A method for locating traveling waves in multi-branch lines based on fault ratio.

Abstract

This invention discloses a method for locating traveling wave faults in multi-branch lines based on fault ratio. The method includes: establishing a fault location algorithm based on the inherent length of the line and the reflection of the traveling wave; establishing fault location determination rules for different fault locations in multi-branch lines based on the fault location algorithm and utilizing the numerical relationship between the distance from each end to the fault point and the length of the line branch; collecting fault voltage traveling wave data; processing the fault voltage traveling wave data based on a trained neural network model to obtain the arrival time of the traveling wave front reflecting fault distance information; and using the fault location determination rules to locate the fault based on the arrival time of the traveling wave front reflecting fault distance information. This method enables accurate location of faults in distribution network lines.

Description

Technical Field

[0001] This invention relates to the field of power distribution network fault identification technology, and in particular to a method for locating traveling waves in multi-branch lines based on fault ratio. Background Technology

[0002] As a key link in power transmission, the accurate location of line faults in the distribution network is of great significance for achieving rapid repair of faulty lines and improving the reliability of power supply in the distribution network.

[0003] Distribution networks differ significantly from traditional transmission networks in their topology. Their radial distribution and complex multi-branch structure make it difficult for traditional methods such as single-ended and double-ended traveling wave methods to accurately pinpoint the fault location. Variations in line parameters have a significant impact on the accuracy of traveling wave location, and the accuracy of traveling wave ranging is also affected by the traveling wave velocity. Using wave velocity to locate faults can introduce varying degrees of error.

[0004] Therefore, there is an urgent need for a more accurate method for locating faults in power distribution network lines. Summary of the Invention

[0005] The main objective of this invention is to provide a multi-branch line traveling wave location method based on fault ratio, which can improve the accuracy of fault location.

[0006] To achieve the above objectives, the first aspect of this application provides a method for locating traveling waves in multi-branch lines based on fault ratio, the method comprising:

[0007] Establish a fault location algorithm based on the inherent length of the line and the reflection of traveling waves;

[0008] Based on the fault location algorithm in multi-branch lines, fault location determination rules are established for different fault locations by utilizing the numerical relationship between the distance from each end to the fault point and the length of the line branch.

[0009] Collect fault voltage traveling wave data;

[0010] Based on the trained neural network model, the fault voltage traveling wave data is processed to obtain the arrival time of the traveling wave front, which reflects the fault distance information.

[0011] Based on the arrival time of the traveling wave front, which reflects the fault distance information, the fault location is determined using the fault location determination rule.

[0012] A second aspect of this application provides a multi-branch line traveling wave location device based on a fault ratio, comprising:

[0013] A module is established to develop a fault location algorithm based on the inherent length of the line and the reflection of traveling waves;

[0014] The establishment module is also used to establish fault location determination rules for different fault locations in multi-branch lines based on the fault location algorithm and by utilizing the numerical relationship between the distance from each end to the fault point and the length of the line branch.

[0015] The acquisition module is used to acquire fault voltage traveling wave data;

[0016] The prediction module is used to process the fault voltage traveling wave data based on a trained neural network model to obtain the arrival time of the traveling wave front, which reflects the fault distance information.

[0017] The processing module is used to locate the fault by using the fault location determination rules based on the arrival time of the traveling wave front reflecting the fault distance information.

[0018] A third aspect of this application provides an electronic device including a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform steps as described in the first aspect and any possible implementation thereof.

[0019] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the steps of the method described in the first aspect.

[0020] This application provides a traveling wave localization method for multi-branch lines based on fault ratio. It establishes a fault localization algorithm based on the inherent length of the line and the reflection of the traveling wave. In multi-branch lines, based on the fault localization algorithm, it establishes fault location determination rules for different fault locations using the numerical relationship between the distance from each end to the fault point and the length of the line branch. It collects fault voltage traveling wave data; processes the fault voltage traveling wave data based on a trained neural network model to obtain the arrival time of the traveling wave front, reflecting fault distance information; and uses the fault location determination rules to locate the fault based on the arrival time of the traveling wave front, reflecting fault distance information. By analyzing the inherent length of the line and the traveling wave reflection transmission path, a fault localization method based on reflection of the traveling wave is proposed, eliminating the need to calculate the traveling wave transmission speed and improving the impact of line parameters on the accuracy of traveling wave localization. It establishes the numerical relationship between the distance from multiple ends of the multi-branch line to the fault point and the length of the line branch, locates the fault segment based on the fault location criteria, and simultaneously analyzes fault information from the time-frequency domain perspective using a neural network, thus improving fault localization accuracy. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] in:

[0023] Figure 1 A flowchart illustrating a multi-branch line traveling wave localization method based on fault ratio provided in an embodiment of this application;

[0024] Figure 2 This is a schematic diagram of a power distribution network topology provided in an embodiment of this application;

[0025] Figure 3 This is a schematic diagram of a CNN-LSTM hybrid neural network model structure provided in an embodiment of this application;

[0026] Figure 4 This is a schematic diagram of a multi-branch line traveling wave location device based on fault ratio provided in an embodiment of this application;

[0027] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0029] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0030] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0031] The traveling wave ranging method involved in this application mainly utilizes the transient traveling wave signal generated at the fault point and transmitted along the transmission line at a speed close to the speed of light to each substation in the power grid. The precise time value of the GPS synchronization clock when the fault traveling wave arrives at the measurement point of each substation is recorded, and the precise location of the fault point is calculated from the GPS time difference measured by the substations on both sides of the fault line.

[0032] The basic principle of traveling wave location technology in distribution network fault detection is mainly based on the traveling wave phenomenon in electromagnetic transient processes. When a fault occurs in a distribution network, two types of traveling waves propagate along power cables: voltage traveling waves and current traveling waves. Both types of traveling waves propagate along the line at near the speed of light. Traveling wave location technology utilizes this physical phenomenon by setting up several traveling wave sensors or monitoring terminals along the distribution line to capture and record traveling wave signals in real time. Once a fault occurs, these sensors receive the initial traveling wave signal generated by the fault and send its time and amplitude information to a data analysis center. The data analysis center uses the traveling wave ranging principle, through precise measurement and calculation of the arrival time of the traveling waves received by each sensor, combined with the known propagation speed of the traveling waves, to estimate the approximate location of the fault.

[0033] The embodiments of this application are described below with reference to the accompanying drawings.

[0034] Please see Figure 1 This is a flowchart illustrating a multi-branch line traveling wave localization method based on fault ratio provided in an embodiment of this application. Figure 1 As shown, the method includes:

[0035] 101. Establish a fault location algorithm based on the inherent length of the line and the reflection of the traveling wave.

[0036] The execution subject of the method in this application embodiment can be a multi-branch line traveling wave location device based on the fault ratio. In practical applications, it can be implemented using terminal equipment, such as a server or computer.

[0037] Specifically, when a fault occurs in a transmission line, the fault traveling wave will be refracted and reflected multiple times in the line. However, the transmission speed of the traveling wave is closely related to the line parameters, and changes in the parameters will cause unstable changes in the traveling wave speed. To avoid this situation, the method in this application embodiment improves upon the traditional single- and double-ended positioning, obtaining a new fault point calculation formula, omitting the preset and calculation of the traveling wave speed, and reducing the impact of line parameters on positioning.

[0038] In one optional implementation, the fault location algorithm based on the inherent length of the line and the reflection of traveling waves includes a single-end fault location formula and a double-end fault location formula, wherein the single-end fault location formula includes:

[0039]

[0040] The above formula for determining the distance to a fault point at both ends includes:

[0041]

[0042] Where d represents the distance from the fault point to the measuring end, and L represents the total length of the line;

[0043] Δt=t M -t N , Δt fM =t fM -t M , Δt fN =t fN -t N , t M and t N It is the time t for the preceding traveling wave to reach the traveling wave detection devices at both ends. fM and t fN It is the time it takes for the reflected traveling wave from the fault point to reach the traveling wave detection devices at both ends first.

[0044] The improved single- and double-ended positioning calculation formulas described above omit the preset and calculation of traveling wave velocity, reducing the impact of line parameters on positioning. Single-ended fault location focuses on measuring the signal at one end; after obtaining the arrival time of the preceding traveling wave, it only needs to identify the reflected traveling wave front of the signal at that end. Double-ended fault location focuses on measuring the signals at both ends; after obtaining the arrival time of the preceding traveling wave, it still needs to identify the reflected traveling wave fronts of the signals at both ends. However, the latter can solve the problem of clock synchronization error in the acquisition device during positioning.

[0045] 102. Based on the above fault location algorithm in multi-branch lines, establish fault location determination rules under different fault locations by utilizing the numerical relationship between the distance from each end to the fault point and the length of the line branch.

[0046] In multi-branch lines, the complex distribution network topology has one or more nodes at each end, resulting in more complex node reflections and refractions, making two-end localization difficult to apply. Therefore, based on the aforementioned fault location algorithm, the fault location determination rules in multi-branch lines can be extended, mainly by the numerical relationship between the distance from each end to the fault point and the length of the line branch. This combines branch determination and fault location algorithms to solve for the fault location.

[0047] In one optional implementation, step 102 includes:

[0048] Based on the above fault location algorithm, in multi-branch lines, the algorithm is extended to the distance between the fault point and the endpoints between any two ends, and the fault point ranging formula for multi-branch lines is obtained.

[0049] Based on the fault location formula for multi-branch lines, the fault ratio relationship is defined, and the fault location determination rules under different fault locations are determined.

[0050] Further, optionally, the fault location formula for the above-mentioned multi-branch lines includes:

[0051]

[0052] in, The fault-preceding traveling wave arrives at line M. x M y The time difference between the two ends, The fault reflected wave arrives at line M respectively x M y The time at both ends.

[0053] Figure 2 This is a schematic diagram of a power distribution network topology provided as an embodiment of this application. Figure 2 As shown, an example analysis is conducted based on this topology:

[0054] In such Figure 2 In the multi-branch line shown, the fault point may be located in the MxPx section or in the PxPy section.

[0055] When the fault point is located on the M1P1 section, the difference in distance between the two leading traveling waves and the time it takes for the reflected traveling wave to arrive at M1 can be calculated by measuring with M1 and M2 as the two ends:

[0056]

[0057] If the fault point is in section P1P2:

[0058]

[0059] The time difference between the arrival of the M1 terminal and the preceding traveling wave.

[0060] Therefore, in complex multi-branch lines, the fault point and endpoint M between any two endpoints are extended. x Distance can be expressed as:

[0061]

[0062] Where MxMy is Figure 2 The lines at any two endpoints of the multi-branch line shown are... The fault-preceding traveling wave arrives at line M. x M y The time difference between the two ends, The fault reflected wave arrives at line M respectively x M y The time at both ends.

[0063] The above equation shows that after the fault traveling wave is generated, the fault point and the endpoint M x Distance is divided into endpoint M x Distance to the fault point and endpoint M x To neighboring node P x The distance between any line endpoint and its nearest neighbor is constant when the distribution network system structure is known. Therefore, the ratio S is defined as:

[0064]

[0065] Indicates line M x P y The fixed length; since the fault point can occur at any location on the line, the S values ​​of each endpoint are known as follows:

[0066]

[0067] Therefore, different fault conditions can be analyzed based on different S values, and the faulty branch can be determined.

[0068] 103. Collect fault voltage traveling wave data.

[0069] The accuracy of fault location is closely related to time. Regardless of the traveling wave location method, it is necessary to detect the arrival time of the traveling wave front at the detection location. Therefore, the calibration of the traveling wave front affects the accuracy. In this embodiment, a neural network is used to extract the arrival time sequence of typical wave fronts that reflect fault distance information, thereby achieving accurate fault location in the distribution network.

[0070] In practical applications, fault voltage traveling wave data can be collected, and then steps 104 and 105 can be executed. The arrival time of the traveling wave front, reflecting fault distance information, can be obtained through a trained neural network model. Then, the fault location can be achieved using the aforementioned established fault location determination rules. The aforementioned fault voltage traveling wave data can be real data or simulated data.

[0071] 104. Based on the trained neural network model, the above-mentioned fault voltage traveling wave data is processed to obtain the arrival time of the traveling wave front, which reflects the fault distance information.

[0072] In this embodiment, the input to the trained neural network model can be the fault voltage traveling wave waveform X(t), and the output can be the time T for the fault-preceding traveling wave to reach the detection devices at both ends. Mx T My and the time T for the reflected traveling wave to reach both ends of the line PxMx T PxMy Parameters such as these.

[0073] In one optional implementation, the neural network model is a CNN-LSTM hybrid neural network model, wherein CNN is used to extract the frequency distribution features of each wavefront, and LSTM is used to extract the time correlation features of typical wavefronts.

[0074] Figure 3 This is a schematic diagram of a CNN-LSTM hybrid neural network model structure provided in an embodiment of this application. Figure 3 As shown, the CNN-LSTM hybrid neural network model includes CNN layers and LSTM layers.

[0075] A convolutional neural network layer is a type of network layer in deep learning used to process data with a grid-like structure (such as images). It extracts features from the input data by using convolution operations.

[0076] Long Short-Term Memory (LSTM) layers are a special type of recurrent neural network (RNN) layer that can learn and remember long-term dependent information. LSTM performs exceptionally well in processing sequential data, such as time series analysis, natural language processing, and speech recognition, because it effectively avoids the vanishing and exploding gradient problems found in traditional RNNs.

[0077] Specifically, the CNN-LSTM hybrid neural network model constructed in this embodiment may include an input layer, convolutional layers, pooling layers, LSTM layers, fully connected layers, and an output layer. Deep learning models have numerous hidden layers, consisting of multiple layers of non-linear operations, which increases model complexity and the number of parameters during computation while enhancing learning capabilities. When training data is limited and other factors influence the model, overfitting can occur. To mitigate the interference of overfitting on the decision-making process, the Dropout algorithm is used to reduce overfitting during data training, thereby increasing the number of Dropout layers. In this embodiment, the CNN extracts the frequency distribution features of each wavefront, and the LSTM extracts the temporal correlation features of typical wavefronts.

[0078] 105. Based on the arrival time of the traveling wave front reflecting the fault distance information, the fault location is determined using the fault location determination rules.

[0079] As mentioned above, in multi-branch lines, based on the above fault location algorithm, the numerical relationship between the distance from each end of the line to the fault point and the length of the branch line is used to establish the fault location judgment criteria under different fault locations. In practical applications, the fault branch can be determined according to the established judgment criteria. At the same time, the above neural network is used to analyze the fault information from the time and frequency domain perspective to improve the fault location accuracy.

[0080] The neural network model in the embodiments of this application will be described in detail below.

[0081] The training process for the aforementioned hybrid neural network model may include:

[0082] A power distribution network simulation model was built. Traveling wave detection points were set at the power distribution network bus, and multiple simulated fault points were set at all points of impedance discontinuity and on any branch. At each simulated fault point, parameters of various fault types were set for batch simulation. The traveling wave front was detected by a traveling wave sensor to obtain simulation data.

[0083] A time-domain traveling wave signal database is constructed by integrating the simulation data and experimental data;

[0084] The time-domain traveling wave signal database is divided into a training set and a test set. The neural network model is iteratively trained using the training set data to obtain the trained neural network model.

[0085] Specifically, during the data acquisition phase, a distribution network simulation model is built in PSCAD / EMTDC, and traveling wave detection points are set at the distribution network bus. Multiple simulated fault points are set at all points of impedance discontinuity and on any branch. At each simulated fault point, parameters for various fault types are set for batch simulation. The traveling wave front is detected by a traveling wave sensor to obtain the fault voltage traveling wave waveform. Simulation data and experimental data can be fused to construct a time-domain traveling wave signal database, thus establishing a multi-point fault holographic database.

[0086] During the training phase, the database is divided into training and testing sets. A CNN-LSTM hybrid neural network model is built, and iterative training is performed using the training set data. The fault voltage traveling wave waveform X(t) is input, and the trained hybrid neural network model is localized. The training results are output, and the arrival time T of the fault-preceding traveling wave at both ends of the detection device is obtained. Mx T My and the time T for the reflected traveling wave to reach both ends of the line PxMx T PxMy Parameters such as these are used to obtain accurate time.

[0087] The following fault location methods have been proposed:

[0088] 1. In complex distribution network topology, the traveling wave detection device is located at the end of the branch line. Based on the arrival time of the fault traveling wave at the end of each line, a fault traveling wave arrival time matrix is ​​established. The difference feature of the time matrix before and after the traveling wave change is used to realize fault location. However, this method has certain limitations in distribution network structures with multiple branches along the line.

[0089] 2. Utilizing the numerical relationship between the length of multiple branch lines and the location of fault points, a fault location method for multi-branch transmission lines based on double-ended traveling waves is proposed, combining branch determination and fault location algorithms to solve for the fault location. However, this method requires calculation and location based on a pre-set traveling wave velocity. When line parameters change, the changing wave velocity will have a significant impact on the accuracy of traveling wave ranging.

[0090] 3. The traveling wave fault location method based on natural frequency extracts the relationship between the main natural frequency and multiple natural frequency values ​​and the fault distance, which can realize the fault location of transmission lines and effectively avoid the problem of traveling wave calibration difficulty. However, in the numerous outgoing lines of each node in a complex distribution network, frequency aliasing will occur, the natural frequency cannot be correctly identified, and the fault location accuracy is not high.

[0091] This application proposes a fault location method based on the inherent length of the line and the reflection of traveling waves, ignoring the traveling wave velocity and reducing the impact of line parameters on location accuracy, thus making the location results more accurate. In multi-branch lines, based on the above fault location algorithm, the numerical relationship between the distance from each end of the line to the fault point and the length of the branch line is used to establish the fault location determination criteria under different fault locations, enabling accurate location of fault points in multi-branch lines. Furthermore, a CNN-LSTM hybrid neural network model is employed. The CNN layer can quickly extract spatial features of the input data, the LSTM layer can extract long-term and short-term dependencies in time series data, introducing fault distance information, the convolutional layer can mine the frequency distribution of each wavefront in the traveling wave, and the long short-term memory layer can mine time correlation information, achieving accurate fault location in the distribution network.

[0092] Based on the description of the foregoing method embodiments, this application also provides a multi-branch line traveling wave location device based on the fault ratio.

[0093] Figure 4 This is a schematic diagram of a multi-branch line traveling wave location device based on a fault ratio, provided as an embodiment of this application. Figure 4 As shown, the multi-branch line traveling wave location device 400 based on fault ratio includes:

[0094] Module 410 is established to create a fault location algorithm based on the inherent length of the line and the reflection of traveling waves;

[0095] The establishment module 410 is also used to establish fault location determination rules for different fault locations in multi-branch lines based on the fault location algorithm and by utilizing the numerical relationship between the distance from each end to the fault point and the length of the line branch.

[0096] Acquisition module 420 is used to acquire fault voltage traveling wave data;

[0097] Prediction module 430 is used to process the fault voltage traveling wave data based on a trained neural network model to obtain the arrival time of the traveling wave front, which reflects the fault distance information.

[0098] Processing module 440 is used to locate the fault using the fault location determination rule based on the arrival time of the traveling wave front reflecting the fault distance information.

[0099] Understandably, this involves Figure 4 The relevant content of each module in the above method embodiments has been described in detail, and you can refer to the content of the method embodiments for details; that is... Figure 4 The provided multi-branch line traveling wave location device 400 based on fault ratio can perform the following: Figure 1Any steps in the illustrated embodiments are not described in detail here. The multi-branch line traveling wave location device 400 based on fault ratio in this application embodiment proposes a fault location method based on catadioptric traveling waves by analyzing the inherent length of the line and the traveling wave reflection transmission path. This eliminates the need to calculate the traveling wave transmission speed, thus improving the impact of line parameters on the accuracy of traveling wave location. It establishes a numerical relationship between the distance from multiple ends of the multi-branch line to the fault point and the length of the line branches, locates the section according to the fault location criteria, and simultaneously uses a hybrid neural network to analyze fault information from the time-frequency domain perspective, thereby improving the fault location accuracy.

[0100] In one embodiment of this application, an electronic device is also provided. See also... Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 500 includes a processor 501 and a memory 502. The memory 502 stores a computer program, which, when executed by the processor 501, will perform actions such as... Figure 1 Any step in the method embodiment shown. The electronic device 500 may also include input / output devices, etc. In a specific embodiment, the electronic device may be a terminal device, etc.

[0101] In one embodiment, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor 501, causes the processor 501 to perform any of the steps in the above method embodiments.

[0102] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0103] 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.

[0104] 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 patent application should be determined by the appended claims.

Claims

1. A method for locating traveling waves in multi-branch lines based on fault ratio, characterized in that, The method includes: Establish a fault location algorithm based on the inherent length of the line and the reflection of traveling waves; Based on the fault location algorithm in multi-branch lines, fault location determination rules are established for different fault locations by utilizing the numerical relationship between the distance from each end to the fault point and the length of the line branch. Collect fault voltage traveling wave data; Based on the trained neural network model, the fault voltage traveling wave data is processed to obtain the arrival time of the traveling wave front, which reflects the fault distance information. Based on the arrival time of the traveling wave front reflecting the fault distance information, the fault location is determined using the fault location determination rule. The fault location algorithm includes a single-end fault point ranging formula and a double-end fault point ranging formula, wherein the single-end fault point ranging formula includes: The formula for determining the distance between two-end fault points includes: Where d represents the distance from the fault point to the measurement end, L represents the total length of the line, and v represents the traveling wave velocity; Δt=t M -t N , Δt fM =t fM -t M , Δt fN =t fN -t N , t M and t N It is the time t for the preceding traveling wave to reach the traveling wave detection devices at both ends. fM and t fN It is the time it takes for the traveling wave reflected from the fault point to reach the traveling wave detection devices at both ends first. The method for determining fault location in multi-branch lines, based on the fault location algorithm, utilizes the numerical relationship between the distance from each end to the fault point and the length of the line branch to establish fault location determination rules for different fault locations, including: Based on the fault location algorithm, in multi-branch lines, the distance between the fault point and the endpoints between any two ends is extended to obtain the fault point ranging formula for multi-branch lines; based on the fault point ranging formula for multi-branch lines, the fault ratio relationship is defined, and the fault location determination rules under different fault locations are determined. The fault location formula for the multi-branch line includes: in, These represent the arrival times of the fault-preceding traveling wave on line M. x / M y Terminal time; These represent the reflected waves from the fault point between the lines reaching line M. x / M y Terminal time; Each represents any branch P between lines. x The reflected wave from the node reaches line M x / M y Terminal time; Indicates the use of line M x M y The endpoint data then leads to the fault point to endpoint M. x The distance between them; For line M x M y The distance between them; The fault ratio relationship is as follows: Among them, M x M y L represents the line between any two endpoints in a multi-branch line. MxPx Indicates endpoint M x To neighboring node P x The distance; The fault location determination rules under different fault locations include:

2. The multi-branch line traveling wave localization method based on fault ratio as described in claim 1, characterized in that, The neural network model is a CNN-LSTM hybrid neural network model, where CNN is used to extract the frequency distribution features of each wavefront, and LSTM is used to extract the time correlation features of typical wavefronts.

3. The multi-branch line traveling wave localization method based on fault ratio as described in claim 2, characterized in that, The training process of the hybrid neural network model includes: A power distribution network simulation model was built. Traveling wave detection points were set at the power distribution network bus, and multiple simulated fault points were set at all points of impedance discontinuity and on any branch. At each simulated fault point, parameters of various fault types were set for batch simulation. The traveling wave front was detected by a traveling wave sensor to obtain simulation data. A time-domain traveling wave signal database is constructed by integrating the simulation data and experimental data; The time-domain traveling wave signal database is divided into a training set and a test set. The neural network model is iteratively trained using the training set data to obtain the trained neural network model.

4. A traveling wave location device for multi-branch lines based on fault ratio, characterized in that, Includes a module that performs the method as described in any one of claims 1-3.

5. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1-3.

6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the processor performs the steps of the method as described in any one of claims 1-3.

Images