Method, device, equipment and storage medium for determining partial discharge type of GIS equipment

Abstract

The present invention discloses a method, device, equipment and storage medium for determining the local discharge type of GIS equipment, including: obtaining local discharge data; converting the local discharge data into a pulse sequence phase distribution PRPS image; extracting image features of the PRPS image using a complete local binary pattern method; obtaining a range interval corresponding to a parameter pair of an SVM recognition model; generating multiple parameter pairs within the range interval to form a parameter pair set; performing a search operation on the parameter pair set until the fitness of each parameter pair is obtained; determining a target SVM recognition model according to the parameter pair with the largest fitness; inputting the image features of the PRPS image into a target SVM recognition model to obtain the local discharge type of the GIS equipment corresponding to the local discharge data; the present invention realizes accurate identification of the local discharge type, effectively reduces the misjudgment rate of faults, reduces the influence of sample differences of different fault types on model performance in actual engineering, and improves the fault recognition accuracy.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of power equipment status monitoring, and in particular to a method, device, equipment and storage medium for determining the partial discharge type of GIS equipment. Background Art

[0002] Gas Insulated Switchgear (GIS) is the main equipment in substations because of its good insulation performance, high reliability and small size. However, due to the complex working environment, insulation aging often causes partial discharge, which in turn leads to insulation breakdown. Partial discharge is an important sign of insulation aging of power equipment and a precursor of insulation failure. It has important engineering value for GIS equipment insulation status monitoring.

[0003] Since the apparent discharge amount is different when different types of partial discharge faults occur in GIS equipment, the damage to the insulation is also different. Therefore, accurate identification of the type of partial discharge fault is helpful to judge the severity of the fault and accurately judge the health status of the equipment. At present, the detection of partial discharge type of power equipment mainly relies on the signal characteristics of partial discharge signals. However, due to the presence of background noise in the partial discharge signal acquisition process and the difficulty in removing it, it will affect the identification of the fault type and reduce the accuracy of fault identification. Summary of the invention

[0004] The embodiments of the present invention provide a method, device, equipment and storage medium for determining the local discharge type of GIS equipment, so as to realize accurate identification of the local discharge type, effectively reduce the misjudgment rate of faults, reduce the impact of the differences in samples of different fault types in actual projects on the model performance, improve the fault identification accuracy, and enhance the health status detection level of GIS equipment.

[0005] In a first aspect, an embodiment of the present invention provides a method for determining a partial discharge type of GIS equipment, comprising:

[0006] Obtain partial discharge data;

[0007] Convert the partial discharge data into a pulse sequence phase distribution PRPS (Phase Resolved Pulse Sequence) image;

[0008] The Completed Local Binary Patterns (CLBP) method is used to extract the image features of PRPS images;

[0009] Get the range interval corresponding to the parameter pair of the SVM recognition model;

[0010] Generate multiple parameter pairs within the range interval to form a parameter pair set;

[0011] Perform a search operation on the parameter pair set until the fitness of each parameter pair is obtained;

[0012] Determine the target SVM recognition model based on the parameter pair with the largest fitness;

[0013] Inputting the image features of the PRPS image into a target SVM recognition model to obtain a GIS equipment local discharge type corresponding to the local discharge data;

[0014] The search operation includes:

[0015] Searching the parameter pair set to obtain parameter pairs to be selected;

[0016] The fitness of the candidate parameter pair is determined based on the fitness function.

[0017] Optionally, the fitness of the candidate parameter pair is determined based on a fitness function, including:

[0018] Acquire an image sample set, wherein the image sample set includes: PRPS image samples and GIS equipment local discharge types corresponding to the PRPS image samples;

[0019] Generate a first SVM recognition model based on the candidate parameter pair;

[0020] Inputting the PRPS image samples in the image sample set into a first SVM recognition model to obtain a predicted GIS equipment partial discharge type;

[0021] According to the GIS equipment partial discharge type corresponding to the PRPS image samples in the image sample set and the predicted GIS equipment partial discharge type corresponding to the PRPS image samples, the number of misjudged samples and the number of correctly identified samples corresponding to each GIS equipment partial discharge type are determined;

[0022] Determine the number of PRPS image samples, the number of partial discharge types, and the number of PRPS image samples corresponding to each partial discharge type according to the image sample set;

[0023] The fitness of the candidate parameter pairs is determined according to the number of PRPS image samples, the number of partial discharge types, the number of PRPS image samples corresponding to each partial discharge type, the number of misjudged samples corresponding to each GIS equipment partial discharge type, the number of correctly identified samples corresponding to each GIS equipment partial discharge type, and the fitness function.

[0024] Optionally, the fitness function is:

[0025]

[0026] Among them, fitness is the fitness of the candidate parameter pair, M is the number of PRPS image samples, and M j is the number of PRPS image samples corresponding to the jth GIS equipment partial discharge type, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to the j types of GIS equipment partial discharge.

[0027] Optionally, the fitness of the candidate parameter pair is determined according to the number of PRPS image samples, the number of partial discharge types, the number of PRPS image samples corresponding to each partial discharge type, the number of misjudged samples corresponding to each GIS device partial discharge type, the number of correctly identified samples corresponding to each GIS device partial discharge type, and the fitness function, including:

[0028] Obtain the weight of the number of PRPS image samples, the weight of the number of partial discharge types, the weight of the number of PRPS image samples corresponding to each partial discharge type, the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, and the weight of the number of correctly identified samples corresponding to each GIS equipment partial discharge type;

[0029] Determine the fitness of the candidate parameter pair according to the weight of the number of PRPS image samples, the weight of the number of local discharge types, the weight of the number of PRPS image samples corresponding to each local discharge type, the weight of the number of misjudged samples corresponding to each GIS device local discharge type, the weight of the number of correctly identified samples corresponding to each GIS device local discharge type, the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, the number of correctly identified samples corresponding to each GIS device local discharge type, and the fitness function;

[0030] The fitness function is:

[0031]

[0032] Where M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth type of GIS equipment partial discharge, K1 is the weight of the number of PRPS image samples corresponding to each type of partial discharge, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, K2 is the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, TP jis the number of correctly identified samples corresponding to j types of GIS equipment partial discharge, and K3 is the weight of the number of correctly identified samples corresponding to each type of GIS equipment partial discharge.

[0033] Optionally, multiple parameter pairs are generated within the range interval to form a parameter pair set, including:

[0034] Generate parameter pairs according to the upper and lower bounds of the range interval corresponding to the parameter pairs, the target weight, and the target formula;

[0035] A parameter pair set is determined based on the generated parameter pairs.

[0036] Optionally, the target formula is:

[0037] X i+1 =X min +λ i+1 (X max -X min );

[0038]

[0039] Among them, X i+1 is the i+1th parameter pair, X max is the upper bound of the range interval corresponding to the parameter pair, X min is the lower bound of the range interval corresponding to the parameter pair, λ i+1 is the target weight corresponding to the i+1th parameter pair, λ i is the target weight corresponding to the ith parameter pair. Optionally, the PRPS image samples include at least one of: a PRPS image sample corresponding to a tip discharge fault type, a PRPS image sample corresponding to a suspended discharge fault type, a PRPS image sample corresponding to a surface discharge fault type, a PRPS image sample corresponding to an internal discharge fault type, and a PRPS image sample corresponding to a particle discharge fault type.

[0040] In a second aspect, an embodiment of the present invention provides a device for determining a partial discharge type of a GIS device, comprising:

[0041] A first acquisition module, used for acquiring partial discharge data;

[0042] A conversion module, used for converting partial discharge data into a pulse sequence phase distribution PRPS image;

[0043] An extraction module, for extracting image features of the PRPS image using a complete local binary pattern method;

[0044] The second acquisition module is used to obtain the range interval corresponding to the parameter pair of the SVM recognition model;

[0045] A generating module, used for generating a plurality of parameter pairs within a range interval to form a parameter pair set;

[0046] A search module, used to perform a search operation on the parameter pair set until the fitness of each parameter pair is obtained;

[0047] A first determination module is used to determine a target SVM recognition model according to a parameter pair with the largest fitness;

[0048] The second determination module is used to input the image features of the PRPS image into a target SVM recognition model to obtain the GIS equipment local discharge type corresponding to the local discharge data;

[0049] Wherein, the search module includes:

[0050] A searching unit, configured to search the parameter pair set to obtain a candidate parameter pair;

[0051] The determination unit is used to determine the fitness of the candidate parameter pair based on the fitness function.

[0052] Optionally, determine the unit, specifically for:

[0053] Acquire an image sample set, wherein the image sample set includes: PRPS image samples and GIS equipment local discharge types corresponding to the PRPS image samples;

[0054] Generate a first SVM recognition model based on the candidate parameter pair;

[0055] Inputting the PRPS image samples in the image sample set into a first SVM recognition model to obtain a predicted GIS equipment partial discharge type;

[0056] According to the GIS equipment partial discharge type corresponding to the PRPS image samples in the image sample set and the predicted GIS equipment partial discharge type corresponding to the PRPS image samples, the number of misjudged samples and the number of correctly identified samples corresponding to each GIS equipment partial discharge type are determined;

[0057] Determine the number of PRPS image samples, the number of partial discharge types, and the number of PRPS image samples corresponding to each partial discharge type according to the image sample set;

[0058] The fitness of the candidate parameter pairs is determined according to the number of PRPS image samples, the number of partial discharge types, the number of PRPS image samples corresponding to each partial discharge type, the number of misjudged samples corresponding to each GIS equipment partial discharge type, the number of correctly identified samples corresponding to each GIS equipment partial discharge type, and the fitness function.

[0059] Optionally, the fitness function is:

[0060]

[0061] Where M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth GIS equipment partial discharge type, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to the j types of GIS equipment partial discharge.

[0062] Optionally, the determination unit is further used to:

[0063] Obtain the weight of the number of PRPS image samples, the weight of the number of partial discharge types, the weight of the number of PRPS image samples corresponding to each partial discharge type, the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, and the weight of the number of correctly identified samples corresponding to each GIS equipment partial discharge type;

[0064] Determine the fitness of the candidate parameter pair according to the weight of the number of PRPS image samples, the weight of the number of local discharge types, the weight of the number of PRPS image samples corresponding to each local discharge type, the weight of the number of misjudged samples corresponding to each GIS device local discharge type, the weight of the number of correctly identified samples corresponding to each GIS device local discharge type, the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, the number of correctly identified samples corresponding to each GIS device local discharge type, and the fitness function;

[0065] The fitness function is:

[0066]

[0067] Where M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth type of GIS equipment partial discharge, K1 is the weight of the number of PRPS image samples corresponding to each type of partial discharge, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, K2 is the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to j types of GIS equipment partial discharge, and K3 is the weight of the number of correctly identified samples corresponding to each type of GIS equipment partial discharge.

[0068] Optionally, the generation module is specifically used to:

[0069] Generate parameter pairs according to the upper and lower bounds of the range interval corresponding to the parameter pairs, the target weight, and the target formula;

[0070] A parameter pair set is determined based on the generated parameter pairs.

[0071] Optionally, the target formula is:

[0072] X i+1 =X min +λ i+1 (X max -X min );

[0073]

[0074] Among them, X i+1 is the i+1th parameter pair, X max is the upper bound of the range interval corresponding to the parameter pair, X min is the lower bound of the range interval corresponding to the parameter pair, λ i+1 is the target weight corresponding to the i+1th parameter pair, λ i is the target weight corresponding to the i-th parameter pair.

[0075] Optionally, the PRPS image samples include: at least one of: PRPS image samples corresponding to a tip discharge fault type, PRPS image samples corresponding to a suspended discharge fault type, PRPS image samples corresponding to a surface discharge fault type, PRPS image samples corresponding to an internal discharge fault type, and PRPS image samples corresponding to a particle discharge fault type.

[0076] In a third aspect, an embodiment of the present invention provides an electronic device, the electronic device comprising:

[0077] at least one processor; and

[0078] a memory communicatively connected to the at least one processor; wherein,

[0079] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the local discharge type of GIS equipment described in any embodiment of the present invention.

[0080] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining the local discharge type of GIS equipment described in any embodiment of the present invention when executed.

[0081] The embodiment of the present invention obtains local discharge data; converts the local discharge data into a PRPS image; extracts image features of the PRPS image through a CLBP algorithm; and inputs the image features of the PRPS image into an AO-SVM recognition model to obtain the local discharge type of the GIS equipment corresponding to the local discharge data. The embodiment of the present invention solves the problem in the prior art that background noise exists in the local discharge signal acquisition process and is difficult to remove, which affects the identification of faults. The embodiment of the present invention realizes accurate identification of the local discharge type, effectively reduces the misjudgment rate of faults, reduces the influence of sample differences of different fault types on model performance in actual projects, improves the fault identification accuracy, and enhances the health status detection level of GIS equipment.

[0082] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0084] Figure 1 A flowchart of a method for determining the type of partial discharge of GIS equipment provided in Embodiment 1 of the present invention;

[0085] Figure 2 A schematic diagram of a PRPS original image and three features extracted by CLBP provided in Example 1 of the present invention;

[0086] Figure 3 A flowchart of determining the fitness of a candidate parameter pair based on a fitness function provided in Embodiment 2 of the present invention;

[0087] Figure 4 A CLBP-AO-SVM algorithm flow chart provided in Embodiment 2 of the present invention;

[0088] Figure 5 A schematic diagram of the structure of a device for determining the type of partial discharge of GIS equipment provided in Embodiment 3 of the present invention;

[0089] Figure 6 A schematic diagram of the structure of an electronic device that can be used to implement an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0090] 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.

[0091] 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 that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0092] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0093] Embodiment 1

[0094] Figure 1 This is a flow chart of a method for determining the type of partial discharge of GIS equipment provided in the first embodiment of the present invention. This embodiment is applicable to the case of determining the type of partial discharge of GIS equipment. The method can be executed by the device for determining the type of partial discharge of GIS equipment in the embodiment of the present invention. The device can be implemented in software and / or hardware. Figure 1 As shown, the method specifically comprises the following steps:

[0095] S110, obtaining partial discharge data.

[0096] The partial discharge data may refer to partial discharge data collected by a UHF sensor.

[0097] In this embodiment, typical partial discharge models of four defects of GIS equipment, namely metal particles, spikes, air gaps, and suspended electrodes, can be pre-constructed; high voltage is passed through the four defect partial discharge models, and partial discharge data is collected using ultra-high frequency sensors.

[0098] S120, converting the partial discharge data into a pulse sequence phase distribution PRPS image.

[0099] The pulse sequence phase distribution image may refer to a pulse signal image of a partial discharge, which displays the phased pulse signal generated by the partial discharge in a three-dimensional coordinate system in chronological order.

[0100] S130, extracting image features of the PRPS image using a complete local binary pattern method.

[0101] Among them, the image features of the pulse sequence phase distribution (PRPS) image can be extracted through the completely local binary pattern (CLBP). Figure 2 A schematic diagram of a PRPS original image and three features extracted by CLBP provided in the first embodiment of the present invention, such as Figure 2 As shown in the figure, CLBP mainly consists of three parts: the CLBP_C operator based on the grayscale of the center point, the CLBP_S operator based on the local sign difference, and the CLBP_M operator based on the local amplitude difference. The expressions of the three operators are:

[0102]

[0103]

[0104]

[0105] Among them, g c Indicates the central pixel of a point, c I represents the average gray value of the image pixels, c is the adaptive threshold value calculated by m in the entire image p The mean value of s p and m p They are d p The sign and magnitude of d p is a local difference vector between the center pixel and the other P neighborhood values.

[0106] In this embodiment, the image features of the PRPS image are extracted by the CLBP algorithm, which not only has good robustness to noise but also has a fast calculation speed, effectively avoiding the interference of background noise during partial discharge data acquisition and improving the efficiency of image feature extraction.

[0107] S140, obtaining a range interval corresponding to a parameter pair of the SVM recognition model.

[0108] Among them, the AO-SVM recognition model can be understood as using the improved Aquila Optimization (AO) algorithm to search for the optimal values ​​of the c parameter and the g parameter of the support vector machine (SVM), and using the SVM with the c parameter and the g parameter as the optimal values ​​to identify the type of partial discharge. The characteristic of the SVM recognition model is that the parameter pair needs to be determined in advance. Therefore, the parameter pair can be a parameter pair consisting of a c parameter and a g parameter. Specifically, the improved AO algorithm can be an improved AO algorithm based on a chaotic mapping function; the range interval corresponding to the parameter pair of the SVM recognition model can be directly obtained through the parameter pair.

[0109] In this embodiment, the AO-SVM model is used to identify the type of GIS partial discharge based on the image features of the PRPS image extracted by CLBP; since the c and g parameters of the SVM recognition model have a huge impact on the recognition performance, the improved AO optimization can be used to optimize the c and g parameters of the SVM, obtain the parameter pair of the SVM recognition model and the range interval corresponding to the parameter pair in advance, and use the SVM with the optimal c and g parameters to identify the type of GIS equipment partial discharge corresponding to the PRPS image, so as to achieve accurate identification of the type of partial discharge and effectively improve the health status detection level of GIS equipment.

[0110] S150: Generate multiple parameter pairs within the range interval to form a parameter pair set.

[0111] Specifically, a plurality of parameter pairs may be generated within a range corresponding to the parameter pairs of the SVM recognition model obtained in advance to form a parameter pair set, and the parameter pair set includes at least two parameter pairs.

[0112] S160, performing a search operation on the parameter pair set until the fitness of each parameter pair is obtained.

[0113] The search operation can be understood as an optimization operation to determine the optimal c and g parameters. The search operation includes: searching the parameter pair set to obtain multiple candidate parameter pairs, and the multiple candidate parameter pairs can all be used as the optimal c and g parameter pairs.

[0114] Specifically, the method of performing a search operation on the parameter pair set until the fitness of each parameter pair is obtained can be: performing a search operation on the parameter pair set based on the AO algorithm until the fitness of each parameter pair is obtained, or performing a search operation on the parameter pair set based on the improved AO algorithm until the fitness of each parameter pair is obtained.

[0115] Specifically, the optimization process of AO algorithm can be divided into four stages, namely, expanding the exploration scope, narrowing the exploration scope, expanding the development space, and narrowing the development space. The specific steps are as follows:

[0116] 1. Expand the search range and update the current optimal value according to the following formula:

[0117]

[0118] X1(t+1) is the solution of the t+1 iteration under the first search method, X best (t) is the optimal solution obtained after t iterations, N is the total number of current solutions, X M (t) is the average value of all current solutions as shown below, t is the current iteration number, T is the maximum iteration number;

[0119]

[0120] 2. Narrow the exploration scope. In this stage, the search scope of the solution is further narrowed. The update formula of the solution is as follows:

[0121] X2(t+1)=X best (t)×Levy(D)+X R (t)+(yx)*rand

[0122] X2(t+1) is the solution of the t+1 iteration in the second upward search mode, X R (t) is the random solution in the range [1,N] at the tth iteration, D is the dimension of the solution, and Levy(D) is the distribution function of the Levy flight:

[0123]

[0124] Where s is a fixed value of 0.01, u and v are random numbers between 0 and 1, and the calculation formula for σ is as follows:

[0125]

[0126] Where β is a constant with a fixed value of 1.5, and y and x are calculated as follows:

[0127] y=r×cos(θ)

[0128] x=r×sin(θ)

[0129] r=r1+U×D1

[0130] θ=-ω×D1+θ1

[0131]

[0132] r1 is a fixed value between [1,20], U is a constant with a fixed value of 0.00565, ω is a constant with a fixed value of 0.005, and D1 is the difference between dim and 1;

[0133] 3. Expand the development space. In this process:

[0134] X3(t+1)=(X best (t)-X M (t))×α

[0135] -rand+((U b -L b )×rand+L b )×δ

[0136] X3(t+1) is the solution of the t+1 iteration under the third upward search method, X best (t) is the random solution in the range [1, N] at the tth iteration, D is the dimension of the solution, and δ is the adjustment parameter value of the search phase, which is 0.1;

[0137] 4. Reduce the development space. In this step, the search space is further reduced, and the optimal solution X is finally obtained. best , the specific process is as follows:

[0138] X4(t+1)=QF×X best (t)-(G1×X(t)×rand)

[0139] -G2×Levy(D)+rand×G1

[0140] X4(t+1) is the solution of the t+1 iteration under the fourth upward search method, QF is a random solution within the quality function of the balanced search strategy, G1 is a variety of search methods in the AO algorithm, and G2 is a decreasing value of [0,2] used to represent the slope between the initial solution and the optimal solution. The specific formula is as follows:

[0141]

[0142] G1=2×rand-1

[0143]

[0144] When the loop end condition t≥T is reached, that is, the maximum number of iterations is reached, the optimal c and g values ​​are output, and multiple candidate parameter pairs are obtained. The fitness of multiple candidate parameter pairs is determined based on the fitness function. Since four types of faults need to be classified, the number of samples of different faults under actual working conditions may be different. At the same time, the misjudgment of the fault type will affect the health status of the equipment. Therefore, the weighted inter-class misjudgment rate can be used as the fitness function:

[0145]

[0146] Where fitness is the fitness of the candidate parameter pair, M is the total number of PRPS image samples, and M j FN is the number of samples of j-type faults. Since there may be multiple types of partial discharge faults, the number of samples of j-type faults can be understood as the number of samples of a certain type of partial discharge fault. j is the number of misjudgment samples of j types of faults, which can be understood as the number of misjudgment samples of a certain type of partial discharge fault; TP j is the number of samples with correct identification of fault j, which can be understood as the number of samples with correct identification of a certain type of partial discharge fault. The fitness of the candidate parameter pair is calculated by the fitness function.

[0147] S170, determining a target SVM recognition model according to the parameter pair with the greatest fitness.

[0148] In this embodiment, the fitness corresponding to multiple candidate parameter pairs is calculated by the fitness function; the multiple candidate parameter pairs are sorted according to the size of the fitness value; the parameter pair with the largest fitness is determined as the target c parameter and the target g parameter; and the target SVM recognition model is determined according to the target c parameter and the target g parameter.

[0149] S180, inputting the image features of the PRPS image into a target SVM recognition model to obtain the GIS equipment partial discharge type corresponding to the partial discharge data.

[0150] Specifically, the image features of the PRPS image extracted by the CLBP algorithm are input into the target SVM recognition model to determine the local discharge type of GIS equipment corresponding to the local discharge data, thereby achieving accurate identification of the local discharge type, effectively reducing the misjudgment rate of faults, and improving the fault recognition accuracy.

[0151] In this embodiment, local discharge data is obtained; the local discharge data is converted into a pulse sequence phase distribution image; the image features of the pulse sequence phase distribution image are extracted by a complete local binary pattern algorithm; the range interval corresponding to the parameter pair of the SVM recognition model is obtained; multiple parameter pairs are generated within the range interval to form a parameter pair set; a search operation is performed on the parameter pair set until the fitness of each parameter pair is obtained; a target SVM recognition model is determined according to the parameter pair with the largest fitness; the image features of the PRPS image are input into the target SVM recognition model to obtain the local discharge type of the GIS equipment corresponding to the local discharge data; wherein the search operation includes: searching the parameter pair set to obtain a candidate parameter pair; determining the fitness of the candidate parameter pair based on the fitness function; solving the problem in the prior art that background noise exists in the local discharge signal acquisition process and is difficult to remove, which affects the identification of faults, and realizes accurate identification of local discharge types; effectively reducing the misjudgment rate of faults, reducing the impact of sample differences of different fault types in actual engineering on model performance, improving the fault identification accuracy, and improving the health status detection level of GIS equipment.

[0152] Embodiment 2

[0153] Figure 3 This is a flowchart of determining the fitness of a candidate parameter pair based on a fitness function provided in the second embodiment of the present invention. The technical solution of this embodiment is further refined on the basis of the above embodiment. Figure 3 As shown, the method includes:

[0154] S210, obtaining an image sample set, wherein the image sample set includes: PRPS image samples and GIS equipment partial discharge types corresponding to the PRPS image samples.

[0155] Specifically, PRPS image samples of corresponding GIS equipment partial discharge types that have been set can be randomly acquired in advance to form an image sample set; the image sample set includes: PRPS image samples and GIS equipment partial discharge types corresponding to the PRPS image samples.

[0156] S220: Generate a first SVM recognition model based on the parameter pair to be selected.

[0157] The first SVM recognition model may refer to a target SVM recognition model. Specifically, the first SVM recognition model may be generated based on a plurality of candidate parameter pairs obtained through searching.

[0158] S230, inputting the PRPS image samples in the image sample set into a first SVM recognition model to obtain a predicted GIS equipment partial discharge type;

[0159] S240, determining the number of misjudged samples and the number of correctly identified samples corresponding to each GIS equipment partial discharge type according to the GIS equipment partial discharge type corresponding to the PRPS image samples in the image sample set and the predicted GIS equipment partial discharge type corresponding to the PRPS image samples.

[0160] Specifically, the number of misjudged samples can be understood as the number of PRPS image samples that incorrectly identify the local discharge type of each GIS device. Identification error can mean that the local discharge type of GIS equipment corresponding to the PRPS image sample in the image sample set and the predicted local discharge type of GIS equipment corresponding to the PRPS image sample are different; the number of correctly identified samples can be understood as the number of PRPS image samples that correctly identify the local discharge type of each GIS device. Correct identification can mean that the local discharge type of GIS equipment corresponding to the PRPS image sample in the image sample set and the predicted local discharge type of GIS equipment corresponding to the PRPS image sample are the same.

[0161] S250: Determine, according to the image sample set, the number of PRPS image samples, the number of partial discharge types, and the number of PRPS image samples corresponding to each partial discharge type.

[0162] The number of partial discharge types may refer to how many types of partial discharge there are.

[0163] In this embodiment, the number of PRPS image samples, the number of local discharge types and the number of PRPS image samples corresponding to each local discharge type can be determined directly based on the PRPS image samples in the image sample set and the local discharge types of GIS equipment corresponding to the PRPS image samples.

[0164] S260, determining the fitness of the candidate parameter pair according to the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, the number of correctly identified samples corresponding to each GIS device local discharge type, and the fitness function.

[0165] Specifically, the fitness of the candidate parameter pair can be determined through a fitness function based on the determined number of PRPS image samples, the number of each local discharge type, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, and the number of correctly identified samples corresponding to each GIS device local discharge type.

[0166] In this embodiment, an image sample set can be obtained in advance, and the image sample set includes: PRPS image samples and GIS equipment local discharge types corresponding to the PRPS image samples; then, a first SVM recognition model is generated based on the candidate parameter pair, and the PRPS image samples in the image sample set are input into the first SVM recognition model to obtain the predicted GIS equipment local discharge type; according to the GIS equipment local discharge type actually corresponding to the PRPS image samples in the image sample set and the corresponding predicted GIS equipment local discharge type obtained by the PRPS image samples through the first SVM recognition model, the number of misjudged samples and the number of correctly identified samples corresponding to each GIS equipment local discharge type are determined; the number of PRPS image samples, the number of each local discharge type, and the number of PRPS image samples corresponding to each local discharge type can be determined according to the image sample set; the fitness of the candidate parameter pair is determined through a fitness function according to the number of PRPS image samples, the number of each local discharge type, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS equipment local discharge type, and the number of correctly identified samples corresponding to each GIS equipment local discharge type, which can effectively reduce the misjudgment rate of faults, reduce the impact of sample differences of different fault types in actual engineering on model performance, and improve the fault recognition accuracy.

[0167] Optionally, the fitness function is:

[0168]

[0169] Among them, fitness is the fitness of the candidate parameter pair, M is the number of PRPS image samples, and M j is the number of PRPS image samples corresponding to the jth GIS equipment partial discharge type, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to the j types of GIS equipment partial discharge.

[0170] In this embodiment, the fitness of the candidate parameter pair can be determined by the fitness function, which can effectively reduce the misjudgment rate of faults, reduce the impact of sample differences of different fault types in actual engineering on model performance, and improve the fault recognition accuracy.

[0171] Optionally, the fitness of the candidate parameter pair is determined according to the number of PRPS image samples, the number of partial discharge types, the number of PRPS image samples corresponding to each partial discharge type, the number of misjudged samples corresponding to each GIS device partial discharge type, the number of correctly identified samples corresponding to each GIS device partial discharge type, and the fitness function, including:

[0172] Obtain the weight of the number of PRPS image samples, the weight of the number of partial discharge types, the weight of the number of PRPS image samples corresponding to each partial discharge type, the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, and the weight of the number of correctly identified samples corresponding to each GIS equipment partial discharge type;

[0173] The fitness of the candidate parameter pair is determined according to the weight of the number of PRPS image samples, the weight of the number of local discharge types, the weight of the number of PRPS image samples corresponding to each local discharge type, the weight of the number of misjudged samples corresponding to each GIS device local discharge type, the weight of the number of correctly identified samples corresponding to each GIS device local discharge type, the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, the number of correctly identified samples corresponding to each GIS device local discharge type and the fitness function.

[0174] Among them, the weight of the number of PRPS image samples, the weight of the number of local discharge types, the weight of the number of PRPS image samples corresponding to each local discharge type, the weight of the number of misjudged samples corresponding to each GIS device local discharge type, and the weight of the number of correctly identified samples corresponding to each GIS device local discharge type can be pre-set. For example, the weight of the number of PRPS image samples can be set to a1, the weight of the number of local discharge types can be set to a2, the weight of the number of PRPS image samples corresponding to each local discharge type can be set to a3, the weight of the number of misjudged samples corresponding to each GIS device local discharge type can be set to a4, and the weight of the number of correctly identified samples corresponding to each GIS device local discharge type can be set to a5. a1, a2, a3, a4, and a5 can be set equal, that is, a1=a2=a3=a4=a5=1; it can also be determined according to actual conditions, and this embodiment does not limit this.

[0175] The fitness function is:

[0176]

[0177] Where M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth type of GIS equipment partial discharge, K1 is the weight of the number of PRPS image samples corresponding to each type of partial discharge, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, K2 is the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, TP jis the number of correctly identified samples corresponding to j types of GIS equipment partial discharge, and K3 is the weight of the number of correctly identified samples corresponding to each type of GIS equipment partial discharge.

[0178] In this embodiment, the fitness of the candidate parameter pair can be determined through a fitness function based on the weight of the number of PRPS image samples, the weight of the number of local discharge types, the weight of the number of PRPS image samples corresponding to each local discharge type, the weight of the number of misjudged samples corresponding to each GIS device local discharge type, the weight of the number of correctly identified samples corresponding to each GIS device local discharge type, the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, and the number of correctly identified samples corresponding to each GIS device local discharge type.

[0179] Optionally, multiple parameter pairs are generated within the range interval to form a parameter pair set, including:

[0180] Generate parameter pairs according to the upper and lower bounds of the range interval corresponding to the parameter pairs, the target weight, and the target formula;

[0181] A parameter pair set is determined based on the generated parameter pairs.

[0182] The upper and lower bounds of the range interval corresponding to the parameter pair may refer to the search upper and lower bounds within the range interval corresponding to the parameter pair, and are not necessarily the maximum value or the minimum value; each parameter pair has its corresponding target weight. Specifically, the target weight corresponding to the parameter pair may be determined first, and then the parameter pair may be generated according to the upper and lower bounds of the range interval corresponding to the parameter pair, the target weight, and the target formula, and the generated parameter pairs may be grouped to generate a parameter pair set.

[0183] In this embodiment, parameter pairs can be generated according to the upper and lower bounds of the range interval corresponding to the parameter pairs, the target weight and the target formula; the efficiency of generating parameter pairs is improved, and the parameter pair set is determined according to the generated parameter pairs.

[0184] Optionally, the target formula is:

[0185] X i+1 =X min +λ i+1 (X max -X min );

[0186]

[0187] Among them, X i+1 is the i+1th parameter pair, X max is the upper bound of the range interval corresponding to the parameter pair, X minis the lower bound of the range interval corresponding to the parameter pair, λ i+1 is the target weight corresponding to the i+1th parameter pair, λ i is the target weight corresponding to the i-th parameter pair.

[0188] In this embodiment, the target weight corresponding to the parameter pair may be determined first, and then the parameter pair may be generated according to the upper and lower bounds of the range interval corresponding to the parameter pair, the target weight, and the target formula.

[0189] Optionally, the PRPS image samples include: at least one of: PRPS image samples corresponding to a tip discharge fault type, PRPS image samples corresponding to a suspended discharge fault type, PRPS image samples corresponding to a surface discharge fault type, PRPS image samples corresponding to an internal discharge fault type, and PRPS image samples corresponding to a particle discharge fault type.

[0190] For example, Figure 4 A CLBP-AO-SVM algorithm flow chart provided in the second embodiment of the present invention is as follows: Figure 4 As shown, in this embodiment, the steps of determining the GIS equipment local discharge type corresponding to the local discharge data by using the CLBP-AO-SVM algorithm are as follows:

[0191] 1. Input PRPS image;

[0192] 2. Extract the image features of PRPS images through CLBP algorithm;

[0193] 3. Initialize the parameters of the improved AO algorithm and search for the upper and lower bounds U b , L b , population number M, solution space X, current solution vector X i (c, g), X i The dimension dim=2, the maximum number of iterations is T.

[0194] 4. Through the fitness function, that is, the weighted inter-class misjudgment rate: Calculate the current fitness function value.

[0195] 5. Update the c and g parameters according to the calculated fitness function value.

[0196] 6. Determine whether the maximum number of iterations has been reached or the preset fitness threshold has been met. If not, calculate the fitness value again until the maximum number of iterations has been reached or the preset threshold has been met. Then stop the iteration and output the optimal c and g parameters.

[0197] 7. Input the optimal c parameters and g parameters into the SVM model to obtain the local discharge type (i.e. fault type) of the GIS equipment corresponding to the PRPS image.

[0198] In this embodiment, a PRPS image is input; image features of the PRPS image are extracted by the CLBP algorithm; parameters of the improved AO algorithm are initialized; the current fitness function value is calculated by the fitness function; c and g parameters are updated according to the calculated fitness function value and it is determined whether the maximum number of iterations is reached to determine the optimal c parameter and g parameter; the optimal c parameter and g parameter are input into the SVM model to obtain the local discharge type of GIS equipment corresponding to the local discharge data, which solves the problem in the prior art that background noise exists in the local discharge signal acquisition process and is difficult to remove, thereby affecting the identification of faults, realizes accurate identification of local discharge types, effectively reduces the misjudgment rate of faults, reduces the influence of sample differences of different fault types in actual projects on model performance, improves the fault identification accuracy, and enhances the health status detection level of GIS equipment.

[0199] Embodiment 3

[0200] Figure 5 This is a schematic diagram of the structure of a device for determining the type of partial discharge of GIS equipment provided in the third embodiment of the present invention. Figure 5 As shown, the device comprises:

[0201] A first acquisition module 310, used to acquire partial discharge data;

[0202] A conversion module 320, for converting partial discharge data into a pulse sequence phase distribution PRPS image;

[0203] An extraction module 330 is used to extract image features of the PRPS image using a complete local binary pattern method;

[0204] The second acquisition module 340 is used to obtain the range interval corresponding to the parameter pair of the SVM recognition model;

[0205] A generating module 350, configured to generate a plurality of parameter pairs within a range interval to form a parameter pair set;

[0206] A search module 360, configured to perform a search operation on the parameter pair set until the fitness of each parameter pair is obtained;

[0207] A first determination module 370, for determining a target SVM recognition model according to the parameter pair with the maximum fitness;

[0208] The second determination module 380 is used to input the image features of the PRPS image into a target SVM recognition model to obtain the GIS equipment local discharge type corresponding to the local discharge data;

[0209] The search module 360 ​​includes:

[0210] A searching unit, configured to search the parameter pair set to obtain a candidate parameter pair;

[0211] The determination unit is used to determine the fitness of the candidate parameter pair based on the fitness function.

[0212] Optionally, determine the unit, specifically for:

[0213] Acquire an image sample set, wherein the image sample set includes: PRPS image samples and GIS equipment local discharge types corresponding to the PRPS image samples;

[0214] Generate a first SVM recognition model based on the candidate parameter pair;

[0215] Inputting the PRPS image samples in the image sample set into a first SVM recognition model to obtain a predicted GIS equipment partial discharge type;

[0216] According to the GIS equipment partial discharge type corresponding to the PRPS image samples in the image sample set and the predicted GIS equipment partial discharge type corresponding to the PRPS image samples, the number of misjudged samples and the number of correctly identified samples corresponding to each GIS equipment partial discharge type are determined;

[0217] Determine the number of PRPS image samples, the number of partial discharge types, and the number of PRPS image samples corresponding to each partial discharge type according to the image sample set;

[0218] The fitness of the candidate parameter pairs is determined according to the number of PRPS image samples, the number of partial discharge types, the number of PRPS image samples corresponding to each partial discharge type, the number of misjudged samples corresponding to each GIS equipment partial discharge type, the number of correctly identified samples corresponding to each GIS equipment partial discharge type, and the fitness function.

[0219] Optionally, the fitness function is:

[0220]

[0221] Where M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth GIS equipment partial discharge type, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to the j types of GIS equipment partial discharge.

[0222] Optionally, the determination unit is further used to:

[0223] Obtain the weight of the number of PRPS image samples, the weight of the number of partial discharge types, the weight of the number of PRPS image samples corresponding to each partial discharge type, the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, and the weight of the number of correctly identified samples corresponding to each GIS equipment partial discharge type;

[0224] Determine the fitness of the candidate parameter pair according to the weight of the number of PRPS image samples, the weight of the number of local discharge types, the weight of the number of PRPS image samples corresponding to each local discharge type, the weight of the number of misjudged samples corresponding to each GIS device local discharge type, the weight of the number of correctly identified samples corresponding to each GIS device local discharge type, the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, the number of correctly identified samples corresponding to each GIS device local discharge type, and the fitness function;

[0225] The fitness function is:

[0226]

[0227] Where M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth type of GIS equipment partial discharge, K1 is the weight of the number of PRPS image samples corresponding to each type of partial discharge, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, K2 is the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to j types of GIS equipment partial discharge, and K3 is the weight of the number of correctly identified samples corresponding to each type of GIS equipment partial discharge.

[0228] Optionally, the generation module is specifically used to:

[0229] Generate parameter pairs according to the upper and lower bounds of the range interval corresponding to the parameter pairs, the target weight, and the target formula;

[0230] A parameter pair set is determined based on the generated parameter pairs.

[0231] Optionally, the target formula is:

[0232] X i+1 =X min +λ i+1 (X max -X min );

[0233]

[0234] Among them, X i+1 is the i+1th parameter pair, X max is the upper bound of the range interval corresponding to the parameter pair, X min is the lower bound of the range interval corresponding to the parameter pair, λ i+1 is the target weight corresponding to the i+1th parameter pair, λ i is the target weight corresponding to the i-th parameter pair.

[0235] Optionally, the PRPS image samples include: at least one of: PRPS image samples corresponding to a tip discharge fault type, PRPS image samples corresponding to a suspended discharge fault type, PRPS image samples corresponding to a surface discharge fault type, PRPS image samples corresponding to an internal discharge fault type, and PRPS image samples corresponding to a particle discharge fault type.

[0236] The device for determining the partial discharge type of GIS equipment provided in the embodiment of the present invention can execute the method for determining the partial discharge type of GIS equipment provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0237] Embodiment 4

[0238] Figure 6 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0239] like Figure 6As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0240] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0241] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the method for determining the type of partial discharge of GIS equipment.

[0242] In some embodiments, the method for determining the type of partial discharge of GIS equipment may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for determining the type of partial discharge of GIS equipment described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the method for determining the type of partial discharge of GIS equipment in any other appropriate manner (e.g., by means of firmware).

[0243] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0244] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0245] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0246] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0247] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0248] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0249] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0250] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining the type of partial discharge of GIS equipment, characterized in that: include: Obtain partial discharge data; Convert partial discharge data into pulse sequence phase distribution PRPS images; The image features of PRPS images are extracted using the complete local binary pattern method; Get the range interval corresponding to the parameter pair of the SVM recognition model; Generate multiple parameter pairs within the range interval to form a parameter pair set; Perform a search operation on the parameter pair set until the fitness of each parameter pair is obtained; Determine the target SVM recognition model based on the parameter pair with the largest fitness; Inputting the image features of the PRPS image into a target SVM recognition model to obtain a local discharge type of the GIS equipment corresponding to the local discharge data; The search operation includes: Searching the parameter pair set to obtain parameter pairs to be selected; Determine the fitness of the candidate parameter pair based on the fitness function; The step of determining the fitness of the candidate parameter pair based on the fitness function comprises: Acquire an image sample set, wherein the image sample set includes: PRPS image samples and GIS equipment local discharge types corresponding to the PRPS image samples; Generate a first SVM recognition model based on the candidate parameter pair; Inputting the PRPS image samples in the image sample set into a first SVM recognition model to obtain a predicted GIS equipment partial discharge type; According to the GIS equipment partial discharge type corresponding to the PRPS image samples in the image sample set and the predicted GIS equipment partial discharge type corresponding to the PRPS image samples, the number of misjudged samples and the number of correctly identified samples corresponding to each GIS equipment partial discharge type are determined; Determine the number of PRPS image samples, the number of partial discharge types, and the number of PRPS image samples corresponding to each partial discharge type according to the image sample set; The fitness of the candidate parameter pairs is determined according to the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS equipment local discharge type, the number of correctly identified samples corresponding to each GIS equipment local discharge type, and the fitness function.

2. The method according to claim 1, characterized in that The fitness function is: ； in, is the fitness of the candidate parameter pair, M is the number of PRPS image samples, and M j is the number of PRPS image samples corresponding to the jth GIS equipment partial discharge type, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to the j types of GIS equipment partial discharge.

3. The method according to claim 1, characterized in that The fitness of the candidate parameter pair is determined according to the number of PRPS image samples, the number of partial discharge types, the number of PRPS image samples corresponding to each partial discharge type, the number of misjudged samples corresponding to each GIS equipment partial discharge type, the number of correctly identified samples corresponding to each GIS equipment partial discharge type, and the fitness function, including: Obtain the weight of the number of PRPS image samples, the weight of the number of partial discharge types, the weight of the number of PRPS image samples corresponding to each partial discharge type, the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, and the weight of the number of correctly identified samples corresponding to each GIS equipment partial discharge type; Determine the fitness of the candidate parameter pair according to the weight of the number of PRPS image samples, the weight of the number of local discharge types, the weight of the number of PRPS image samples corresponding to each local discharge type, the weight of the number of misjudged samples corresponding to each GIS device local discharge type, the weight of the number of correctly identified samples corresponding to each GIS device local discharge type, the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, the number of correctly identified samples corresponding to each GIS device local discharge type, and the fitness function; The fitness function is: ； Where, M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth type of GIS equipment partial discharge, is the weight of the number of PRPS image samples corresponding to each partial discharge type, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, is the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to the jth GIS equipment partial discharge type, The weight of the number of correctly identified samples corresponding to each GIS equipment partial discharge type.

4. The method according to claim 1, characterized in that: Generate multiple parameter pairs within the range interval to form a parameter pair set, including: Generate parameter pairs according to the upper and lower bounds of the range interval corresponding to the parameter pairs, the target weight, and the target formula; A parameter pair set is determined based on the generated parameter pairs.

5. The method according to claim 4, wherein the target formula is: ； ； in, is the i+1th parameter pair, is the upper bound of the range interval corresponding to the parameter pair, is the lower bound of the range interval corresponding to the parameter pair, is the target weight corresponding to the i+1th parameter pair, is the target weight corresponding to the i-th parameter pair.

6. The method according to claim 1, characterized in that The PRPS image samples include at least one of: PRPS image samples corresponding to a tip discharge fault type, PRPS image samples corresponding to a suspended discharge fault type, PRPS image samples corresponding to a surface discharge fault type, PRPS image samples corresponding to an internal discharge fault type, and PRPS image samples corresponding to a particle discharge fault type.

7. A device for determining the type of partial discharge of GIS equipment, characterized in that: include: A first acquisition module, used for acquiring partial discharge data; A conversion module, used for converting partial discharge data into a pulse sequence phase distribution PRPS image; An extraction module, for extracting image features of the PRPS image using a complete local binary pattern method; The second acquisition module is used to obtain the range interval corresponding to the parameter pair of the SVM recognition model; A generating module, used for generating a plurality of parameter pairs within a range interval to form a parameter pair set; A search module, used to perform a search operation on the parameter pair set until the fitness of each parameter pair is obtained; A first determination module is used to determine a target SVM recognition model according to a parameter pair with the largest fitness; The second determination module is used to input the image features of the PRPS image into a target SVM recognition model to obtain the GIS equipment local discharge type corresponding to the local discharge data; Wherein, the search module includes: A searching unit, configured to search the parameter pair set to obtain a candidate parameter pair; A determination unit, used for determining the fitness of the candidate parameter pair based on the fitness function; The determining unit is specifically configured to: Acquire an image sample set, wherein the image sample set includes: PRPS image samples and GIS equipment local discharge types corresponding to the PRPS image samples; Generate a first SVM recognition model based on the candidate parameter pair; Inputting the PRPS image samples in the image sample set into a first SVM recognition model to obtain a predicted GIS equipment partial discharge type; According to the GIS equipment partial discharge type corresponding to the PRPS image samples in the image sample set and the predicted GIS equipment partial discharge type corresponding to the PRPS image samples, the number of misjudged samples and the number of correctly identified samples corresponding to each GIS equipment partial discharge type are determined; Determine the number of PRPS image samples, the number of partial discharge types, and the number of PRPS image samples corresponding to each partial discharge type according to the image sample set; The fitness of the candidate parameter pairs is determined according to the number of PRPS image samples, the number of partial discharge types, the number of PRPS image samples corresponding to each partial discharge type, the number of misjudged samples corresponding to each GIS device partial discharge type, the number of correctly identified samples corresponding to each GIS device partial discharge type, and the fitness function.

8. The device according to claim 7, characterized in that The fitness function is: ； Where, M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth GIS equipment partial discharge type, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to the j types of GIS equipment partial discharge.

9. The device according to claim 7, characterized in that Determine the unit, specifically also used for: Obtain the weight of the number of PRPS image samples, the weight of the number of partial discharge types, the weight of the number of PRPS image samples corresponding to each partial discharge type, the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, and the weight of the number of correctly identified samples corresponding to each GIS equipment partial discharge type; Determine the fitness of the candidate parameter pair according to the weight of the number of PRPS image samples, the weight of the number of local discharge types, the weight of the number of PRPS image samples corresponding to each local discharge type, the weight of the number of misjudged samples corresponding to each GIS device local discharge type, the weight of the number of correctly identified samples corresponding to each GIS device local discharge type, the number of PRPS image samples, the number of local discharge types, the number of PRPS image samples corresponding to each local discharge type, the number of misjudged samples corresponding to each GIS device local discharge type, the number of correctly identified samples corresponding to each GIS device local discharge type, and the fitness function; The fitness function is: ； Where, M is the number of PRPS image samples, M j is the number of PRPS image samples corresponding to the jth type of GIS equipment partial discharge, is the weight of the number of PRPS image samples corresponding to each partial discharge type, FN j is the number of misjudged samples corresponding to the jth GIS equipment partial discharge type, is the weight of the number of misjudged samples corresponding to each GIS equipment partial discharge type, TP j is the number of correctly identified samples corresponding to the jth GIS equipment partial discharge type, The weight of the number of correctly identified samples corresponding to each GIS equipment partial discharge type.

10. The device according to claim 7, characterized in that The generation module is specifically used for: Generate parameter pairs according to the upper and lower bounds of the range interval corresponding to the parameter pairs, the target weight, and the target formula; A parameter pair set is determined based on the generated parameter pairs.

11. The device according to claim 10, wherein the target formula is: ； ； in, is the i+1th parameter pair, is the upper bound of the range interval corresponding to the parameter pair, is the lower bound of the range interval corresponding to the parameter pair, is the target weight corresponding to the i+1th parameter pair, is the target weight corresponding to the i-th parameter pair.

12. The device according to claim 7, characterized in that The PRPS image samples include at least one of: PRPS image samples corresponding to a tip discharge fault type, PRPS image samples corresponding to a suspended discharge fault type, PRPS image samples corresponding to a surface discharge fault type, PRPS image samples corresponding to an internal discharge fault type, and PRPS image samples corresponding to a particle discharge fault type.

13. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the local discharge type of GIS equipment according to any one of claims 1-6.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining the partial discharge type of GIS equipment according to any one of claims 1 to 6 when executed.

Images