Defect type identification method and device, computer device, and storage medium

Abstract

This application relates to a defect type identification method, apparatus, computer equipment, storage medium, and computer program product. The method includes: acquiring a partial discharge signal from a gas-insulated combined electrical equipment; identifying the partial discharge signal using a support vector machine (SVM); if no identification result is obtained from the SVM output, uploading the feature parameters of the partial discharge signal to a target evaluation system for evaluation to determine the target defect type; if the target defect type is not a preset defect type, training the SVM using the feature parameters to obtain a target SVM, which is then used to identify the defect type of the partial discharge signal to be identified. This method solves the problem of low accuracy in defect type identification for gas-insulated combined electrical equipment and improves the accuracy of defect type identification.

Description

Technical Field

[0001] This application relates to the field of power detection technology, and in particular to a defect type identification method, apparatus, computer equipment, storage medium, and computer program product. Background Technology

[0002] Gas-insulated switchgear (GIS) equipment is a new type of electrical equipment that encloses all electrical components, such as circuit breakers, disconnectors, grounding switches, instrument transformers, surge arresters, busbars, connectors, and outgoing terminals, in a grounded metal casing and fills it with sulfur hexafluoride (SF6) insulating gas with high electrical strength. It has the advantages of small footprint, light weight, anti-interference, and high reliability.

[0003] Because gas-insulated switchgear is a fully enclosed structure, accidents can lead to complex repair procedures, long repair times, and wide-ranging impacts. To ensure power grid safety, it is necessary to periodically inspect gas-insulated switchgear for partial discharge signals. Based on these signals, partial discharge maps are generated, which are then visually examined by highly experienced experts to determine the type of defect in the gas-insulated switchgear and formulate corresponding maintenance strategies.

[0004] However, the method of identifying defect types by visually inspecting partial discharge patterns by experts relies heavily on their professional skills and experience, thus resulting in a low accuracy rate in identifying defect types in gas-insulated combined electrical equipment. Summary of the Invention

[0005] Therefore, it is necessary to provide a defect type identification method, device, computer equipment, computer-readable storage medium, and computer program product that can improve the accuracy of defect type identification in gas-insulated combined electrical equipment in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a method for identifying defect types. The method includes:

[0007] Acquire partial discharge signals from gas-insulated combined electrical equipment;

[0008] The partial discharge signal was identified using a support vector machine.

[0009] If the recognition result output by the support vector machine is not obtained, the feature parameters of the partial discharge signal are uploaded to the target evaluation system so that the feature parameters of the partial discharge signal can be evaluated by the target evaluation system to obtain the target defect type of the partial discharge signal.

[0010] If the target defect type is not a preset defect type, the feature parameters of the partial discharge signal are used to train the support vector machine to obtain the target support vector machine, which is then used to identify the defect type of the partial discharge signal to be identified.

[0011] In one embodiment, the method further includes:

[0012] If the target defect type is the preset defect type, then the support vector machine is trained using the feature parameters of the partial discharge signal to obtain the target support vector machine.

[0013] In one embodiment, the method further includes:

[0014] Obtain characteristic parameters from simulated partial discharge sample data of different preset defect types;

[0015] Based on the feature parameters of the partial discharge sample data, a pre-defined support vector machine is trained to obtain the support vector machine.

[0016] In one embodiment, the characteristic parameters for obtaining simulated partial discharge sample data of different preset defect types further include:

[0017] The partial discharge sample data was normalized to obtain normalized partial discharge sample data.

[0018] The normalized partial discharge sample data is then normalized to obtain its characteristic parameters.

[0019] In one embodiment, training a preset support vector machine based on the feature parameters of the partial discharge sample data, and obtaining the support vector machine further includes:

[0020] The feature parameters of the partial discharge sample data are reduced in dimension to obtain the reduced feature parameters.

[0021] Based on the feature parameters after dimensionality reduction, the preset support vector machine is trained to obtain the support vector machine.

[0022] In one embodiment, based on the feature parameters after dimensionality reduction, a preset support vector machine is trained, and the support vector machine is further obtained by:

[0023] Based on the radial basis kernel function, the parameters of the preset support vector machine are optimized using the grid search method to obtain the support vector machine.

[0024] Secondly, this application also provides a defect type identification device. The device includes:

[0025] The acquisition module is used to acquire the partial discharge signal of the gas-insulated combined electrical equipment;

[0026] The identification module is used to identify the partial discharge signal using a support vector machine;

[0027] The evaluation module is used to upload the feature parameters of the partial discharge signal to the target evaluation system if the recognition result output by the support vector machine is not obtained, so that the feature parameters of the partial discharge signal can be evaluated by the target evaluation system to obtain the target defect type of the partial discharge signal.

[0028] The update module is used to train the support vector machine using the feature parameters of the partial discharge signal if the target defect type is not a preset defect type, so as to obtain a target support vector machine and use the target support vector machine to identify the defect type of the partial discharge signal to be identified.

[0029] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of any of the methods described above.

[0030] Fourthly, this application also provides a computer-readable storage medium. This computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0031] Fifthly, this application also provides a computer program product. This computer program product includes a computer program that, when executed by a processor, implements the steps of any of the methods described above.

[0032] The aforementioned defect type identification method, apparatus, computer equipment, storage medium, and computer program product acquire partial discharge signals from gas-insulated combined electrical equipment; use a support vector machine (SVM) to identify the partial discharge signals; if no identification result is obtained from the SVM, the feature parameters of the partial discharge signal are uploaded to a target evaluation system for evaluation to determine the target defect type; if the target defect type is not a preset defect type, the SVM is trained using the feature parameters to obtain a target SVM, which is then used to identify the defect type of the partial discharge signal to be identified. By first identifying the defect type of the partial discharge signal using the SVM, and then further identifying the feature parameters of the partial discharge signal for which the SVM cannot output a defect type identification result, and then using these feature parameters to train the SVM again, the resulting target SVM can more accurately identify the defect type of the partial discharge signal to be identified, thereby improving the accuracy of defect type identification for gas-insulated combined electrical equipment. Attached Figure Description

[0033] Figure 1 This is a flowchart illustrating a defect type identification method provided in an embodiment of this application;

[0034] Figure 2 This is a flowchart illustrating a support vector machine acquisition method provided in an embodiment of this application;

[0035] Figure 3 This is a flowchart illustrating a method for obtaining characteristic parameters of partial discharge sample data provided in an embodiment of this application.

[0036] Figure 4 This is a flowchart illustrating a support vector machine training method provided in an embodiment of this application.

[0037] Figure 5 This is a schematic diagram of the structure of a defect type identification device provided in the embodiments of this application;

[0038] Figure 6 This is a diagram showing the internal structure of a computer device in an embodiment of this application. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0040] This embodiment illustrates the application of the method to a computer device. It is understood that the method can also be applied to a server, and to a system that includes both computer devices and servers, and can be implemented through the interaction between the computer devices and servers.

[0041] Figure 1 This is a flowchart illustrating a defect type identification method provided in an embodiment of this application. The method is applied to computer equipment, such as an industrial control computer. In one embodiment, as shown... Figure 1 As shown, it includes the following steps:

[0042] S101, acquire the partial discharge signal of the gas-insulated combined electrical equipment.

[0043] In this embodiment, when the industrial control computer detects a partial discharge signal in the gas-insulated switchgear, it acquires the partial discharge signal of the gas-insulated switchgear as the discrimination input for the subsequent support vector machine.

[0044] S102 uses a support vector machine to identify partial discharge signals.

[0045] In this embodiment, the industrial control computer has a support vector machine (SVM) that has been trained on a preset SVM. The partial discharge signal obtained in S101 is used as the input of this SVM. The SVM performs defect identification on the partial discharge signal. If the SVM model determines that the partial discharge signal is one of the preset defect types, it outputs the determination result and completes one defect type identification.

[0046] Since the accuracy of defect type identification obtained by experts in traditional technologies is low, while the defect type identification in this embodiment is achieved by using support vector machines, which is not limited by the technical knowledge and experience of experts, thus improving the accuracy of defect type identification.

[0047] S103, if the recognition result output by the support vector machine is not obtained, the feature parameters of the partial discharge signal are uploaded to the target evaluation system so that the feature parameters of the partial discharge signal can be evaluated by the target evaluation system to obtain the target defect type of the partial discharge signal.

[0048] In this embodiment, if the support vector machine cannot obtain a discrimination result after the identification in S102 above, and the industrial control computer cannot obtain a defect type identification result, the industrial control computer can transmit the feature parameters of this partial discharge signal to the target evaluation system and use the target evaluation system to evaluate the feature parameters of the undiscriminable partial discharge signal.

[0049] The target evaluation system is an intelligent computer program system that incorporates a wealth of expert knowledge and experience in the field of partial discharge spectrum recognition. It leverages the knowledge and problem-solving methods of human experts to handle problems in this area. The system features automatic analysis and fault diagnosis, display of characteristic spectra from recorded detection data, display of historical trends of characteristic parameters and provision of data backtracking, and detailed operation and diagnostic logs. It also supports remote monitoring, allowing experts and users to monitor the system's operation remotely. When the system is unable to evaluate and identify uploaded characteristic parameters, or when other issues arise, experts can remotely assist in confirming the fault via the network. It should be noted that the target evaluation system can run on an industrial control computer or other different computer devices.

[0050] S104. If the target defect type is not the preset defect type, the support vector machine is trained using the feature parameters of the partial discharge signal to obtain the target support vector machine, which is then used to identify the defect type of the partial discharge signal to be identified.

[0051] In this embodiment, if the evaluation result of the target evaluation system is not one of the preset defect types, the evaluation result of the target evaluation system is treated as a new defect type. The feature parameters of this partial discharge signal are added to the feature parameters corresponding to the new defect type, and the support vector machine is trained using the feature parameters corresponding to the new defect type to obtain a retrained target support vector machine. The target support vector machine is then used to identify the partial discharge signal to be identified. Because the support vector machine is trained using the feature parameters corresponding to the new defect type, if the defect type of the partial discharge signal to be identified belongs to either the new defect type or the preset defect type, the target support vector machine can identify the defect type of the partial discharge signal to be identified as either the new defect type or the preset defect type.

[0052] The defect type identification method provided in this embodiment acquires the partial discharge signal of the gas-insulated combined electrical equipment and uses a support vector machine (SVM) to identify the partial discharge signal. If no identification result is obtained from the SVM, the feature parameters of the partial discharge signal are uploaded to a target evaluation system for evaluation. The target evaluation system then evaluates the feature parameters of the partial discharge signal to determine the target defect type. If the target defect type is not a preset defect type, the feature parameters of the partial discharge signal are used to train the SVM to obtain a target SVM, which is then used to identify the defect type of the partial discharge signal to be identified. Because the partial discharge signal is first identified using the SVM, and for partial discharge signals for which the SVM cannot output a defect type identification result, the target evaluation system further identifies the feature parameters of the partial discharge signal and uses these feature parameters to train the SVM again. This allows the resulting target SVM to more accurately identify the defect type of the partial discharge signal to be identified, thereby improving the accuracy of defect type identification for gas-insulated combined electrical equipment.

[0053] Optionally, the above defect type identification method can also be implemented in the following ways:

[0054] If the target defect type is a preset defect type, the support vector machine is trained using the feature parameters of the partial discharge signal to obtain the target support vector machine.

[0055] In this embodiment, if a preset defect type exists that is identical to the target defect type, the feature parameters of this partial discharge signal are added to the feature parameters corresponding to the preset defect type that is identical to the target defect type, thereby updating the feature parameters corresponding to the preset defect type that is identical to the target defect type. For ease of description, the preset defect type that is identical to the target defect type will be referred to as the target preset defect type. The feature parameters of this partial discharge signal can then be used to train the support vector machine to obtain the target support vector machine.

[0056] Alternatively, the support vector machine can be trained using the feature parameters corresponding to the updated target preset defect type to obtain the target support vector machine, and then the target support vector machine can be used to identify the partial discharge signal to be identified, thereby completing the identification of the defect type.

[0057] In this embodiment, if the target defect type is a preset defect type, the feature parameters of the partial discharge signal are used to train the support vector machine to obtain the target support vector machine, thereby enriching the training sample data of the support vector machine. Then, the target support vector machine is used to complete the identification of the defect type of the partial discharge signal to be identified, thereby improving the accuracy of defect type identification of gas-insulated combined electrical equipment.

[0058] Figure 2 This is a flowchart illustrating a support vector machine acquisition method provided in an embodiment of this application. (Refer to...) Figure 2 This embodiment relates to an optional implementation of a support vector machine. Based on the above embodiment, the defect type identification method further includes the following steps:

[0059] S201, Obtain the characteristic parameters of simulated partial discharge sample data of different preset defect types.

[0060] In this embodiment, S201, obtaining the characteristic parameters of simulated partial discharge sample data of different defect types can be achieved in the following way:

[0061] Method 1: Normalize the partial discharge sample data to obtain normalized partial discharge sample data, extract the feature parameters of the normalized partial discharge sample data, and use the extracted feature parameters of the normalized partial discharge sample data as feature parameters of simulated partial discharge sample data of different defect types.

[0062] Method 2: The partial discharge sample data is normalized to obtain normalized partial discharge sample data. The feature parameters of the normalized partial discharge sample data are extracted and used as feature parameters of the simulated partial discharge sample data for different defect types.

[0063] In this embodiment, M different preset defect types are simulated, for example, M is 4, and N samples are collected for each preset defect type, for example, N is 1000, that is, 1000 partial discharge sample data corresponding to the four preset defect types are obtained. Since the four common defect types are high-voltage conductor surface pinpoint defects, free metal particle defects, floating electrode defects, and solid insulation air gap defects, these four defect types are used as the four preset defect types, and the simulation methods are as follows:

[0064] Needle tip defects on the surface of high voltage conductors: A copper wire with a diameter of 1 mm and a length of 5 mm is placed radially along the high voltage conductor to simulate needle tip discharge;

[0065] Free metal particle defects: Ten copper wires, each 1 cm long and 1 mm in diameter, are randomly placed on the surface of the insulator of a gas-insulated combined electrical equipment. When a high voltage is applied, the copper wires will jump randomly, simulating the discharge of free metal particles.

[0066] Floating electrode defect: A stainless steel nut is attached to the high-voltage wire, and a 0.1mm thick epoxy resin plate is placed between the nut and the wire to simulate floating electrode discharge;

[0067] Solid insulation air gap defect: Seal a 1mm wide air gap between two flat electrodes by sealing a 10mm thick epoxy resin plate.

[0068] The aforementioned defect model was placed inside the casing of a gas-insulated combined electrical equipment (GEE), and the test voltage was slowly increased to detect partial discharge within the casing. When a discharge pulse was detected, the pressure was stopped. Once the partial discharge signal stabilized, the intensity, frequency, and phase of the partial discharge signal over a certain time period were recorded, and a corresponding partial discharge spectrum was generated. The partial discharge spectrum was then denoised using wavelet transform to remove noise from the signal. Wavelet transform denoising is a commonly used technique in this field and will not be elaborated upon here.

[0069] In this embodiment, the partial discharge sample data is normalized and standardized to obtain the characteristic parameters of the partial discharge sample data. Alternatively, either method one or method two can be used to obtain the characteristic parameters of the partial discharge sample data. This embodiment does not impose specific restrictions on the method of obtaining the characteristic parameters of the partial discharge sample data.

[0070] S202, based on the feature parameters of partial discharge sample data, a preset support vector machine is trained to obtain the support vector machine.

[0071] In this embodiment, S202, based on the feature parameters of the partial discharge sample data, a preset support vector machine is trained to obtain the support vector machine, which can be achieved in the following way:

[0072] Method 1: Without processing the feature parameters of the partial discharge sample data, the feature parameters of the partial discharge sample data obtained by S201 are directly used to train a preset support vector machine to obtain the support vector machine.

[0073] Method 2: Smooth the feature parameters of the partial discharge sample data, and use the smoothed feature parameters of the partial discharge sample data to train a preset support vector machine to obtain the support vector machine.

[0074] In this embodiment, the feature parameters of the partial discharge sample data are reduced in dimensionality. Based on the feature parameters of the dimensionality-reduced partial discharge sample data, a preset support vector machine is trained to obtain the support vector machine. Alternatively, either method one or method two can be used to train the preset support vector machine to obtain the support vector machine. This embodiment does not impose specific restrictions on the method of obtaining the support vector machine.

[0075] This embodiment acquires feature parameters of simulated partial discharge sample data of different preset defect types, and trains a preset support vector machine based on the feature parameters of the partial discharge sample data to obtain the support vector machine. Then, the support vector machine can be used to identify the defect type of the partial discharge signal, thereby improving the accuracy of defect type identification of gas-insulated combined electrical equipment.

[0076] Figure 3 This is a flowchart illustrating a method for obtaining characteristic parameters of partial discharge sample data provided in an embodiment of this application. (Refer to...) Figure 3 This embodiment relates to an optional implementation method for obtaining characteristic parameters of simulated partial discharge sample data of different defect types. Based on the above embodiment, S201 further includes the following steps:

[0077] S301, normalize the partial discharge sample data to obtain normalized partial discharge sample data.

[0078] In this embodiment, each power frequency cycle is divided into two half cycles, namely the positive half cycle and the negative half cycle, and the discharge amplitude sequence in the partial discharge sample data is normalized according to formula (1).

[0079]

[0080] In equation (1), P k P′ represents the normalized amplitude of the k-th discharge signal in the discharge amplitude sequence. k P′ represents the amplitude of the k-th discharge signal in the discharge amplitude sequence. min P′ represents the minimum amplitude in the discharge amplitude sequence. max This represents the maximum amplitude value in the discharge amplitude sequence.

[0081] S302, normalize the normalized partial discharge sample data to obtain the characteristic parameters of the partial discharge sample data.

[0082] In this embodiment, the average discharge amplitude E of the positive and negative half-cycles is calculated respectively. mag_p E mag_n Standard deviation S of discharge amplitude during positive and negative half-cycles mag_p S mag_nand the total number of discharges N in the positive and negative half-cycles. mag_p N mag_n Based on formulas (2), (3), and (4), the three statistical characteristic parameters E describing the discharge amplitude are calculated respectively. mag S mag N mag :

[0083] E mag =max(E mag_p E mag_n (2)

[0084] S mag =max(S) mag_p S mag_n (3)

[0085] N mag =N mag_p +N mag_n (4)

[0086] Calculate the average discharge time interval E for the positive and negative half-cycles respectively. int_p E int_n and the total number N of discharge intervals in the positive and negative half-cycles int_p N int_n Based on formulas (5) and (6), two statistical characteristic parameters E describing the characteristics of the discharge time interval are calculated. int and E int :

[0087] E int =max(E int_p E int_n (5)

[0088] N int =N int_p +N int_n (6)

[0089] The characteristic ΔT describing the discharge interval is calculated according to formula (7). mean :

[0090]

[0091] In equation (7), ω represents the ωth discharge interval, and N represents the total number of discharge intervals.

[0092] The characteristic r and N used to describe the discharge distribution region are calculated according to formulas (8) and (9). reg :

[0093]

[0094] Nreg =N reg_n +N reg_p (9)

[0095] Therefore, the characteristic parameter E of the partial discharge sample data can be obtained. mag S mag N mag E int N int ΔT mean r and N reg A feature space can be constructed based on the characteristic parameters of partial discharge sample data. Since the numerical distributions of the various characteristic parameters differ significantly, normalization is necessary to remove unit limitations and transform them into dimensionless pure numerical values. This facilitates comparison and weighting of features of different magnitudes; for example, deviation normalization can be used to transform the characteristic parameters.

[0096] In this embodiment, the partial discharge sample data is normalized to obtain normalized partial discharge sample data. The normalized partial discharge sample data is then standardized to obtain the feature parameters of the partial discharge sample data. This improves the accuracy of the feature parameters of the partial discharge sample data, and further improves the accuracy of defect type identification of the support vector machine based on the feature parameters of the partial discharge sample data.

[0097] Figure 4 This is a flowchart illustrating a support vector machine training method provided in an embodiment of this application. (Refer to...) Figure 4 This embodiment relates to an optional implementation of how to train a support vector machine. Based on the above embodiment, S202 further includes the following steps:

[0098] S401 performs dimensionality reduction processing on the feature parameters of the partial discharge sample data to obtain the dimensionality-reduced feature parameters.

[0099] In this embodiment, the feature space constructed from the feature parameters obtained through the above steps often has a high dimensionality, but the contribution of each feature parameter to the subsequent support vector machine classification and recognition is not the same. In order to find those features that are beneficial to recognition and remove redundant information, it is necessary to perform dimensionality reduction processing on the feature parameters of the above feature space. For example, principal component analysis, kernel principal component analysis, etc. can be used to perform dimensionality reduction processing on the feature parameters of the above feature space. This embodiment does not limit the method of dimensionality reduction processing.

[0100] S402, based on the feature parameters after dimensionality reduction, train a preset support vector machine to obtain the support vector machine.

[0101] In this embodiment, S402, based on the feature parameters after dimensionality reduction, a preset support vector machine is trained to obtain the support vector machine, which can be achieved in the following way:

[0102] Method 1: Use a one-to-one algorithm to construct a support vector machine classifier between any two sample data, resulting in M×(M-1) / 2 support vector machine classifiers. Train the support vector machine classifiers based on the feature parameters of the partial discharge sample data.

[0103] Method 2: Use a hybrid kernel function-based support vector machine. Use a one-to-one algorithm to build a support vector machine classifier between any two sample data, resulting in M×(M-1) / 2 support vector machine classifiers. Train the support vector machine classifiers based on the feature parameters of the partial discharge sample data.

[0104] In this embodiment, the parameters of the preset support vector machine are optimized using a grid search method based on the radial basis kernel function to obtain the support vector machine. Alternatively, the preset support vector machine can be trained using either method one or method two to obtain the support vector machine. This embodiment does not impose specific restrictions on the method of training the support vector machine.

[0105] This embodiment reduces the dimensionality of the feature parameters of partial discharge sample data to obtain the dimensionality-reduced feature parameters. Based on the dimensionality-reduced feature parameters, a preset support vector machine is trained to obtain the support vector machine. The support vector machine is then used to identify the partial discharge signal, thereby completing the identification of the defect type and improving the accuracy of defect type identification for gas-insulated combined electrical equipment.

[0106] Optionally, the above-mentioned training of a preset support vector machine based on the feature parameters after dimensionality reduction can also be achieved in the following ways:

[0107] Based on the radial basis kernel function, the parameters of the preset support vector machine are optimized using the grid search method to obtain the support vector machine.

[0108] In this embodiment, a preset support vector machine is trained based on the feature parameters after dimensionality reduction. The partial discharge sample data corresponding to different preset defect types obtained from the above simulation are used as the original test samples. After screening, the original test samples form a pattern sample library. Some samples in the pattern sample library are selected as training samples, and the rest are used as test samples. The training samples are not used as test samples.

[0109] Using a radial basis function kernel, after selecting the kernel function, it is necessary to determine the support vector kernel function parameters and control structure risk parameters of the preset support vector machine (SVM), namely the penalty coefficient C and parameter w, in order to control the generalization ability of the SVM. A grid search method is used as the optimization algorithm to obtain the optimal parameters of the penalty coefficient C and parameter w. The learning process of training samples is essentially finding the optimal parameters of the preset SVM's penalty coefficient C and parameter w to obtain the trained SVM. Alternatively, either method one or method two can be used to obtain the SVM. The SVM is then used to identify partial discharge signals, thereby completing the defect type identification and improving the accuracy of defect type identification for gas-insulated combined electrical equipment.

[0110] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0111] Based on the same inventive concept, this application also provides a defect type identification device for implementing the defect type identification method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more defect type identification device embodiments provided below can be found in the limitations of the defect type identification method described above, and will not be repeated here.

[0112] Reference Figure 5 , Figure 5 This is a schematic diagram of a defect type identification device provided in an embodiment of this application. The device 500 includes: an acquisition module 501, an identification module 502, an evaluation module 503, and an update module 504, wherein:

[0113] The acquisition module 501 is used to acquire the partial discharge signal of the gas-insulated combined electrical equipment.

[0114] The identification module 502 is used to identify partial discharge signals using a support vector machine.

[0115] The evaluation module 503 is used to upload the feature parameters of the partial discharge signal to the target evaluation system if the recognition result output by the support vector machine is not obtained, so that the feature parameters of the partial discharge signal can be evaluated by the target evaluation system to obtain the target defect type of the partial discharge signal.

[0116] The update module 504 is used to train the support vector machine with the feature parameters of the partial discharge signal if the target defect type is not the preset defect type, so as to obtain the target support vector machine and use the target support vector machine to identify the defect type of the partial discharge signal to be identified.

[0117] The defect type identification device provided in this embodiment acquires partial discharge signals from gas-insulated combined electrical equipment and uses a support vector machine (SVM) to identify these signals. If no identification result is obtained from the SVM, the feature parameters of the partial discharge signal are uploaded to a target evaluation system for evaluation. The target evaluation system then evaluates these feature parameters to determine the target defect type of the partial discharge signal. If the target defect type is not a preset defect type, the feature parameters of the partial discharge signal are used to train the SVM, resulting in a target SVM. This target SVM is then used to identify the defect type of the partial discharge signal to be identified. Because the SVM is first used to identify the defect type of the partial discharge signal, and for partial discharge signals for which the SVM cannot output a defect type identification result, the target evaluation system further identifies the feature parameters of the partial discharge signal and uses these feature parameters to train the SVM again. This allows the resulting target SVM to more accurately identify the defect type of the partial discharge signal to be identified, thereby improving the accuracy of defect type identification for gas-insulated combined electrical equipment.

[0118] Optionally, update module 504 is also used for:

[0119] If the target defect type is a preset defect type, the support vector machine is trained using the feature parameters of the partial discharge signal to obtain the target support vector machine.

[0120] Optionally, the defect type identification device 500 may also include:

[0121] The first acquisition unit is used to acquire the characteristic parameters of simulated partial discharge sample data of different preset defect types.

[0122] The second determining unit is used to train a preset support vector machine based on the feature parameters of the partial discharge sample data to obtain the support vector machine.

[0123] Optionally, the first acquisition unit is specifically used to normalize the partial discharge sample data to obtain normalized partial discharge sample data; and to standardize the normalized partial discharge sample data to obtain the feature parameters of the partial discharge sample data.

[0124] Optionally, the second determining unit includes:

[0125] The dimension reduction sub-unit is used to reduce the dimension of the feature parameters of the partial discharge sample data to obtain the dimension-reduced feature parameters.

[0126] The training subunit is used to train a pre-defined support vector machine based on the feature parameters after dimensionality reduction, thus obtaining the support vector machine.

[0127] Optionally, a training subunit is used to optimize the parameters of a pre-defined support vector machine based on the radial basis kernel function using a grid search method to obtain the support vector machine.

[0128] Each module in the aforementioned defect type identification device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0129] Figure 6 This is an internal structural diagram of a computer device in an embodiment of this application. In this embodiment, a computer device is provided, which may be a terminal, and its internal structural diagram may be as follows. Figure 6 As shown. The computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a defect type identification method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0130] Those skilled in the art will understand that Figure 6The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0131] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the energy storage system scheduling method provided in the above embodiments. Its implementation principle and technical effects are similar to those of the above method embodiments, and will not be repeated here.

[0132] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the steps of the energy storage system scheduling method provided in the above embodiments. Its implementation principle and technical effects are similar to those of the above method embodiments, and will not be repeated here.

[0133] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the energy storage system scheduling method provided in the above embodiments. Its implementation principle and technical effects are similar to those of the above method embodiments, and will not be repeated here.

[0134] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0135] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0136] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0137] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A defect type identification method characterized by, The method comprises: acquiring a partial discharge signal of a gas insulated combined electric device; identifying the partial discharge signal by using a support vector machine; if an identification result output by the support vector machine is not obtained, uploading feature parameters of the partial discharge signal to a target evaluation system to evaluate the feature parameters of the partial discharge signal by the target evaluation system to obtain a target defect type of the partial discharge signal, wherein the target evaluation system has functions of automatic analysis and fault diagnosis, display of a characteristic map of detection data record, display of a historical trend of feature parameters and provision of data backtracking, operation and diagnosis log, and the target evaluation system supports remote monitoring; if the target defect type is not a preset defect type, training the support vector machine by using the feature parameters of the partial discharge signal to obtain a target support vector machine to identify a defect type of a to-be-identified partial discharge signal by using the target support vector machine.

2. The method of claim 1, wherein, The method further comprises: if the target defect type is the preset defect type, training the support vector machine by using the feature parameters of the partial discharge signal to obtain the target support vector machine.

3. The method of claim 1, wherein, The method further comprises: acquiring feature parameters of simulated partial discharge sample data of different preset defect types; training a preset support vector machine based on the feature parameters of the partial discharge sample data to obtain the support vector machine.

4. The method of claim 3, wherein, The acquiring of the feature parameters of the simulated partial discharge sample data of different preset defect types comprises: performing normalization processing on the partial discharge sample data to obtain normalized partial discharge sample data; performing normalization processing on the normalized partial discharge sample data to obtain the feature parameters of the partial discharge sample data.

5. The method of claim 3, wherein, The training of the preset support vector machine based on the feature parameters of the partial discharge sample data to obtain the support vector machine comprises: performing dimension reduction processing on the feature parameters of the partial discharge sample data to obtain dimension-reduced feature parameters; training the preset support vector machine based on the dimension-reduced feature parameters to obtain the support vector machine.

6. The method of claim 5, wherein, The training of the preset support vector machine based on the dimension-reduced feature parameters to obtain the support vector machine comprises: optimizing parameters of the preset support vector machine by using a grid search method based on a radial basis kernel function to obtain the support vector machine.

7. A defect type identification apparatus characterized by comprising: The device comprises: an acquiring module configured to acquire a partial discharge signal of a gas insulated combined electric device; an identifying module configured to identify the partial discharge signal by using a support vector machine; An evaluation module is configured to, if the recognition result output by the support vector machine is not obtained, upload the characteristic parameters of the partial discharge signal to a target evaluation system to evaluate the characteristic parameters of the partial discharge signal by the target evaluation system to obtain a target defect type of the partial discharge signal, wherein the target evaluation system has functions of automatic analysis and fault diagnosis, display of characteristic maps of detection data records, display of historical trends of characteristic parameters and provision of data backtracking, operation and diagnosis logs, and the target evaluation system supports remote monitoring; An updating module is configured to, if the target defect type is not a preset defect type, train the support vector machine by using the characteristic parameters of the partial discharge signal to obtain a target support vector machine to recognize the defect type of a to-be-recognized partial discharge signal.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

Images