Fuzzy neural network-based distribution line fault prediction method and system

Abstract

The invention relates to the field of power distribution network fault prediction, and discloses a power distribution line fault prediction method and system based on a fuzzy neural network, and the method comprises the steps: obtaining power distribution network line data; preprocessing the power distribution network line data, and introducing a clustering algorithm to classify the data; using an adaptive fuzzy neural network as a basic model to analyze the classified data, wherein the adaptive fuzzy neural network comprises an input layer, a membership function layer, a rule layer, a normalization layer and an output layer; analyzing a fault mechanism of the target equipment, determining a fault rate distribution function type, selecting a membership function, applying the selected membership function to a membership function layer, and performing parameter setting; and testing and evaluating the basic model, integrating the basic model to a power grid monitoring system, and predicting a distribution line fault in real time. The method can be widely applied to the field of fault prediction and health management of industrial equipment, and lays a foundation for improving the operation and maintenance management efficiency of the power distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network fault prediction, and particularly to a distribution line fault prediction method and system based on a fuzzy neural network. Background Art

[0002] As the end link of the power system, the distribution network is directly connected to users, and its operating state has a direct impact on the power consumption quality of users. Users have extremely high requirements for the reliability of power supply. Any power outage event may affect the user experience and cause adverse social impacts. According to statistics, most user power outages are caused by distribution network faults. Therefore, effective prediction of distribution network faults can perform operation and maintenance in advance, reduce power outage events, and improve power supply reliability.

[0003] In modern industrial systems, equipment fault prediction technology is crucial for ensuring the stable operation of equipment and production efficiency. Existing fault prediction technologies are mainly divided into two types: model-based and data-driven, but both have deficiencies. Model-based technologies need to deeply understand the fault mechanism and are difficult to handle the diverse fault modes of complex systems; although data-driven technologies get rid of the dependence on the fault mechanism, they have high requirements for data quality, are easily affected by noise, and the prediction accuracy is limited. The proportion of cable lines in the distribution network is increasing day by day, and they are buried underground, making it difficult to detect problems. Once a fault occurs, it has a wide impact. The distribution network has a complex structure, a large scale, and numerous equipment. It is difficult to comprehensively monitor with limited resources, resulting in maintenance of some lines only after a fault occurs. Therefore, how to effectively monitor and predict all line faults has become a major problem. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a distribution line fault prediction method and system based on a fuzzy neural network to solve the problem of how to effectively monitor and predict all line faults including buried cables in a complex and large-scale distribution network, so as to perform operation and maintenance in advance, reduce power outage events, and improve power supply reliability, while overcoming the limitations of existing fault prediction technologies, such as strong dependence on the fault mechanism, high requirements for data quality, and susceptibility to noise.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for predicting faults in a distribution line based on a fuzzy neural network, including: acquiring distribution network line data; preprocessing the distribution network line data and introducing a clustering algorithm to classify the data; using an adaptive fuzzy neural network as a basic model to analyze the classified data, where the adaptive fuzzy neural network includes an input layer, a membership function layer, a rule layer, a normalization layer, and an output layer; analyzing the fault mechanism of the target device, determining the type of failure rate distribution function and selecting a membership function, applying the selected membership function to the membership function layer, and performing parameter setting; testing and evaluating the basic model, and integrating the basic model into the power grid monitoring system to predict faults in the distribution line in real time.

[0008] As a preferred embodiment of the method for predicting faults in a distribution line based on a fuzzy neural network according to the present invention, wherein: the acquisition of the distribution network line fault data includes fault time data, fault characteristic data, equipment status monitoring data, and environmental information.

[0009] As a preferred embodiment of the method for predicting faults in a distribution line based on a fuzzy neural network according to the present invention, wherein: the preprocessing includes: data cleaning, data dimensionality reduction, and data standardization;

[0010] In the data preprocessing stage, a clustering algorithm is introduced to classify the fault data, group the fault samples with similar distribution characteristics, and optimize a set of membership functions for each type of distribution separately to accurately match its characteristics.

[0011] As a preferred embodiment of the method for predicting faults in a distribution line based on a fuzzy neural network according to the present invention, wherein: the adaptive fuzzy neural network includes an input layer, a membership function layer, a rule layer, a normalization layer, and an output layer;

[0012] The input layer receives the classified equipment status monitoring data;

[0013] The membership function layer maps the input data to fuzzy membership degrees and processes it using a preset set of membership functions; the set of membership functions has a total of n×r nodes, where n is the number of nodes in the input layer and r is the number of nodes in the regular layer, and each node represents a certain membership function, which is expressed by the formula:

[0014]

[0015] where, u ij (t), c ij (t) and respectively represent the output value, expected value, and variance of the input value x i (t) on the jth membership function;

[0016] The rule layer performs fuzzy inference based on the membership degree and outputs according to the excitation intensity of the operator on the fuzzy rule space. Among them, the calculation of the activation intensity of the i-th rule is as follows:

[0017]

[0018] Among them, ω i represents the activation intensity of the i-th rule, m ij represents the membership degree of the j-th input variable in the i-th rule, m represents the total number of rules, and n represents the total number of input variables;

[0019] The normalization layer normalizes the output of the rule layer. Let O i represent the output of the i-th rule, and O total represent the sum of the outputs of all rules. Then the output N i of the normalization layer is expressed as:

[0020]

[0021] The output layer calculates the final output. This layer is an output layer with a single node and is used to calculate the sum of all input signals.

[0022] As a preferred scheme of the distribution line fault prediction method based on the fuzzy neural network according to the present invention, among them: the selected membership function includes:

[0023] Set the number of iterations and the minimum allowable error as the termination conditions for the training of the adaptive fuzzy neural network to adjust the prediction accuracy and running time of the model;

[0024] Preset a set of membership function sets MF = {MF1, MF2, ···, MF i}, where MF i represents the preset membership function, and initialize the minimum error of the current loop to infinity;

[0025] Randomly select a membership function MF i from the membership function set to construct the adaptive fuzzy neural network, and pre-train the network. After training, save the prediction error E% of the current loop and compare it with the minimum error IterErro of the current loop;

[0026] If the prediction error E% of the current loop is less than the minimum error MinError, update the minimum error value and the corresponding membership function information, otherwise do not update the IterError value and membership function information of the adaptive fuzzy neural network;

[0027] Delete the membership function MFi used in this round, and check whether the set of membership functions is empty. If the set is not empty, continue the loop for selection and construction; if the set is empty, it means that all preset membership functions have been used to construct the adaptive fuzzy neural network.

[0028] As a preferred solution of the distribution line fault prediction method based on a fuzzy neural network according to the present invention, wherein: testing and evaluating the basic model includes:

[0029] Input the test data set into the trained adaptive fuzzy neural network model, and calculate the error between its output result and the actual fault data;

[0030] Use an error index to evaluate the prediction accuracy of the model, and perform statistical analysis on the evaluation results.

[0031] As a preferred solution of the distribution line fault prediction method based on a fuzzy neural network according to the present invention, wherein: integrate the adaptive fuzzy neural network model into the power grid monitoring system, and continuously adjust and optimize the model according to the actual operation situation to predict the distribution line fault in real time.

[0032] In a second aspect, the present invention provides a distribution line fault prediction system based on a fuzzy neural network, including:

[0033] A data acquisition module for acquiring distribution network line data;

[0034] A data processing module for preprocessing the distribution network line data and introducing a clustering algorithm to classify the data;

[0035] A model analysis module for analyzing the classified data by using an adaptive fuzzy neural network as a basic model, the adaptive fuzzy neural network including an input layer, a membership function layer, a rule layer, a normalization layer and an output layer; analyzing the fault mechanism of the target device, determining the type of failure rate distribution function and selecting a membership function, applying the selected membership function to the membership function layer, and performing parameter setting;

[0036] An evaluation and prediction module for testing and evaluating the basic model and integrating the basic model into the power grid monitoring system to predict the distribution line fault in real time.

[0037] In a third aspect, the present invention provides an electronic device, including:

[0038] A memory and a processor;

[0039] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the distribution line fault prediction method based on the fuzzy neural network are realized.

[0040] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps of the distribution line fault prediction method based on the fuzzy neural network are realized.

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows: By introducing an adaptive fuzzy neural network combined with an optimized membership function, the present invention provides a distribution line fault prediction method and system based on a fuzzy neural network, which not only integrates the advantages of high accuracy of fault prediction technology but also inherits the characteristics of data-driven methods that do not require building complex mathematical models and have strong adaptability. By preprocessing and classifying the distribution network line data and using the powerful self-learning ability of the neural network, the potential information in the fault data is fully mined to achieve efficient and accurate prediction of distribution line faults. In addition, this model has high accuracy, fast convergence, and wide applicability, can effectively overcome the problems of strong dependence on fault mechanisms and susceptibility to noise of traditional prediction technologies, significantly improve the reliability and operation and maintenance efficiency of the power supply system, is applicable to the field of fault prediction and health management of industrial equipment, and provides strong technical support for the stable operation of the power system. Description of the Drawings

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0043] Figure 1 It is a schematic diagram of the overall process logic of the distribution line fault prediction method based on the fuzzy neural network according to an embodiment of the present invention;

[0044] Figure 2 It is a flowchart of an adaptive fuzzy neural network with the optimal membership function for the distribution line fault prediction method according to an embodiment of the present invention. Detailed Embodiments

[0045] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0046] Embodiment 1

[0047] Referring to Figure 1 - Figure 2 For an embodiment of the present invention, a method for predicting distribution line faults based on a fuzzy neural network is provided, which provides a fault prediction model with high precision, high adaptability, and fast convergence. By optimizing the membership function, efficient fitting of different data distributions is achieved, thereby improving the reliability and practicality of fault prediction, laying a foundation for improving the operation and maintenance management efficiency of the distribution network, and effectively enhancing the operation and maintenance management level of the current distribution network under disaster meteorological conditions. As Figure 1 shown, it specifically includes the following steps:

[0048] S100: Obtain distribution network line data;

[0049] S200: Preprocess the distribution network line data and introduce a clustering algorithm to classify the data;

[0050] S300: Use an adaptive fuzzy neural network as the basic model to analyze the classified data, where the adaptive fuzzy neural network includes an input layer, a membership function layer, a rule layer, a normalization layer, and an output layer;

[0051] S400: Analyze the fault mechanism of the target device, determine the type of failure rate distribution function and select a membership function, apply the selected membership function to the membership function layer, and perform parameter settings;

[0052] S500: Test and evaluate the basic model and integrate the basic model into the power grid monitoring system to predict distribution line faults in real time.

[0053] It should be noted that by introducing an adaptive fuzzy neural network combined with an optimized membership function, the present invention provides a method and system for predicting distribution line faults based on a fuzzy neural network, which not only integrates the advantages of high-precision fault prediction technology but also inherits the characteristics of data-driven methods that do not require the construction of complex mathematical models and have strong adaptability. By preprocessing and classifying the distribution network line data and utilizing the powerful self-learning ability of the neural network, the potential information in the fault data is fully mined to achieve efficient and accurate prediction of distribution line faults. In addition, this model has high precision, fast convergence, and wide applicability, can effectively overcome the problems of strong dependence on fault mechanisms and susceptibility to noise in traditional prediction technologies, significantly improve the reliability and operation and maintenance efficiency of the power supply system, is applicable to the field of fault prediction and health management of industrial equipment, and provides strong technical support for the stable operation of the power system.

[0054] In the embodiment of the present application, in the above step S100, the distribution network line data is acquired, and the distribution network line fault data includes fault time data, fault feature data, equipment status monitoring data, and environmental information.

[0055] In the embodiment of the present application, in the above step S200, the distribution network line data is preprocessed, and a clustering algorithm is introduced to classify the data; the preprocessing includes data cleaning, data dimensionality reduction, and data normalization;

[0056] In the data preprocessing stage, a clustering algorithm is introduced to classify the fault data, and the fault samples with similar distribution characteristics are grouped. A set of membership functions is optimized separately for each class of distribution to accurately match its characteristics.

[0057] It should be noted that the above step S200 can effectively clean and standardize the original data, remove noise and outliers, and improve the data quality. The application of the clustering algorithm enables automatic grouping of similar data points according to the inherent characteristics of the data, thereby identifying line segments or equipment categories with different characteristics, providing a cleaner and more targeted input for the adaptive fuzzy neural network, and further improving the learning efficiency and prediction accuracy of the model. In addition, by pre-classifying the data reasonably in advance, customized analysis strategies can be adopted for different types of fault modes, enhancing the adaptability and flexibility of the entire prediction system.

[0058] In the embodiment of the present application, in the above step S300, the adaptive fuzzy neural network includes an input layer, a membership function layer, a rule layer, a normalization layer, and an output layer;

[0059] Specifically, the input layer receives the classified equipment status monitoring data;

[0060] Specifically, the membership function layer maps the input data to fuzzy membership degrees and processes it using a preset set of membership functions. Among them, the set of membership functions has a total of n×r nodes, where n is the number of nodes in the input layer and r is the number of nodes in the regular layer. Each node represents a certain membership function. In this embodiment, the membership function is taken as a Gaussian function, and the formula is expressed as:

[0061]

[0062] where u ij (t), c ij (t) and respectively represent the output value, expectation, and variance of the input value x i (t) on the j-th membership function;

[0063] Specifically, the rule layer performs fuzzy reasoning based on the membership degrees and outputs according to the operator's excitation intensity for the fuzzy rule space. Among them, the calculation of the activation intensity of the i-th rule is:

[0064]

[0065] where ω i represents the activation intensity of the i-th rule, m ij represents the membership degree of the j-th input variable in the i-th rule, m represents the total number of rules, and n represents the total number of input variables;

[0066] Specifically, the normalization layer normalizes the output of the rule layer. The output of the rule layer is the unnormalized intensity of each rule, and they need to be converted into probability form so that the correct result can be obtained when performing weighted averaging in the output layer. Let O i represent the output of the i-th rule, and O total represent the sum of the outputs of all rules. Then the output N i of the normalization layer is expressed as:

[0067]

[0068] Specifically, the output layer calculates the final output. This layer is an output layer with a single node and is used to calculate the sum of all input signals.

[0069] It should be noted that in the above step S300, the adaptive fuzzy neural network (AFNN) is used as the basic model to analyze the classified data. Combining the advantages of fuzzy logic and neural networks, it can not only process uncertain and fuzzy information, but also automatically adjust parameters through learning algorithms to optimize performance. The input layer, membership function layer, rule layer, normalization layer, and output layer included in AFNN enable it to simulate complex non-linear relationships and accurately capture the characteristic patterns of distribution network line faults. In addition, the strong self-learning ability of AFNN can extract valuable information from a large amount of historical data, continuously improve the prediction accuracy, and quickly adapt to new fault patterns, thus providing strong technical support for real-time fault prediction and significantly improving the reliability and maintenance efficiency of the power system.

[0070] In the embodiment of the present application, as Figure 2 shown, the above step S400 includes:

[0071] Set the number of iterations (NumIter) and the minimum allowable error (MinError) as the termination conditions for the training of the adaptive fuzzy neural network to adjust the prediction accuracy and running time of the model;

[0072] Preset a set of membership function sets MF = {MF1, MF2, ···, MFi}, where MFi represents the preset membership function, and initialize the minimum error of the current loop to infinity, so as to compare and select the best membership function in the subsequent steps;

[0073] Exemplarily, select five of them to establish the membership function set MF = {guassmf, gbellmf, sigmf, psigmf, dsigmf}, which are the Gaussian membership function (gaussmf), the generalized bell-shaped membership function (gbellmf), the S-shaped membership function (sigmf), the double S-shaped product membership function (psigmf), and the double S-shaped difference membership function (dsigmf). Preset NumIter to 20 and MinError to 0.001.

[0074] Randomly select a membership function MFi from the membership function set to construct an adaptive fuzzy neural network, and pre-train the network. After training, save the prediction error E% of the current loop and compare it with the minimum error IterErro of the current loop;

[0075] If the prediction error E% of the current loop is less than the minimum error MinError, update the minimum error value and the corresponding membership function information, otherwise do not update the IterError value and membership function information of the adaptive fuzzy neural network;

[0076] Delete the membership function MFi used in this round and check whether the set of membership functions is empty. If the set is not empty, continue the loop for selection and construction; if the set is empty, it means that all preset membership functions have been used to construct the adaptive fuzzy neural network.

[0077] It should be noted that through in-depth analysis of the fault mechanism of the target device and determination of the type of failure rate distribution function in the above step S400, the most suitable membership function can be selected and applied to the membership function layer of the adaptive fuzzy neural network (AFNN), so as to perform accurate parameter setting. This process can more accurately reflect the actual fault behavior and probability characteristics, enhancing the pertinence and reliability of the prediction model. The analysis based on the fault mechanism not only reduces the blind dependence of the model on data, but also improves its generalization ability under different working conditions; and the reasonable selection of membership functions further optimizes the fuzzy reasoning process, enabling the model to capture early fault signals more sensitively and accurately, warning of potential problems in advance, and effectively improving the preventive maintenance level and operation safety of the power system.

[0078] In the embodiment of the present application, the above step S500 includes:

[0079] Specifically, the testing and evaluation of the basic model include: inputting the test data set into the trained adaptive fuzzy neural network model, calculating the error between its output result and the actual fault data; using the error index to evaluate the prediction accuracy of the model, and conducting statistical analysis on the evaluation results.

[0080] It should be noted that for the fault prediction of complex equipment, for complex equipment affected by multiple failure distribution functions, a hybrid membership function set integrating multiple failure distribution functions is established to improve the adaptability of the prediction model to the faults of complex equipment.

[0081] Specifically, integrate the adaptive fuzzy neural network model into the power grid monitoring system, and continuously adjust and optimize the model according to the actual operation situation to predict the faults of distribution lines in real time.

[0082] It should be noted that the above step S500 ensures the reliability and stability of the model in practical applications, can timely detect and warn of potential faults, thus significantly reducing the occurrence of sudden power outage events. The real-time prediction function enables the power company to plan maintenance and repair in advance, optimize resource allocation, and improve the response speed and service quality. In addition, embedding the advanced fault prediction model into the existing power grid monitoring system not only enhances the intelligent level and self-diagnosis ability of the system, but also provides strong technical support for the automated management and efficient operation of the smart grid, significantly improving the power supply reliability of the distribution network and user satisfaction.

[0083] Embodiment 2

[0084] In this embodiment, a distribution line fault prediction system based on a fuzzy neural network is provided, including:

[0085] A data acquisition module for acquiring distribution network line data;

[0086] A data processing module for preprocessing the distribution network line data and introducing a clustering algorithm to classify the data;

[0087] A model analysis module for analyzing the classified data using an adaptive fuzzy neural network as the basic model, where the adaptive fuzzy neural network includes an input layer, a membership function layer, a rule layer, a normalization layer, and an output layer; analyzing the fault mechanism of the target device, determining the type of failure rate distribution function and selecting a membership function, applying the selected membership function to the membership function layer, and performing parameter settings;

[0088] An evaluation and prediction module for testing and evaluating the basic model and integrating the basic model into the power grid monitoring system to predict distribution line faults in real time.

[0089] It should be noted that the technical solution of the system for predicting distribution line faults based on a fuzzy neural network belongs to the same concept as the technical solution of the method for predicting distribution line faults based on a fuzzy neural network described above. For the details not described in detail in the technical solution of the system for predicting distribution line faults based on a fuzzy neural network in this embodiment, reference can be made to the description of the technical solution of the method for predicting distribution line faults based on a fuzzy neural network described above.

[0090] The above-mentioned unit modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0091] This embodiment also provides an electronic device, which includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an 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 medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for predicting distribution line faults based on a fuzzy neural network. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0092] This embodiment also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the method proposed in the above embodiment.

[0093] The storage medium proposed in this embodiment and the method proposed in the above embodiment belong to the same inventive concept. For technical details not described in detail in this embodiment, reference can be made to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0094] Through the above description of the implementation manners, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general hardware. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such an understanding, the technical solution of the present invention, in essence, or the part that makes contributions to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk, or optical disc of a computer, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the method of the embodiment of the present invention.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

[0096] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages.

[0097] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0098] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0099] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0100] Although the preferred embodiments of the present application have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn of the basic creative concept. Therefore, the appended claims are intended to be interpreted to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present application.

[0101] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A distribution line fault prediction method based on a fuzzy neural network, characterized in that, Including: Obtain the distribution network line data; Preprocess the distribution network line data and introduce a clustering algorithm to classify the data; Use an adaptive fuzzy neural network as the basic model to analyze the classified data. The adaptive fuzzy neural network includes an input layer, a membership function layer, a rule layer, a normalization layer, and an output layer; Analyze the failure mechanism of the target device, determine the type of failure rate distribution function and select the membership function, apply the selected membership function to the membership function layer, and perform parameter settings; Test and evaluate the basic model, and integrate the basic model into the power grid monitoring system to predict the distribution line faults in real time.

2. The fault prediction method for distribution lines based on a fuzzy neural network according to claim 1, characterized in that The obtaining of the distribution network line fault data includes fault time data, fault feature data, equipment status monitoring data, and environmental information.

3. The method for predicting faults in a distribution line based on a fuzzy neural network according to claim 2, wherein, The preprocessing includes: data cleaning, data dimensionality reduction, and data standardization; In the data preprocessing stage, introduce a clustering algorithm to classify the fault data, group the fault samples with similar distribution characteristics, and optimize a set of membership functions for each class of distribution to match its characteristics.

4. The method for predicting distribution line faults based on a fuzzy neural network according to claim 3, wherein, The adaptive fuzzy neural network includes an input layer, a membership function layer, a rule layer, a normalization layer, and an output layer; The input layer receives the classified equipment status monitoring data; The membership function layer maps the input data to fuzzy membership degrees and processes it using a preset set of membership functions. The set of membership functions has a total of n×r nodes, where n is the number of nodes in the input layer and r is the number of nodes in the regular layer. Each node represents a certain membership function, and the formula is expressed as: where, u ij (t), c ij (t) and represent the output value, the expectation, and the variance of the input value x i (t) on the j-th membership function, respectively; The rule layer performs fuzzy reasoning based on the membership degrees and outputs the excitation intensity of the fuzzy rule space according to the operator. Among them, the calculation of the activation intensity of the i-th rule is: where ω i represents the activation strength of the i-th rule, and m ij represents the membership degree of the j-th input variable in the i-th rule. m represents the total number of rules, and n represents the total number of input variables; The normalization layer normalizes the output of the rule layer, where O i represents the output of the i-th rule, and O total represents the sum of the outputs of all rules. Then the output N i of the normalization layer is expressed as: The output layer calculates the final output. This layer is an output layer with a single node and is used to calculate the sum of all input signals.

5. The fault prediction method for distribution lines based on a fuzzy neural network according to claim 4, characterized in that The selection of the membership function includes: Set the number of iterations and the minimum allowable error as the termination condition for the training of the adaptive fuzzy neural network to adjust the prediction accuracy and running time of the model; Preset a set of membership function sets MF = {MF1, MF2, ···, MF i}, where MF i represents the preset membership function, and initialize the minimum error of the current loop to infinity; Randomly select a membership function MF from the set of membership functions i Construct the adaptive fuzzy neural network and pre-train the network. After training, save the prediction error E% of the current cycle and compare it with the minimum error IterErro of the current cycle; If the prediction error E% of the current cycle is less than the minimum error MinError, update the minimum error value and the corresponding membership function information, otherwise do not update the IterError value and membership function information of the adaptive fuzzy neural network; Delete the membership function MFi used in this round and check whether the set of membership functions is empty. If the set is not empty, continue the loop selection and construction process; if the set is empty, it means that all preset membership functions have been used to construct the adaptive fuzzy neural network.

6. The fault prediction method for distribution lines based on a fuzzy neural network according to claim 5, wherein, The test and evaluation of the basic model include: Input the test data set into the trained adaptive fuzzy neural network model, and calculate the error between its output result and the actual fault data; Use the error index to evaluate the prediction accuracy of the model, and perform statistical analysis on the evaluation results.

7. The fault prediction method for distribution lines based on a fuzzy neural network according to claim 6, characterized in that, Integrate the adaptive fuzzy neural network model into the power grid monitoring system, and continuously adjust and optimize the model according to the actual operation situation to predict the distribution line faults in real time.

8. A system applying the distribution line fault prediction method based on a fuzzy neural network as described in any one of claims 1 to 7, characterized in that, Including: A data acquisition module for obtaining distribution network line data; A data processing module, which is used to preprocess the distribution network line data and introduce a clustering algorithm to classify the data; A model analysis module, which is used to analyze the classified data by using an adaptive fuzzy neural network as a basic model. The adaptive fuzzy neural network includes an input layer, a membership function layer, a rule layer, a normalization layer and an output layer; analyze the failure mechanism of the target device, determine the type of the failure rate distribution function and select a membership function, apply the selected membership function to the membership function layer, and perform parameter settings; An evaluation and prediction module, which is used to test and evaluate the basic model and integrate the basic model into the power grid monitoring system to predict the distribution line faults in real time.

9. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method described in claims 1 to 7 are implemented.

10. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that: When the computer-executable instructions are executed by the processor, the steps of the method described in claims 1 to 7 are implemented.

Images