A mountain wind farm lightning stroke fault identification method and system based on zero sequence current and related device
By deeply mining the zero-sequence current characteristics and using a fuzzy logic dynamic weight allocation algorithm, the accuracy and adaptability issues of lightning fault identification in mountainous wind farms were solved, enabling rapid and accurate lightning fault identification and real-time early warning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUANENG POWER INT INC HEBEI CLEAN ENERGY BRANCH
- Filing Date
- 2026-03-25
- Publication Date
- 2026-05-29
Smart Images
Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] This invention belongs to the field of intelligent operation and maintenance and fault diagnosis technology for mountain wind farms, specifically involving a method, system and related device for identifying lightning strike faults in mountain wind farms based on zero-sequence current. Background Technology
[0002] With the large-scale development of mountain wind farms, frequent failures of box-type transformers in the power collection system due to lightning strikes have become a major factor affecting the availability of wind farms. The complex terrain, uneven spatial distribution of soil resistivity, and variable grounding conditions in mountainous areas result in a high degree of uncertainty in the current path of lightning faults, leading to serious limitations in traditional methods for identifying lightning faults. (1) Existing methods rely on complex coupling analysis of voltage traveling waves or multiphase current signals, resulting in redundant characteristic parameters, high computational complexity, and failure to fully utilize the unique time-frequency domain characteristics of zero-sequence current under three-phase unbalanced conditions caused by lightning strikes, leading to insufficient identification accuracy under complex operating conditions.
[0003] (2) Traditional criteria have fixed weights and cannot adapt to environmental differences such as dynamic changes in grounding resistance and seasonal fluctuations in soil moisture in mountain wind farms. They have poor adaptability and high false alarm and false alarm rates.
[0004] (3) Some methods focus on the overvoltage propagation characteristics, neglecting the strong characterization ability and easy acquisition of zero-sequence current signals for fault types, resulting in insufficient real-time fault identification and engineering practicality.
[0005] Therefore, there is an urgent need for an intelligent method for identifying lightning strike faults that is based on zero-sequence current deep mining, can adapt to mountainous environments, has high identification accuracy, and is computationally efficient. Summary of the Invention
[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method, system, and related devices for identifying lightning strike faults in mountainous wind farms based on zero-sequence current. This invention achieves rapid, accurate, and robust identification of three types of faults—side-strike lightning, direct lightning strike, and grounding grid backflashover—by deeply mining the multidimensional characteristics of zero-sequence current in the time and frequency domains and introducing an environment-adaptive intelligent weight allocation algorithm.
[0007] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for identifying lightning strike faults in mountain wind farms based on zero-sequence current, comprising the following steps: Acquire the zero-sequence current signal of the current measurement point set in the box-type transformer under test in the mountain wind farm; The obtained zero-sequence current signal was subjected to time-frequency analysis, and six key feature parameters were extracted. Calculate the membership score of each key feature parameter for the specified fault type; Dynamically calculate the feature weight of each associated feature parameter; Based on the obtained membership scores and feature weights, the lightning fault type of the box-type transformer under test is determined.
[0008] Preferably, the six key characteristic parameters are the zero-sequence current peak value, the zero-sequence current rise steepness, the zero-sequence current harmonic distortion rate, the zero-sequence current decay time constant, the waveform factor, and the ratio of the dominant frequency components.
[0009] Preferably, the membership score of each key feature parameter for a given fault type is calculated, specifically by: The fault types are defined as including lightning strike faults that bypass lightning strikes, lightning strike faults that occur directly, and grounding grid backflash faults. Based on the preset normalized interval mapping function corresponding to each type of fault, the membership score corresponding to each key feature parameter is calculated.
[0010] Preferably, the method for obtaining the preset normalized interval mapping function is as follows: Based on the historical fault data of the transformer under test, the preset normalized interval mapping function corresponding to each type of fault is statistically obtained.
[0011] Preferably, based on the environmental parameters of the mountain wind farm acquired in real time, the feature weight of each associated feature parameter is calculated using the fuzzy logic dynamic method.
[0012] Preferably, based on the obtained membership scores and feature weights, the lightning fault type of the transformer under test is determined. The specific method is as follows: Based on the obtained membership scores and feature weights, calculate the comprehensive confidence score for each type of fault. If the maximum comprehensive confidence score is greater than or equal to the preset threshold, then the fault type corresponding to the maximum comprehensive confidence score is the fault type of the box-type transformer under test. If the maximum overall confidence level is less than the preset threshold, the fault type of the transformer under test needs to be manually confirmed.
[0013] Secondly, the present invention provides a lightning fault identification system for mountain wind farms based on zero-sequence current, comprising: Zero-sequence current signal acquisition unit is used to acquire the zero-sequence current signal of the current measurement point set by the box-type transformer under test in the mountain wind farm; The signal time-frequency analysis unit is used to perform time-frequency analysis on the obtained zero-sequence current signal and extract six key feature parameters; The membership calculation unit is used to calculate the membership score of each key feature parameter for the specified fault type. The feature weight calculation unit is used to dynamically calculate the feature weight of each associated feature parameter; The fault type confirmation unit is used to confirm the lightning fault type of the transformer under test based on the obtained membership score and feature weight.
[0014] Thirdly, the present invention provides an electronic device including a processor and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the method described thereon.
[0015] Fourthly, the present invention provides a computer program product, the computer program product including computer-executable instructions, which, when executed, implement the method described.
[0016] Fifthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the method described herein.
[0017] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a lightning fault identification method for mountain wind farms based on zero-sequence current. By deeply mining the single-channel signal of zero-sequence current, it fundamentally solves the problems of feature redundancy, high computational complexity, and poor engineering practicality caused by existing technologies that rely on multi-phase current or voltage traveling wave analysis. Specifically, this method extracts six key feature parameters—zero-sequence current peak value, rise steepness, harmonic distortion rate, decay time constant, waveform factor, and dominant frequency component ratio—and constructs a highly discriminative feature space from five physical dimensions: energy, transient, spectrum, morphology, and attenuation. This significantly enhances the identification capability for three types of faults: lightning strikes around the ground, direct lightning strikes, and grounding grid backflashover, overcoming the shortcomings of traditional methods such as single feature and low identification accuracy. Simultaneously, this method introduces a fuzzy logic dynamic weight allocation mechanism, which can adaptively adjust the weights of each feature parameter based on real-time acquired environmental parameters such as grounding resistance and soil moisture. This effectively solves the problems of poor adaptability and high false alarm / false negative rates of traditional fixed-weight methods in complex mountain environments, greatly improving the robustness and environmental adaptability of the identification. Furthermore, this method relies solely on zero-sequence current signals, reducing the difficulty of sensor deployment and the system's requirements for multi-channel signal synchronization. It is particularly suitable for mountainous wind farms where wiring is difficult and interference is strong. It has the advantages of simple deployment, low cost, and high real-time performance, ultimately achieving rapid, accurate, and intelligent identification of lightning strike faults, providing reliable technical support for the intelligent operation and maintenance of mountainous wind farms. Attached Figure Description
[0018] Figure 1 This is a flowchart illustrating the lightning strike fault identification process according to an embodiment of the present invention. Figure 2 This is a measured waveform of the zero-sequence current under a lightning strike fault, according to an embodiment of the present invention. Figure 3 This is a measured waveform of zero-sequence current under a direct lightning strike fault according to an embodiment of the present invention. Figure 4 This is a measured waveform of zero-sequence current under a grounding grid backflash fault according to an embodiment of the present invention. Detailed Implementation
[0019] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.
[0020] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.
[0021] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.
[0022] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."
[0023] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0024] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.
[0025] Example 1 This embodiment provides a method for identifying lightning strike faults in mountain wind farms based on zero-sequence current, which includes the following steps: Step 1: Zero-sequence current signal acquisition and preprocessing Current measuring points are arranged on the high-voltage side, low-voltage side, or key nodes of the grounding grid of the box-type transformer. These current measuring points are equipped with broadband current sensors to collect the three-phase current signals of each current measuring point in real time during a lightning strike. , The sampling frequency is not less than 1MHz, and the sampling duration covers the transient process of lightning strike.
[0026] Calculate the zero-sequence current signal corresponding to the current measuring point based on Kirchhoff's current law. ):
[0027] Wavelet packet transform threshold denoising algorithm is used for each zero-sequence current signal Preprocessing effectively removes complex electromagnetic interference and measurement noise from mountainous terrain, resulting in a clean zero-sequence current signal with a high signal-to-noise ratio. .
[0028] Step 2: Extraction of multidimensional feature parameters of zero-sequence current Pure zero-sequence current signal at each current measurement point Joint time-frequency analysis was performed to extract six key feature parameters characterizing the physical nature of lightning strike faults, and a high-dimensional feature vector was constructed: 1. Zero-sequence current peak value : Reflects the intensity of the energy injected by a lightning strike.
[0029]
[0030] 2. Zero-sequence current rise steepness : Characterizes the impact intensity of the transient current of a lightning strike.
[0031]
[0032] In the formula, This is the time it takes for the zero-sequence current to rise from 10% of its peak value to 90% of its peak value.
[0033] 3. Zero-sequence current harmonic distortion rate It reflects the degree of spectral distortion of the current waveform.
[0034]
[0035] In the formula, The amplitude of the zero-sequence current fundamental wave. The amplitude of the nth harmonic of the zero-sequence current (n≥2) reflects the spectral distortion characteristics of the zero-sequence current, and N is the highest harmonic order considered.
[0036] 4. Zero-sequence current decay time constant : Characterizes the rate at which energy dissipates in a faulty loop.
[0037] An exponential fit is performed on the zero-sequence current decay stage after the lightning peak, and the fitting formula is as follows:
[0038] The decay time constant is solved using the least squares method. For the constant term, introduce This is to improve the fitting accuracy, specifically the DC offset or steady-state residual value during the attenuation process. In the actual lightning current attenuation process, due to zero-point drift of the measurement system itself or the presence of a small power frequency component after a fault, the signal may not completely attenuate to zero. Therefore, option C makes the fitting model closer to the actual physical process.
[0039] 5. Waveform factor Quantify the sharpness of the current waveform to distinguish between oscillating and pulse-type faults.
[0040]
[0041] In the formula, This represents the absolute average value of the current during the first complete cycle after the lightning strike.
[0042] 6. Core frequency component ratio : Capture the dominant frequency components in the waveform to distinguish between high-frequency resonance and power frequency dominant faults.
[0043] In the formula, This refers to the frequency component amplitude corresponding to the largest amplitude value in the amplitude spectrum after performing a Fast Fourier Transform on the signal.
[0044] Step 3: Feature parameter normalization and intelligent weight dynamic allocation Based on historical fault data, the normalized interval mapping function corresponding to each type of fault is statistically obtained. For the three types of faults—flashover, direct lightning strike, and grounding grid backflashover—the six key feature parameters extracted in step 2 are normalized to the [0,1] interval to obtain the membership score of each feature parameter for each type of fault. . The feature weights of each associated feature parameter are dynamically calculated using fuzzy logic: 1. Input variables: Real-time monitored mountain environmental parameters, including grounding resistance. (Ω) and soil volumetric water content (%).
[0045] 2. Obfuscation: This refers to the process of blurring the grounding resistance. and soil volumetric water content The data is divided into fuzzy subsets such as "low", "medium", and "high", and a membership function is set for each fuzzy subset.
[0046] 3. Fuzzy Rule Base: Based on expert experience and data-driven approaches, fuzzy rules are established. In this embodiment, the fuzzy rules are: .
[0047] 4. Reasoning and Defuzzification: The Mamdani reasoning method is applied, and the centroid method is used for defuzzification to obtain the dynamic adjustment factors of the weights corresponding to each key feature parameter. .
[0048] 5. Weight Calculation: Combined with the preset initial weights. Calculate the final weights:
[0049] To achieve adaptive adjustment of weights according to the environment, in this embodiment, the preset initial weights are obtained based on the following method: Based on expert experience, in the early stages of a project or when a large amount of historical data is lacking, key characteristic parameters with strong discriminative power (such as the zero-sequence current peak value I_0m and the rise steepness k_0) can be assigned higher initial weights, while features with relatively weaker discriminative power can be assigned lower initial weights, based on domain knowledge of power system fault diagnosis. This reflects the practicality and flexibility of the method.
[0050] Based on historical data-driven presets: With sufficient historical fault data, statistical analysis (such as using machine learning algorithms like random forests and linear discriminant analysis) can be used to calculate the average importance of each feature parameter to fault classification, and the normalized result is used as the initial weight. This method makes the weight setting more objective and scientific.
[0051] Step 4: Intelligent identification of lightning strike fault type Calculate the overall confidence score for the three types of faults:
[0052] Set the identification threshold T=0.5. If the score for a certain type of fault... S Fault i Highest and satisfy S If the fault i ≥ T, it is determined to be this type of fault; if the highest score is lower than T, it is determined to be an uncertain or atypical lightning strike fault, and the manual review process is initiated.
[0053] The beneficial effects of this invention are: 1. Enhanced feature dimension and high discriminative power: Six key features are extracted from five physical dimensions, namely energy, transient state, spectrum, morphology and decay, to construct a highly discriminative feature space. This makes the three types of faults linearly separable in the feature space, fundamentally solving the problem of insufficient recognition accuracy of single features or shallow features.
[0054] 2. The algorithm is intelligent and adaptive with strong robustness: It adopts a fuzzy logic system to replace the fixed formula for dynamic weight allocation, which enables the algorithm to simulate expert experience and flexibly adapt to the continuous changes in mountain environmental parameters, significantly improving the engineering applicability and robustness of the method.
[0055] 3. Empirically driven and highly practical: The method is based on the analysis and feature extraction of a large amount of measured waveform data. The identification logic is clear and the physical meaning is well-defined. The high accuracy of the method in complex real-world scenarios has been verified through examples. It is easy to integrate into the smart operation and maintenance platform of wind farms to achieve minute-level accurate fault location and early warning.
[0056] 4. Single-signal solution, easy to deploy: Relying only on a single zero-sequence current signal, it reduces the system's requirements for multi-sensor synchronization and reliability, making it particularly suitable for scenarios where wiring is difficult in mountainous wind farms, thus reducing construction and maintenance costs.
[0057] Example 2 This embodiment provides a method for identifying lightning strike faults in mountain wind farms based on zero-sequence current, which includes the following steps: Step 1: Setting simulation model parameters Box-type transformer: rated voltage 0.69 / 37kV, capacity 2.25MVA, high voltage winding capacitance to ground 800pF, low voltage winding capacitance to ground 1500pF; Collection line: 35kV double-circuit overhead line, conductor type LGJ-240 / 30, tower height 45m, grounding resistance ( R g =8 Omega), soil moisture (w=25%); Lightning current parameters: amplitude 100kA, wavefront time 2.6μs, half peak time 50μs (compliant with IEC standards).
[0058] Step 2: Zero-sequence current signal acquisition and preprocessing Three-phase current signals under three types of faults are collected by current sensors. The zero-sequence current is calculated and then denoised using wavelet thresholding to obtain a clean zero-sequence current waveform. Step 3: Fault identification verification based on measured waveforms To verify the effectiveness and practicality of the method of this invention, typical waveforms of lightning strikes, direct lightning strikes, and grounding grid backflash faults were selected from historical lightning fault recording data of a mountain wind farm (e.g., [examples of waveforms]). Figures 2 to 4 An empirical analysis was conducted (as shown in the figure).
[0059] 1. Feature parameter extraction results Preprocessing and feature extraction were performed on the three sets of measured waveforms, and the key feature parameters are shown in the table below:
[0060] 2. Intelligent Recognition Process Normalization: Input the data in the table above into a normalization function calibrated based on a large amount of measured data to obtain the membership scores Sj of each feature to the three types of faults.
[0061] Weight allocation: Assume the current environment parameters are R g =15Ω, w=30%, input to fuzzy logic system, calculate dynamic weight vector ω=[0.28,0.26,0.18,0.12,0.09,0.07].
[0062] Comprehensive assessment: Calculate the total score for all types of faults.
[0063] 3. Recognition Results Waveform of lightning strike around the target: S_strike = 0.82 (maximum), correctly identified.
[0064] Direct lightning strike waveform: S_direct_strike = 0.78 (maximum), correctly identified.
[0065] Grounding grid backflash waveform: S backflash = 0.85 (maximum), correctly identified.
[0066] This embodiment demonstrates that the method of the present invention has an extremely high accuracy rate in identifying actual fault waveforms, verifying its strong engineering practical value.
[0067] Example 3 This embodiment provides a lightning strike fault identification system for mountain wind farms based on zero-sequence current, including: Zero-sequence current signal acquisition unit is used to acquire the zero-sequence current signal of the current measurement point set by the box-type transformer under test in the mountain wind farm; The signal time-frequency analysis unit is used to perform time-frequency analysis on the obtained zero-sequence current signal and extract six key feature parameters; The membership calculation unit is used to calculate the membership score of each key feature parameter for the specified fault type. The feature weight calculation unit is used to dynamically calculate the feature weight of each associated feature parameter; The fault type confirmation unit is used to confirm the lightning fault type of the transformer under test based on the obtained membership score and feature weight.
[0068] Example 4 This embodiment also provides a computing device. The computing device includes a bus, a processor, a memory, and a communication interface. The processor, memory, and communication interface communicate with each other via the bus. The computing device can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memory in the computing device.
[0069] A bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, a bus can include a path for transmitting information between various components of a computing device (e.g., memory, processor, communication interfaces).
[0070] The processor may include any one or more of the following: central processing unit (CPU), graphics processing unit (GPU), tensor processing unit (TPU), application specific integrated circuit (ASIC), field-programmable gate array (FPGA), microprocessor (MP), or digital signal processor (DSP).
[0071] Memory can include volatile memory, such as random access memory (RAM). Processors can also include non-volatile memory. volatile memory, such as read-only memory (ROM). ROM (memory only), flash memory, hard disk drive (HDD), or solid state drive (SSD).
[0072] The memory stores executable program code, which the processor executes to implement the functions of the aforementioned units, thereby achieving, for example, the method described in Embodiment 1. That is, the memory may store instructions for the methods and functions relating to the computing device in any of the above embodiments.
[0073] The communication interface uses transceiver modules such as, but not limited to, network interface cards and transceivers to enable communication between computing devices and other devices or communication networks.
[0074] Example 5 This embodiment also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, cause the processor to perform the methods and functions of the computing device involved in any of the above embodiments.
[0075] Generally, the various embodiments of this disclosure can be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects can be implemented in hardware, while others can be implemented in firmware or software, which can be executed by a controller, microprocessor, or other computing device. Although various aspects of the embodiments of this disclosure are shown and described as block diagrams, flowcharts, or represented using some other illustration, it should be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as, as non-limiting examples, in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.
[0076] Example 6 This embodiment provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods as described above with reference to the accompanying drawings. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.
[0077] Computer program code used to implement the methods of this disclosure may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the computer or other programmable data processing apparatus, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be performed. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.
[0078] In the context of this disclosure, computer program code or related data may be carried on any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and so on. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.
[0079] Computer-readable media can be any tangible medium that contains or stores programs for or relating to an instruction execution system, apparatus, or device, or a data storage device such as a data center containing one or more available media. Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. More detailed examples of computer-readable storage media include electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical storage devices, magnetic storage devices, or any suitable combination thereof.
[0080] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A method for identifying lightning strike faults in mountain wind farms based on zero-sequence current, characterized in that, Includes the following steps: Acquire the zero-sequence current signal of the current measurement point set in the box-type transformer under test in the mountain wind farm; The obtained zero-sequence current signal was subjected to time-frequency analysis, and six key feature parameters were extracted. Calculate the membership score of each key feature parameter for the specified fault type; Dynamically calculate the feature weight of each associated feature parameter; Based on the obtained membership scores and feature weights, the lightning fault type of the box-type transformer under test is determined.
2. The method for identifying lightning faults in mountain wind farms based on zero-sequence current according to claim 1, characterized in that, The six key characteristic parameters are the peak value of zero-sequence current, the steepness of the rise of zero-sequence current, the harmonic distortion rate of zero-sequence current, the decay time constant of zero-sequence current, the waveform factor, and the ratio of the dominant frequency components.
3. The method for identifying lightning faults in mountain wind farms based on zero-sequence current according to claim 1, characterized in that, The membership score for each key feature parameter corresponding to a given fault type is calculated using the following method: The fault types are defined as including lightning strike faults that bypass lightning strikes, lightning strike faults that occur directly, and grounding grid backflash faults. Based on the preset normalized interval mapping function corresponding to each type of fault, the membership score corresponding to each key feature parameter is calculated.
4. The method for identifying lightning faults in mountain wind farms based on zero-sequence current according to claim 3, characterized in that, The method for obtaining the preset normalized interval mapping function is as follows: Based on the historical fault data of the transformer under test, the preset normalized interval mapping function corresponding to each type of fault is statistically obtained.
5. The method for identifying lightning faults in mountain wind farms based on zero-sequence current according to claim 1, characterized in that, Based on the environmental parameters of the mountain wind farm acquired in real time, the feature weight of each associated feature parameter is calculated using the fuzzy logic dynamic method.
6. The method for identifying lightning faults in mountain wind farms based on zero-sequence current according to claim 1, characterized in that, Based on the obtained membership scores and feature weights, the lightning fault type of the transformer under test is determined. The specific method is as follows: Based on the obtained membership scores and feature weights, calculate the comprehensive confidence score for each type of fault. If the maximum comprehensive confidence score is greater than or equal to the preset threshold, then the fault type corresponding to the maximum comprehensive confidence score is the fault type of the box-type transformer under test. If the maximum overall confidence level is less than the preset threshold, the fault type of the transformer under test needs to be manually confirmed.
7. A lightning fault identification system for mountain wind farms based on zero-sequence current, characterized in that, include: Zero-sequence current signal acquisition unit is used to acquire the zero-sequence current signal of the current measurement point set by the box-type transformer under test in the mountain wind farm; The signal time-frequency analysis unit is used to perform time-frequency analysis on the obtained zero-sequence current signal and extract six key feature parameters; The membership calculation unit is used to calculate the membership score of each key feature parameter for the specified fault type. The feature weight calculation unit is used to dynamically calculate the feature weight of each associated feature parameter; The fault type confirmation unit is used to confirm the lightning fault type of the transformer under test based on the obtained membership score and feature weight.
8. An electronic device, characterized in that, It includes a processor and a memory, the memory storing computer instructions that, when executed by the processor, cause the electronic device to perform the method of any one of claims 1 to 6.
9. A computer program product, characterized in that, The computer program product includes computer-executable instructions that, when executed, implement the method of any one of claims 1 to 6.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when executed by a processor, implement the method of any one of claims 1 to 6.