial discharge ultra-high frequency single pulse signal, extracting key characteristic parameters of the product of the characteristic parameter matrix and its transposed matrix, and obtaining key characteristic parameter matrices of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals based on the key characteristic parameters;

[0009] Quantifying the clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals into a height score; composing the three-dimensional characteristic parameter matrix of the multi-source mixed partial discharge ultra-high frequency single pulse signal with the height score and the key characteristic parameter matrix;

[0010] The three-dimensional characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals is clustered and analyzed to obtain the separation results.

[0011] As a further improvement of the present invention, a characteristic parameter sequence of a multi-source mixed partial discharge ultra-high frequency single pulse signal is extracted, including:

[0012] Extract the amplitude of the ultra-high frequency single pulse signal of multi-source mixed partial discharge to form a finite-length one-dimensional time series;

[0013] Perform windowing on the one-dimensional time series to form several vectors;

[0014] All vectors are combined into a trajectory matrix;

[0015] The eigenvalues ​​of the trajectory matrix are obtained, and the eigenvalues ​​greater than zero are used as the characteristic parameter sequence of the multi-source mixed partial discharge ultra-high frequency single pulse signal.

[0016] As a further improvement of the present invention, the key characteristic parameter of extracting the product of the characteristic parameter matrix and its transposed matrix, and obtaining the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals based on the key characteristic parameter comprises:

[0017] Normalizing the characteristic parameter matrix, and calculating the covariance matrix according to the normalized characteristic parameter matrix;

[0018] Calculate the eigenvalues ​​and eigenvectors of the covariance matrix;

[0019] Sort the eigenvalues ​​of the covariance matrix from large to small, select the two largest eigenvalues, and form an eigenvector matrix with the eigenvectors corresponding to these two eigenvalues;

[0020] Based on the eigenvector matrix, the characteristic parameter matrices of a plurality of multi-source mixed partial discharge ultra-high frequency single pulse signals are converted into key characteristic parameter matrices of a plurality of multi-source mixed partial discharge ultra-high frequency single pulse signals.

[0021] As a further improvement of the present invention, a solitary forest algorithm is used to quantify the clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals into a height score.

[0022] As a further improvement of the present invention, the clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals is quantified as a height score, including:

[0023] Selecting sub-sample sets and sub-sample eigenvalues ​​of key characteristic parameter matrices of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals for training, and obtaining the solitary tree parameters;

[0024] Establish a solitary tree based on the solitary tree parameters and test the solitary tree with new samples:

[0025] If the test passes, the height score is calculated based on this set of lone trees; otherwise, the training parameters are adjusted or the samples are updated to retrain until the test passes, and the height score is calculated based on this set of lone trees.

[0026] As a further improvement of the present invention, a cluster analysis is performed on the three-dimensional characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals, and the separation results obtained include:

[0027] Each multi-source mixed partial discharge ultra-high frequency single pulse signal is regarded as a data point, and the characteristic parameter of the multi-source mixed partial discharge ultra-high frequency single pulse signal is regarded as the coordinate of the data point;

[0028] Select the initial cluster centroid and divide all data points into the nearest category according to their distance to the cluster centroid, forming several different initial clusters;

[0029] Modify the value of the cluster centroid to the average value of all data points in this class, recalculate the distance between each data point and the cluster centroid, and classify the data point into the nearest cluster centroid according to the distance, repeat the iterative classification and modification of the cluster centroid until the convergence condition is met;

[0030] According to the clustering results, the number of cluster centroids is adjusted to complete the classification and obtain the multi-source mixed partial discharge separation results.

[0031] As a further improvement of the present invention, the optimization function of cluster analysis is:

[0032] In the formula, C represents the class C i The set of; P is the value of the optimization function, that is, the optimization objective function; the centroid of each cluster is o i ;p ij For data point a j Belongs to Class C i (1≤j≤V); s is the number of partial discharge data points; V is the number of artificially set classes; ω is a weight parameter that can control the nearest class center.

[0033] In a second aspect, the present invention provides a multi-source hybrid partial discharge separation system for electric power equipment, comprising:

[0034] Key characteristic parameter extraction module, used to extract characteristic parameter sequence of multi-source mixed partial discharge ultra-high frequency single pulse signal;

[0035] A key characteristic parameter matrix construction module is used to form a characteristic parameter matrix of a multi-source mixed partial discharge ultra-high frequency single pulse signal from a plurality of characteristic parameter sequences of the multi-source mixed partial discharge ultra-high frequency single pulse signals, extract key characteristic parameters of the product of the characteristic parameter matrix and its transposed matrix, and obtain key characteristic parameter matrices of a plurality of multi-source mixed partial discharge ultra-high frequency single pulse signals based on the key characteristic parameters;

[0036] A three-dimensional characteristic parameter matrix construction module is used to quantify the clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals into a height score; the height score and the key characteristic parameter matrix are combined to form a three-dimensional characteristic parameter matrix of the multi-source mixed partial discharge ultra-high frequency single pulse signal;

[0037] The separation module is used to cluster the three-dimensional characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals to obtain separation results.

[0038] In a third aspect, the present invention provides an electronic device, comprising:

[0039] at least one processor; and,

[0040] a memory communicatively connected to the at least one processor; wherein,

[0041] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform steps to implement the above-mentioned method for separating multi-source hybrid partial discharge of electric power equipment.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the multi-source hybrid partial discharge separation method for power equipment as described above.

[0043] In a fifth aspect, the present invention provides a computer program product, which stores a multi-source hybrid local discharge separation program for electric power equipment, and when the multi-source hybrid local discharge separation program for electric power equipment is executed by a processor, the steps of the multi-source hybrid local discharge separation method for electric power equipment as described above are implemented.

[0044] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0045] The present invention provides a method for separating multi-source mixed partial discharge signals of electric power equipment. After analyzing and extracting characteristic values, high-dimensional features are subjected to dimensionality reduction processing. Considering that two-dimensional plane parameters cannot intuitively reflect the difference in separation results in some cases, outlier detection is introduced to quantify the clustering degree of multi-source partial discharge data in the feature space as a height score. One-dimensional features are added on the basis of the original features. Based on the multi-source partial discharge mixed signal separation method proposed by the present invention, separation and extraction of multi-source partial discharge signals are realized, and the accuracy of subsequent partial discharge identification and positioning work is improved.

[0046] Furthermore, the solitary forest algorithm is used to quantify the clustering degree of multi-source partial discharge data in the feature space as a height score. As an unsupervised learning method, the solitary forest algorithm does not need to be trained with labeled data. It directly classifies multi-source partial discharges according to the waveform characteristics of the obtained partial discharge signal itself. It can quantify the clustering tightness between different cluster signals in the feature space of multi-source partial discharge signals, and can effectively remove noise or separate multi-source signals with large differences in the number of discharge points, thereby improving the accuracy of subsequent partial discharge diagnosis and positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flow chart of the method for separating multi-source mixed partial discharge of electric power equipment according to the present invention;

[0048] Figure 2 This is a single partial discharge ultra-high frequency pulse waveform diagram;

[0049] Figure 3 This is the separation result diagram of the separation algorithm for two mixed discharges;

[0050] Figure 4 This is the separation result diagram of the separation algorithm for three mixed discharges;

[0051] Figure 5 A structural block diagram of a multi-source hybrid partial discharge separation system for electric power equipment provided by an embodiment of the present invention;

[0052] Figure 6 A block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0054] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0055] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0056] Reference Figure 1 In one embodiment of the present invention, a method for separating multi-source mixed partial discharge of power equipment is provided, comprising the following steps:

[0057] Step 1: Acquire multiple multi-source mixed partial discharge ultra-high frequency single pulse signals, the number of which is denoted by s.

[0058] Step 2: extract the characteristic parameter sequences W of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals, and form the characteristic parameter sequences of these multi-source mixed partial discharge ultra-high frequency single pulse signals into a characteristic parameter matrix Q of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals, wherein the characteristic parameter matrix Q of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals = [W1, W2, W3, …, W m ], use the principal component analysis method to extract the characteristic parameter matrix and the characteristic parameter matrix transpose multiplication, that is, QQ T Based on the key characteristic parameters, a key characteristic parameter matrix R of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals is obtained.

[0059] In some embodiments, methods such as linear discriminant analysis and t-distribution neighborhood embedding algorithm can also be used to convert high-dimensional features into low-dimensional features, wherein the present invention extracts matrix-matrix transposition multiplication, i.e., QQ T The key feature parameters are used to reduce the dimensionality of high-dimensional features.

[0060] Furthermore, in the specific implementation process, step 2 specifically includes:

[0061] Step 21, obtaining a plurality of multi-source mixed partial discharge ultra-high frequency single pulse signals, the number of which is denoted as s. Performing singular spectrum analysis on the s multi-source mixed partial discharge ultra-high frequency single pulse signals, obtaining characteristic parameter sequences of the s multi-source mixed partial discharge ultra-high frequency single pulse signals;

[0062] The following takes a single multi-source mixed partial discharge ultra-high frequency pulse signal as an example to illustrate the process of performing singular spectrum analysis on the multi-source mixed partial discharge ultra-high frequency single pulse signal to obtain the characteristic parameter sequence of the multi-source mixed partial discharge ultra-high frequency single pulse signal: Use signal acquisition equipment to obtain the multi-source mixed partial discharge ultra-high frequency single pulse signal. The waveform of the multi-source mixed partial discharge ultra-high frequency single pulse signal is as follows: Figure 2 The figure shows a waveform diagram drawn from a set of two-dimensional data, including time and amplitude. Singular spectrum analysis (SSA) is performed on each multi-source mixed partial discharge ultra-high frequency single pulse signal to obtain the characteristic parameter sequence of each multi-source mixed partial discharge ultra-high frequency single pulse signal.

[0063] Furthermore, in a specific implementation process, performing singular spectrum analysis on a single multi-source mixed partial discharge ultra-high frequency single pulse signal specifically includes:

[0064] Step 211: extract the ultra-high frequency single pulse signal of multi-source mixed partial discharge, and the amplitude data constitutes a finite-length one-dimensional time series Y, Y = [y1, y2, y3, ..., y n ], the sequence length is recorded as n. i=1,2,3…n, encoding is performed according to the time of signal acquisition, y i It represents the amplitude of the ith collected single pulse signal of source mixed partial discharge ultra-high frequency.

[0065] Step 212: Take the window length as l, and k=n-l+1, and perform windowing on the one-dimensional time series Y extracted from the multi-source mixed partial discharge ultra-high frequency single pulse signal in step 211 to form k vectors X with a length of l. a , a=1,2,3…k, each vector is recorded as:

[0066] X a =[y a ,y a+1 ,y a+2 ,L,y a+l-1 ] T

[0067] Step 213: The k vectors X in step 212 i The trajectory matrix Z is composed as:

[0068]

[0069] Step 214, obtain the eigenvalue of the trajectory matrix: perform matrix-matrix transposition multiplication on the trajectory matrix Z obtained in 13 for the multi-source mixed partial discharge ultra-high frequency single pulse signal to obtain ZZ T , for ZZ T Perform eigenvalue decomposition to obtain l eigenvalues, m of which are greater than 0, and the eigenvalues ​​λ1≥λ2≥λ3≥L≥λ m ≥0, λ represents the characteristic value; that is, l>m. The m characteristic values ​​greater than 0 are taken as the characteristic parameter sequence W of the multi-source mixed partial discharge ultra-high frequency single pulse signal, denoted as W=[λ1,λ2,λ3,…,λ m ].

[0070] Step 22: Combining the characteristic parameter sequences W of the s multi-source mixed partial discharge ultra-high frequency single pulse signals into a characteristic parameter matrix Q of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals of s×m, Q=[W1, W2, W3, …, W m ];

[0071] Step 23, normalize the characteristic parameter matrix Q to obtain a normalized characteristic parameter matrix Q1; calculate the covariance matrix according to the normalized characteristic parameter matrix Q1

[0072] Step 24: Use eigenvalue decomposition to obtain the covariance matrix The eigenvalues ​​and eigenvectors of , where each eigenvector is a row vector of 1×s;

[0073] Step 25: Covariance matrix Sort the eigenvalues ​​from large to small, select the two largest eigenvalues, and form the eigenvector matrix P of 2×s with the eigenvectors corresponding to these two eigenvalues;

[0074] Step 26, converting the characteristic parameter matrix Q of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals into the key characteristic parameter matrix R of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals, where R=PQ, and R is a two-dimensional matrix.

[0075] Step 3: Use the solitary forest algorithm to quantify the clustering degree of the key characteristic parameter matrix R of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals into a height score. The height score is also used as the characteristic parameter of the multi-source mixed partial discharge ultra-high frequency single pulse signal, and finally a three-dimensional characteristic parameter matrix B of the multi-source mixed partial discharge ultra-high frequency single pulse signal is formed.

[0076] Furthermore, in the specific implementation process, step 3 specifically includes:

[0077] Step 31, randomly selecting a sub-sample set and a sub-sample eigenvalue of a key characteristic parameter matrix R of a plurality of multi-source mixed partial discharge ultra-high frequency single pulse signals, and performing training to obtain a solitary tree parameter;

[0078] Specifically, in this step, the selected multiple multi-source mixed partial discharge ultra-high frequency single pulse signals are historical data.

[0079] Step 32, establish a lone tree according to the lone tree parameters, and test the lone tree with new samples. If the test passes, it means that this group of lone trees is reasonably constructed, and the height score can be calculated based on this group of lone trees. If the test fails, adjust the training parameters or update the samples to retrain until the test passes.

[0080] Step 33, based on the established unique tree, calculate the height scores of the key characteristic parameter matrix R of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals, and use the height scores as the characteristic parameters of the multi-source mixed partial discharge ultra-high frequency single pulse signals. The height scores and the key characteristic parameter matrix R finally constitute the three-dimensional characteristic parameter matrix B of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals.

[0081] Step 4: cluster analysis is performed on the three-dimensional characteristic parameter matrix B of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals to obtain separation results.

[0082] Furthermore, in the specific implementation process, step 4 specifically includes:

[0083] Step 41, the three-dimensional characteristic parameter matrix B of s multi-source mixed partial discharge ultra-high frequency single pulse signals is a three-dimensional characteristic parameter matrix, wherein each multi-source mixed partial discharge ultra-high frequency single pulse signal has three characteristic parameters, and each multi-source mixed partial discharge ultra-high frequency single pulse signal is regarded as a data point a j , j = 1, 2, 3…s, the characteristic parameters of the multi-source mixed partial discharge ultra-high frequency single pulse signal are regarded as the coordinates of the data point. The present invention uses the fuzzy C-means (FCM) clustering algorithm for clustering analysis, selects the initial cluster centroid, and divides all data points into the nearest category according to their distance to the cluster centroid, forming V different initial clusters;

[0084] Step 42: Modify the value of the cluster centroid to the average value of all data points in this class, and then recalculate the minimum distance between all data points and each cluster centroid. According to the minimum distance between the data point and the cluster centroid, classify the data point into the nearest cluster centroid. Repeat the process of iterative classification and cluster centroid modification until the convergence condition of the FCM clustering algorithm is met. The optimization function of the FCM clustering algorithm can be defined as:

[0085]

[0086] In the formula, C represents the class C b The set of; P is the value of the optimization function, that is, the optimization objective function; the centroid of each cluster is o b ;p bj For data point a j Belongs to Class C b (1≤b≤V); s is the number of partial discharge data points; V is the number of artificially set classes; ω is a weight parameter that can control the nearest class center.

[0087] Among them, the membership function can be defined as:

[0088]

[0089] Among them, l is the number in a cycle, o l is the centroid of the cluster selected in the loop.

[0090] The cluster centroid can be defined as:

[0091]

[0092] Step 43: According to the clustering result, the number of cluster centroids is adjusted to complete the classification and obtain the multi-source mixed partial discharge separation result.

[0093] The above method is used to separate the collected partial discharge signals, and the final clustering results are as follows: Figure 3 and Figure 4 As shown, Figure 3 The separation algorithm is used to separate the surface discharge and suspension mixed discharge collected by the UHF sensor. The separation accuracy is 98.8%. Figure 4 The separation algorithm is used to separate the mixed signals of local discharge from the tip, surface and suspension sources collected by the high-frequency current sensor, and the separation accuracy is 93.4%.

[0094] See also Figure 5 In yet another embodiment of the present invention, a multi-source hybrid partial discharge separation system for power equipment is provided, comprising:

[0095] Key characteristic parameter extraction module, used to extract characteristic parameter sequence of multi-source mixed partial discharge ultra-high frequency single pulse signal;

[0096] The key characteristic parameter matrix construction module combines the characteristic parameter sequences of the multi-source mixed partial discharge ultra-high frequency single pulse signals into a characteristic parameter matrix of the multi-source mixed partial discharge ultra-high frequency single pulse signal, and extracts the product QQ of the characteristic parameter matrix and its transposed matrix. TBased on the key characteristic parameters, a key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals is obtained;

[0097] A three-dimensional characteristic parameter matrix construction module is used to quantify the clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals into a height score; the height score and the key characteristic parameter matrix are combined to form a three-dimensional characteristic parameter matrix of the multi-source mixed partial discharge ultra-high frequency single pulse signal;

[0098] The separation module is used to cluster the three-dimensional characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals to obtain separation results.

[0099] All relevant contents of each step involved in the aforementioned embodiment of a multi-source hybrid partial discharge separation method for electric power equipment can be referred to the functional description of the functional modules corresponding to a multi-source hybrid partial discharge separation system for electric power equipment in an embodiment of the present invention, and will not be repeated here.

[0100] Reference Figure 6 In another embodiment of the present invention, an electronic device is provided, the electronic device includes a processor and a memory, the processor and the memory are connected through a bus; the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of a multi-source hybrid partial discharge separation method for power equipment. The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6The fact that only one line is used in the diagram does not mean that there is only one bus or only one type of bus.

[0101] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the multi-source hybrid partial discharge separation method of a power equipment in the above embodiment.

[0102] In another embodiment of the present invention, a computer program product is provided, wherein the computer program product stores a multi-source hybrid partial discharge separation program for electric power equipment, and when the multi-source hybrid partial discharge separation program for electric power equipment is executed by a processor, the steps of the multi-source hybrid partial discharge separation method for electric power equipment as described above are implemented. Therefore, no further description will be given here. In addition, no further description will be given of the beneficial effects of adopting the same method. For technical details not disclosed in the computer program product embodiment involved in the present invention, please refer to the description of the method embodiment of the present invention.

[0103] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0104] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0105] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0107] 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 relevant 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 within the scope of protection of the claims of the present invention.

Claims

1. A method for separating multi-source mixed partial discharge of power equipment, characterized in that: include: Extract characteristic parameter sequence of multi-source mixed partial discharge ultra-high frequency single pulse signal; Combining characteristic parameter sequences of several multi-source mixed partial discharge ultra-high frequency single pulse signals into a characteristic parameter matrix of a multi-source mixed partial discharge ultra-high frequency single pulse signal, extracting key characteristic parameters of the product of the characteristic parameter matrix and its transposed matrix, and obtaining key characteristic parameter matrices of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals based on the key characteristic parameters; Quantifying the clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals into a height score; composing the three-dimensional characteristic parameter matrix of the multi-source mixed partial discharge ultra-high frequency single pulse signal with the height score and the key characteristic parameter matrix; The three-dimensional characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals is clustered and analyzed to obtain the separation results.

2. A method for separating multi-source mixed partial discharge of electric power equipment according to claim 1, characterized in that: The method for extracting a characteristic parameter sequence of a multi-source mixed partial discharge ultra-high frequency single pulse signal comprises: Extract the amplitude of the ultra-high frequency single pulse signal of multi-source mixed partial discharge to form a finite-length one-dimensional time series; Perform windowing on the one-dimensional time series to form several vectors; All vectors are combined into a trajectory matrix; The eigenvalues ​​of the trajectory matrix are obtained, and the eigenvalues ​​greater than zero are used as the characteristic parameter sequence of the multi-source mixed partial discharge ultra-high frequency single pulse signal.

3. A method for separating multi-source mixed partial discharge of electric power equipment according to claim 1, characterized in that: The key characteristic parameter of extracting the product of the characteristic parameter matrix and its transposed matrix, and obtaining the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals based on the key characteristic parameter comprises: Normalizing the characteristic parameter matrix, and calculating the covariance matrix according to the normalized characteristic parameter matrix; Calculate the eigenvalues ​​and eigenvectors of the covariance matrix; Sort the eigenvalues ​​of the covariance matrix from large to small, select the two largest eigenvalues, and form an eigenvector matrix with the eigenvectors corresponding to these two eigenvalues; Based on the eigenvector matrix, the characteristic parameter matrices of a plurality of multi-source mixed partial discharge ultra-high frequency single pulse signals are converted into key characteristic parameter matrices of a plurality of multi-source mixed partial discharge ultra-high frequency single pulse signals.

4. A method for separating multi-source mixed partial discharge of electric power equipment according to claim 1, characterized in that: The clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals is quantified as a height score using the solitary forest algorithm.

5. A method for separating multi-source mixed partial discharge of electric power equipment according to claim 1, characterized in that: The clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals is quantified as a height score, including: Selecting sub-sample sets and sub-sample eigenvalues ​​of key characteristic parameter matrices of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals for training, and obtaining the solitary tree parameters; Establish a solitary tree based on the solitary tree parameters and test the solitary tree with new samples: If the test passes, the height score is calculated based on this set of lone trees; otherwise, the training parameters are adjusted or the samples are updated to retrain until the test passes, and the height score is calculated based on this set of lone trees.

6. A method for separating multi-source mixed partial discharge of electric power equipment according to claim 1, characterized in that: The three-dimensional characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals is clustered and analyzed to obtain separation results including: Each multi-source mixed partial discharge ultra-high frequency single pulse signal is regarded as a data point, and the characteristic parameter of the multi-source mixed partial discharge ultra-high frequency single pulse signal is regarded as the coordinate of the data point; Select the initial cluster centroid and divide all data points into the nearest category according to their distance to the cluster centroid, forming several different initial clusters; Modify the value of the cluster centroid to the average value of all data points in this class, recalculate the distance between each data point and the cluster centroid, and classify the data point into the nearest cluster centroid according to the distance, repeat the iterative classification and modification of the cluster centroid until the convergence condition is met; According to the clustering results, the number of cluster centroids is adjusted to complete the classification and obtain the multi-source mixed partial discharge separation results.

7. A method for separating multi-source mixed partial discharge of electric power equipment according to claim 6, characterized in that: The optimization function of the cluster analysis is: In the formula, C represents the class C i The set of; P is the value of the optimization function, that is, the optimization objective function; the centroid of each cluster is o i ;p ij For data point a j Belongs to Class C i (1≤j≤V); s is the number of partial discharge data points; V is the number of artificially set classes; ω is a weight parameter that can control the nearest class center.

8. A multi-source hybrid partial discharge separation system for power equipment, characterized in that: include: Key characteristic parameter extraction module, used to extract characteristic parameter sequence of multi-source mixed partial discharge ultra-high frequency single pulse signal; A key characteristic parameter matrix construction module is used to form a characteristic parameter matrix of a multi-source mixed partial discharge ultra-high frequency single pulse signal from a plurality of characteristic parameter sequences of the multi-source mixed partial discharge ultra-high frequency single pulse signals, extract key characteristic parameters of the product of the characteristic parameter matrix and its transposed matrix, and obtain key characteristic parameter matrices of a plurality of multi-source mixed partial discharge ultra-high frequency single pulse signals based on the key characteristic parameters; A three-dimensional characteristic parameter matrix construction module is used to quantify the clustering degree of the key characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals into a height score; the height score and the key characteristic parameter matrix are combined to form a three-dimensional characteristic parameter matrix of the multi-source mixed partial discharge ultra-high frequency single pulse signal; The separation module is used to cluster the three-dimensional characteristic parameter matrix of multiple multi-source mixed partial discharge ultra-high frequency single pulse signals to obtain separation results.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a multi-source hybrid partial discharge separation method for electric power equipment according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a multi-source hybrid partial discharge separation method for electric power equipment according to any one of claims 1 to 7 is implemented.

Images