Power distribution network fault positioning method, device and equipment based on BAS-PSO algorithm and medium

Abstract

The invention belongs to the technical field of power system fault diagnosis, and particularly discloses a power distribution network fault positioning method, device and equipment based on a BAS-PSO algorithm and a medium, and the method comprises the steps: obtaining a time sequence feature sequence of a power distribution network edge node; inputting the time sequence feature sequence into a pre-trained deep neural network model embedded in the edge node, and outputting a fault probability distribution result; screening candidate fault sections according to a fault probability distribution result; and taking the candidate fault section as an initialized search area, and adopting a BAS-PSO algorithm to obtain a predicted fault position and a corresponding fault probability according to the time sequence feature sequence of the candidate fault section. According to the invention, rapid and accurate positioning of the fault position of the power distribution network can be realized, and the intelligent level of fault positioning is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system fault diagnosis, and in particular relates to a distribution network fault location method, device, equipment and medium based on a BAS-PSO algorithm. Background Art

[0002] The reliability and stability of distribution networks are crucial to the normal operation of power systems. However, with the continuous expansion of power grids and the increase in the integration of new energy sources, the structure of distribution networks has become increasingly complex, making fault location increasingly difficult.

[0003] Traditional fault location methods often rely on manual experience or simple algorithms, which have significant limitations when dealing with complex environments and extreme weather conditions. This can easily lead to inaccurate fault location and slow response times. In the case of new energy stations, extreme weather conditions (such as high temperatures, heavy rain, and typhoons) can significantly impact the stability of distribution networks, increasing the probability of system failures. Existing fault location technologies struggle to effectively address these challenges, necessitating a more advanced fault location approach to improve the system's fault response capabilities and recovery efficiency. Summary of the Invention

[0004] The object of the present invention is to provide a distribution network fault location method, device, equipment and medium based on the BAS-PSO algorithm (Beetle Antennae Search Particle Swarm Optimization) to achieve rapid and accurate positioning of distribution network faults and improve the intelligence level of fault location.

[0005] To achieve the above objectives, the present invention adopts the following technical solutions: According to one aspect of the present invention, a distribution network fault location method based on the BAS-PSO algorithm is provided, comprising the following steps: Obtaining a time series characteristic sequence of a distribution network edge node, wherein the time series characteristic sequence includes voltage, current and zero sequence components; Input the time series feature sequence into a pre-trained deep neural network model embedded in the edge node, and output a fault probability distribution result; filter candidate fault sections according to the fault probability distribution result; The candidate fault segments are uploaded to the central node, which obtains the temporal feature sequence of the candidate fault segments and activates the BAS-PSO algorithm; Taking the candidate fault section as the initial search area, the BAS-PSO algorithm is used to obtain the predicted fault location and the corresponding fault probability according to the temporal feature sequence of the candidate fault section.

[0006] Using this technical solution, edge intelligent computing nodes are deployed at distribution network feeder terminals, switchgear stations, and distributed energy access points. These nodes collect waveform data such as voltage, current, and zero-sequence components in real time, and utilize a deep neural network model for distribution network fault diagnosis, which is more accurate and aligns with real-world applications. A multi-layered fault response architecture combining "edge computing + deep learning + BAS-PSO algorithm" enables rapid local reasoning, determining the presence of a fault and its likely location within milliseconds. This model utilizes a deep neural network model for distribution network fault diagnosis, which is more accurate and aligns with real-world applications. Furthermore, statistical methods based on fault characteristics ensure accurate fault signal extraction, improving the accuracy of fault location.

[0007] According to one embodiment of the present invention, the pre-trained deep neural network model is a CNN (Convolutional Neural Network) model or an LSTM (Long Short-Term Memory) neural network model.

[0008] The lightweight deep learning model embedded in the edge node quickly extracts features and identifies anomalies from the acquired time series feature sequences, determining whether there are power disturbances or potential fault signals, thereby improving response speed and positioning accuracy. Before use, the CNN model or LSTM neural network model is trained using the historical time series feature sequences of the distribution network and the historical fault locations of the distribution network.

[0009] According to one embodiment of the present invention, the BAS-PSO algorithm uses the candidate fault segment as the initial search area and obtains the predicted fault location and corresponding fault probability according to the temporal feature sequence of the candidate fault segment, including the following steps: In the initial stage of fault location, the PSO algorithm is used to search globally to obtain the fault range, and then the BAS algorithm is used for local optimization to obtain the fault point from the fault range; Obtain electrical parameters near the fault point and establish an electrical equivalent model based on the electrical parameters; The error of the electrical equivalent model is minimized, and the predicted fault location and corresponding fault probability are obtained based on the error minimization result.

[0010] Furthermore, the electrical parameters near the fault point include the resistance value and the inductance value between the fault point and the left and right converter stations; the electrical parameters near the fault point also include the impedance value between the fault point and the left and right converter stations.

[0011] Assume that the resistance value between the fault point and the left converter station is R1, the inductance value is L1, the resistance value between the fault point and the right converter station is R2, the inductance value is L2, and the impedance value between the fault point and the left and right converter stations is 、 , where is the angular frequency between the fault point and the left converter station, is the angular frequency between the fault point and the right converter station. and It is difficult to measure, so the four parameters R1, R2, L1, and L2 during short circuit can be identified and measured.

[0012] According to one embodiment of the present invention, in the initial stage of fault location, the PSO global search is used to obtain the fault range, and then the BAS local optimization is used to obtain the fault point from the fault range, including the following steps: In each round of iteration, the inertia weight and learning factor of PSO, as well as the step size and perturbation direction of BAS are dynamically adjusted according to the current search results of PSO and BAS.

[0013] According to one embodiment of the present invention, in each round of iteration, the particle velocity of the PSO algorithm in the next iteration process is updated according to the disturbance direction in the current iteration number of the BAS algorithm; and the step size of the BAS algorithm in the next iteration process is updated according to the current global optimal position obtained by the PSO algorithm and the current global optimal position obtained by the BAS algorithm.

[0014] According to one embodiment of the present invention, the real-time time series feature sequence of the candidate fault section is input into the BAS-PSO algorithm to obtain the predicted fault location and the corresponding fault probability step. Distributed energy and backup power supplies in the distribution network are dispatched according to the predicted fault location and corresponding fault probability.

[0015] According to one embodiment of the present invention, the step of scheduling distributed energy resources and backup power supplies of a distribution network according to predicted fault locations and corresponding fault probabilities includes: Taking the actual power supply power of distributed energy, the power demand of distribution network load and the transmission loss of the power supply path as decision variables, a power deviation loss objective function is constructed. With the goal of minimizing the sum of power deviation and transmission loss, a scheduling plan for distributed energy and backup power supply is obtained.

[0016] By adopting the above technical solution and combining the BAS algorithm with the PSO algorithm, the convergence speed can be improved, effectively solving the problem of low search efficiency in the distribution network fault location process; the location method comprehensively considers the fault characteristics and network topology structure, ensuring stability and reliability in complex network environments.

[0017] According to one aspect of the present invention, a distribution network fault location device based on the BAS-PSO algorithm is provided, comprising: Data acquisition module, used to obtain the time series characteristic sequence of the distribution network; A neural network analysis module is used to input the time series feature sequence into a pre-trained deep neural network model and output a fault probability distribution result; and screen candidate fault sections according to the fault probability distribution result; The fault location module is used to initialize the search area with the candidate fault section and adopt the BAS-PSO algorithm to obtain the predicted fault location and corresponding fault probability according to the temporal feature sequence of the candidate fault section.

[0018] According to one aspect of the present invention, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the distribution network fault location method based on the BAS-PSO algorithm according to any one of the above-mentioned embodiments is implemented.

[0019] According to one aspect of the present invention, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the distribution network fault location method based on the BAS-PSO algorithm according to any of the above embodiments is implemented.

[0020] Compared with the prior art, the present invention has at least the following beneficial effects: 1. This paper utilizes a deep neural network model to diagnose distribution network faults, constructing a multi-layered fault response architecture combining edge computing, deep learning, and a BAS-PSO algorithm. This architecture enables rapid local reasoning, determining the presence of a fault and its possible partition location within milliseconds. The deep neural network model is more accurate and aligns closely with engineering practice. Furthermore, a statistical method based on fault characteristics ensures accurate extraction of fault signals, improving the accuracy of fault location.

[0021] 2. When locating the fault, the BAS algorithm is combined with the PSO algorithm to divide the fault area of ​​the distribution network, and the fault location in each area is optimized and evaluated multiple times to finally achieve accurate positioning of the fault location. This method can fully utilize the global search capability of the PSO algorithm and combine the local optimization characteristics of the BAS algorithm, thereby improving the real-time performance and accuracy of fault response and positioning within milliseconds.

[0022] 3. The method of combining the BAS algorithm with the PSO algorithm can improve the convergence speed and effectively solve the problem of low search efficiency in the distribution network fault location process; the location method comprehensively considers the fault characteristics and network topology structure, ensuring stability and reliability in complex network environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which constitute part of the present invention, are intended to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are intended to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of a distribution network fault location method based on the BAS-PSO algorithm of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other.

[0025] The following detailed description is an exemplary description and is intended to provide further detailed description of the present invention. Unless otherwise indicated, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present invention belongs. The terms used in the present invention are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention.

[0026] Example 1 A distribution network fault location method based on BAS-PSO algorithm is provided. Figure 1 As shown, the following steps are included: Obtaining the time series characteristic sequence of the edge node of the distribution network, the time series characteristic sequence includes voltage, current and zero sequence components; Input the time series feature sequence into the pre-trained deep neural network model embedded in the edge node, and output the fault probability distribution result; filter the candidate fault sections based on the fault probability distribution result; The candidate fault segments are uploaded to the central node, which obtains the temporal feature sequence of the candidate fault segments and activates the BAS-PSO algorithm; Taking the candidate fault section as the initial search area, the BAS-PSO algorithm is used to obtain the predicted fault location and the corresponding fault probability according to the temporal feature sequence of the candidate fault section.

[0027] The pre-trained deep neural network model is a CNN model or an LSTM neural network model.

[0028] The above method utilizes a deep neural network model for fault diagnosis in the distribution network. This model is more accurate and closely aligned with engineering practice. Statistical methods based on fault characteristics ensure accurate fault signal extraction, improving the accuracy of fault location. A multi-layered fault response architecture combining edge computing, deep learning, and the BAS-PSO algorithm enables rapid local reasoning, determining the presence of a fault and its possible partitioned location within milliseconds. When locating the fault, a combination of the BAS algorithm (Bearing Beetle Algorithm) and the PSO algorithm (Particle Swarm Optimization) is used to partition the distribution network's faulty regions. Multiple optimization evaluations are performed on the fault locations within each region, ultimately achieving precise localization of the fault. This method leverages the PSO algorithm's global search capabilities while incorporating the BAS algorithm's local optimization characteristics, thereby improving the real-time performance and accuracy of fault response and location within milliseconds.

[0029] Specifically, taking a six-terminal distribution network as an example, the distribution network fault location method based on the BAS-PSO algorithm of this embodiment is as follows: S1. Obtain the time series feature sequence of the edge nodes of the distribution network.

[0030] Edge intelligent terminals are deployed at edge nodes such as distribution network feeder terminals, switchgear stations and distributed energy access points, and high-frequency sampling devices are used to collect real-time operating data such as local voltage, current, zero-sequence component, etc. The obtained real-time operating data are normalized to obtain a time series feature sequence.

[0031] S2. Use the deep neural network model to obtain and screen candidate fault sections.

[0032] The pre-trained lightweight deep neural network model (CNN model or LSTM neural network model) embedded in the edge node is used to quickly extract features and identify anomalies in the time series feature sequence to determine whether there is power disturbance or potential fault signal.

[0033] During the training phase, the lightweight deep neural network model (CNN model or LSTM neural network model) uses the historical time series feature sequences (including voltage, current, and zero-sequence components) of edge nodes as the training set. The lightweight deep neural network model inputs the historical time series feature sequences containing multi-channel voltage, current, and zero-sequence components within a short time window. The output result is the judgment result of whether there is a power disturbance or fault sign. The cross entropy loss function is used during the training process.

[0034] When used, the trained lightweight deep neural network model inputs the time series feature sequence of the edge nodes of the distribution network and outputs the fault probability distribution results, including the areas where faults may occur and the probability of faults in the areas, etc., to screen out high-risk areas and mark candidate fault sections; generally, areas with a probability of failure higher than 60% are considered high-risk areas and marked as candidate fault sections.

[0035] S3. Use the BAS-PSO algorithm to obtain the predicted fault location and corresponding fault probability.

[0036] The candidate fault area is uploaded to the central node, which activates the BAS-PSO algorithm. The central node's SCADA (Supervisory Control And Data Acquisition) system obtains the time series feature sequence of the candidate fault segment. The time series feature sequence of the candidate fault segment is input into the BAS-PSO algorithm, and a search is performed using the candidate fault segment as the initial search area to obtain the predicted fault location and corresponding fault probability.

[0037] In the BAS-PSO algorithm, in the initial stage of fault location, the PSO algorithm is used to perform a global search on the initial search area to obtain the fault range. The PSO algorithm has a strong global search capability in the early stage and can cover a wider solution space.

[0038] Specifically, the particle positions in the PSO algorithm are initialized to correspond to the candidate fault sections, and the PSO algorithm is iterated according to the following formula: ; ; = + ( ); = + ( ); = + ( ); Where: T is the total number of iterations of the PSO algorithm, t is the current iteration number of the PSO algorithm, is the ratio of the current number of iterations to the total number of iterations; is an intermediate variable; is a control parameter used to control the convergence speed. >1, in this embodiment, set =10; is the inertia weight in the t-th iteration, 、 The upper and lower bounds of the inertia weight are usually set to 1.2 and 0.9; is the cognitive coefficient in the t-th iteration, and They are the upper and lower bounds of the cognitive coefficient, usually set at 2.5 and 0.1; is the social coefficient in the t-th iteration, and They are the upper and lower bounds of the social coefficient, usually set to 3.2 and 0.8.

[0039] and It is the learning factor of the PSO algorithm.

[0040] When the preset number of iterations is reached, the PSO algorithm terminates the iteration. The PSO algorithm dynamically adjusts the inertia weight and learning factors and Perform a global search to quickly locate the approximate location of the fault.

[0041] After using the PSO algorithm to find the approximate location of the fault, the BAS algorithm is used to perform a further fine search on the local range obtained by the PSO algorithm. Further fine search can obtain the fault point from the fault range and improve the accuracy of fault location.

[0042] Specifically, the position of the longhorn beetle in the BAS algorithm is initialized to correspond to the approximate fault location obtained by the PSO algorithm. The BAS algorithm simulates the sensing behavior of the longhorn beetle's left and right whiskers, calculates the evaluation function value of the left and right whisker positions, determines the direction with the smallest error through the sign function, and performs a local fine search in combination with the attenuation step size. The BAS algorithm iterates according to the following formula: + +(1 ; + ( ( ); sign ; =eta ; The left and right beetle's searching behavior: ; ; Where: i Represents the first i Only longhorn beetle; s is the dimension of each longicorn, which is 1 in this embodiment; Represents the current iteration number of the BAS algorithm; Representatives in the k In the iteration, i Only longhorn beetles in the dimension s The position below; Representative refers to k The position of the left beard of the longicorn at the iteration, Representative refers to k The position of the right beetle's whisker at the iteration, and It is used to simulate the longicorn beetle to adjust its position through sensing information with its left and right whiskers to search for the optimal solution; Indicates in k In the iteration, the left beard position of the longicorn is The calculated objective function value; Indicates in k In the iteration, the left beard position of the longicorn is The calculated objective function value; For the k In the iteration i The speed of a longhorn beetle; Representatives in the k In the iteration i Only the extreme value of longhorn beetle; Representatives in the k The extreme value of the longicorn population in the iteration; For the k The increment factor of the position of the longhorn beetle in the iteration; is a random number in the range [0,1]; eta is the step size attenuation factor, which is 0.95 in this embodiment; is the step length; is the search distance, which is set to 0.5 in this embodiment; is the inertia weight; For the k The cognitive coefficient in the iteration, For the k The social coefficient in the iteration; and are random numbers in [0,1].

[0043] When the preset number of iterations is reached, the BAS algorithm terminates the iteration and obtains a more accurate fault point within the approximate range of the fault.

[0044] Furthermore, in the process of optimizing using the BAS-PSO algorithm, a joint fitness function can be used to coordinate the optimization objectives of BAS and PSO. The specific form of the joint fitness function is: = ( x )+ ( x ); in: x Indicates the fault location; is the evaluation function of PSO for the global solution; is the evaluation function of BAS for the local optimal solution; is the dynamic fusion coefficient, ∈ Contributions in the current iteration are dynamically updated.

[0045] ( x ); in: : Multiple energy in location Upper load Actual power supply; :load Power requirements; :Electric energy from the source through the path Losses delivered to the load; n : Indicates the number of loads involved in the power deviation calculation; M : represents the number of power supply paths participating in the scheduling, that is, the number of paths that generate transmission losses and are included in the optimization model; : Artificially set loss weight coefficient.

[0046] ; That is, the current fault location The left and right step lengths are The objective function value is calculated in a small range, and then the absolute value difference is taken as the local evaluation index of the current point. The smaller the value, the closer the current position is to the local minimum, which means it is a stable point with good convergence. On the contrary, The larger the value, the more severe the local fluctuation is, and the location may be far away from the optimal position.

[0047] Furthermore, a bidirectional information sharing mechanism is used to optimize the PSO and BAS algorithms. Specifically, after each iteration, the PSO and BAS algorithms share the following information: the current optimal position, fitness value, historical trajectory, search density, etc. This information is transmitted through an intermediate buffer structure and used to adjust each other's initialization or parameter settings, thereby avoiding invalid searches and overlapping solution spaces.

[0048] The above process is repeated until the termination accuracy or maximum number of generations is met, and the predicted fault location and corresponding fault probability are obtained.

[0049] Furthermore, in each round of iteration, the dynamic parameter coordination mechanism is used to optimize the PSO algorithm and the BAS algorithm. According to the current search results of the BAS algorithm and the PSO algorithm, the inertia weight of the PSO algorithm is dynamically adjusted. , learning factor and , and the step size of the BAS algorithm If the BAS algorithm obtains a high-quality local optimal solution in the current area, the inertia weight of the PSO algorithm is increased to expand the search range; if the PSO algorithm falls into the local optimal area, the BAS algorithm adjusts the step size to improve the ability to escape and avoid falling into the convergence trap.

[0050] Step size of the BAS algorithm Dynamic update is performed based on the global search information of the PSO algorithm. The formula is as follows: ; in: is the step length attenuation coefficient; is the PSO global feedback coefficient; is the current global optimal solution position of PSO (i.e. the global optimal position in the particle swarm); is the current local optimal solution position of BAS.

[0051] To further refine the identification of faulty equipment or nodes, after the fault point is determined using the BAS algorithm, electrical parameters near the fault point, such as voltage, current, resistance, and inductance, are obtained from distribution network sensors or databases. Based on these electrical parameters, mathematical models (such as the resistance-inductance model and the total impedance model) are used to model the distribution network and obtain an electrical equivalent model. This electrical equivalent model is then optimized using an error minimization algorithm (such as the least squares method) to reduce the error between the predicted and actual measured values. The optimized electrical equivalent model is then used to further determine the fault location and probability.

[0052] Among them, resistance-inductance model: Z=R+jL; Where R is the resistance, L is the inductance, and j is the imaginary unit representing the phase of the inductance; Total impedance model: ; Where, is the voltage at the fault point, is the current at the fault point, and Respectively represent the impedance values ​​from the fault point to the two converter stations.

[0053] Specifically, assume that the resistance value between the fault point and the left converter station is R1, the inductance value is L1, the resistance value between the fault point and the right converter station is R2, the inductance value is L2, and the impedance value between the fault point and the left and right converter stations is 、 ,in, is the angular frequency between the fault point and the left converter station, is the angular frequency between the fault point and the right converter station. and It is not easy to measure, so the four parameters R1, R2, L1, and L2 during short circuit can be identified and measured.

[0054] Preferably, the following formula is discretized: ; get: ; ; Where P is the parameter when the system fails , is the sampling constant; 、 They represent the currents in different branches of the circuit, and are physical quantities that describe the directional movement of charges in circuit analysis; They represent the inductance elements in the circuits on the left and right sides of the fault point, which are used to store magnetic field energy. Their characteristics affect the circuit behavior through inductance. 、 Represents the inductor components 、 The voltage across the terminals reflects the voltage drop or rise caused by the inductor's resistance to current changes. They are the identification symbols of two DC voltage sources, which are used to distinguish different DC voltage sources and facilitate the clarification of the action paths and correlations of each power source in circuit analysis and equation establishment; 、 They respectively represent the voltage values ​​of the two DC voltage sources, providing stable DC voltage input for the circuit.

[0055] and are the system A matrices for single-pole short-circuit fault and double-pole fault, respectively, where: ; ; And by: ; ; ; We can get: ; ; in: , r’ and l’ The resistance and inductance per unit length of the line are ，D is the total length of each section of the six-terminal distribution network, is the distance between the fault location and the left converter station, The distance between the fault location and the right converter station.

[0056] Since the above method has only one unknown Therefore, simply solving the above parameters only requires the voltage and current information at a certain moment after the short circuit. However, due to errors in information such as voltage and current, the impedance value may have a large error. Therefore, it is necessary to integrate multiple points. On the basis of the established electrical equivalent model, with the fault location as the only variable, the algorithm is used to optimize the cumulative error of multiple sampling points to achieve high-precision positioning.

[0057] Then, with the goal of minimizing the error, the electrical equivalent model is optimized. The error minimization model is as follows: ; Where, and is the actual measured voltage and current data, and are the voltages and currents predicted based on the electrical equivalent model.

[0058] With the goal of minimizing the error, the electrical model is optimized according to the above formula to obtain the specific location of the fault. The probability of the fault occurring can be calculated based on the timing feature sequence combined with the electrical parameter model.

[0059] To minimize the error in voltage and current information at a single moment, an evaluation function was established to obtain position and probability based on the electrical equivalent model after minimizing the error. The BAS-PSO algorithm was optimized to integrate information from multiple sampling points to minimize the error and achieve precise positioning.

[0060] The evaluation function is: ; ; Where, : Yes f from arrive The comprehensive error measure of the absolute value after summing up; it assists in evaluating the overall error in the BAS-PSO algorithm optimization, by aggregating multiple information, reducing the impact of single sampling point error on fault location.

[0061] : The upper limit of the summation operation, indicating participation Computed The number of sampling points or calculation times reflects the scope of comprehensive information. The larger it is, the more data can be used to reduce error interference.

[0062] : Index variable for sum operation, ranging from 1 to , used to traverse each To accumulate, is to build The identifier that integrates the information of a single sampling point.

[0063] : Indicates the fault location to be optimized; : Function value of a single sampling point or calculation unit, reflected in the fault location The error information of a certain sampling point is the The basic elements, multiple pass The accumulation of To assist algorithm optimization.

[0064] :Evaluation function, for From 1 to At each sampling point, calculate the measured voltage value Calculated value with model The square of the error, plus the measured current value Calculated value with model The error square is summed up. Minimize it through BAS-PSO algorithm , integrate the information of multiple sampling points, eliminate the error of voltage and current information at a single moment, and realize the fault location precise positioning.

[0065] , :Represents the The voltage and current measured at a certain moment or sampling point; , :According to the constructed electrical equivalent model, assuming that the fault occurs at The voltage and current are calculated below.

[0066] The aforementioned BAS-PSO-based distribution network fault location method predicts the distribution network fault location and probability. Distributed energy resources and backup power sources can then be dispatched based on the predicted fault location and corresponding probability. Generally, the BAS algorithm prioritizes the restoration order of critical equipment and loads, and is combined with the PSO algorithm to dynamically optimize the coordinated scheduling of distributed energy resources (wind power, photovoltaics, and energy storage) and backup power sources, rapidly restoring the system's power supply capacity and mitigating the impact of power shortfalls on critical loads.

[0067] Specifically, the method for dispatching distributed energy and backup power supplies of a distribution network includes the following steps: (1) Use the BAS algorithm to prioritize the identification of key equipment and key loads and determine the priority of power restoration.

[0068] (2) Combined with the PSO algorithm, the joint scheduling of distributed energy (wind power, photovoltaic power, energy storage) and backup power sources is dynamically optimized to ensure rapid restoration of power supply and reduce the power gap of critical loads.

[0069] The particle velocity update formula in the PSO algorithm is: = + ( )+ ( ) ; = + ; Where: :particle i exist t +1 generation speed; :particle i exist t The speed of generation; : inertia weight; : Individual learning factor, which indicates the degree to which the particle is attracted by its own historical optimal position, that is, the cognitive coefficient; : Group learning factor, which indicates the degree to which particles are attracted to the global optimal position, that is, the social coefficient; : The value range is Random numbers are used to introduce randomness; :particle i The best historical position; : The current global optimal position of all particles; :particle i exist t The current position of the generation.

[0070] (3) Monitor power demand and energy status in real time and adjust recovery plans dynamically.

[0071] In the step of real-time monitoring of power demand and energy status and dynamically adjusting the recovery plan, in order to optimize power allocation, a power deviation loss objective function is constructed based on the power demand and energy status obtained through monitoring. The goal is to minimize the sum of power deviation and transmission loss. The objective function is solved to obtain the scheduling plan for distributed energy and backup power supply in the distribution network. The objective function is as follows: = ; This function is equivalent to the PSO evaluation function for the global solution defined above. ,Right now: Where: : power deviation loss objective function; : Diversified Energy i Actual power supply; :load i Power requirements; : Power supply path j transmission loss; n : Indicates the number of loads involved in the power deviation calculation; M : represents the number of power supply paths participating in the scheduling, that is, the number of paths that generate transmission losses and are included in the optimization model; : represents the weight coefficient, The value of can be determined based on experience. For example, when the supply and demand balance is prioritized, a smaller value can be set. Values, such as ∈[0.01,0.1]; when reducing transmission loss is the priority, a larger Values, such as ∈[1,10].

[0072] To improve the speed and reliability of distribution network fault location, this embodiment proposes a fault location method based on BAS-PSO. The BAS algorithm and PSO algorithm are the basis of the fault location method. By combining global search capabilities and local optimization characteristics, it adapts to the high-dimensional search requirements of multi-objective optimization and complex power grid environments, and can achieve efficient fault location and recovery under extreme weather conditions. This method adopts a staged closed-loop optimization process, and the algorithm runs iteratively according to the following stages: Phase 1: The PSO algorithm leads the global search to determine the suspected fault area; Phase 2: The BAS algorithm conducts a detailed search in the target area to determine the local optimum. The third stage: the PSO algorithm and the BAS algorithm exchange results and reinitialize parameters, and feedback optimization; The above process is repeated until the termination accuracy or the maximum number of generations is met.

[0073] First, the distribution network is divided into fault zones. A network topology analysis system is used to accurately assess fault nodes in each zone to ensure the accuracy and reliability of fault location. Then, using fault signal strength, equipment damage level, and system recovery time as optimization objectives, the BAS-PSO algorithm is used to develop a comprehensive distribution network fault location strategy, improving the network's fault recovery capability and reliability. Furthermore, the BAS algorithm prioritizes the identification of key equipment and load recovery sequences, and combines this with the PSO algorithm to dynamically optimize the coordinated scheduling of wind power, photovoltaics, energy storage, and backup power sources. Prioritizing the reliability of power supply to critical loads during system recovery significantly improves system recovery efficiency.

[0074] Example 2 A distribution network fault location method based on the BAS-PSO algorithm is provided, which differs from Example 1 in that: In the step of obtaining the predicted fault location and the corresponding fault probability using the BAS-PSO algorithm, in order to improve the adaptability and global optimization efficiency of the BAS-PSO algorithm in complex search spaces, this embodiment proposes a fusion strategy of the "dynamic parameter coordination mechanism" and the "hybrid search incremental mechanism".

[0075] First, the dynamic parameter coordination mechanism can dynamically adjust the parameters of PSO and BAS according to the current search status in each iteration. , learning factor 、 , and the step length of BAS , directional disturbance Both can be adaptively adjusted as fitness changes.

[0076] The inertia weight update follows a linear decreasing function: ; in is the maximum number of iterations of the PSO algorithm, is the current generation of the PSO algorithm. Secondly, the hybrid search increment mechanism introduces the local perturbation vector provided by BAS into the PSO velocity update formula to improve the fine-tuning ability of particles in high-density areas. The revised particle velocity update formula is: ; in: :particle The speed in generation t+1; : Current position of the particle; : The best historical position, : global optimal position; : BAS local optimal direction vector; : Directional gain factor, which adjusts the influence of BAS information on speed update; : random number.

[0077] The feedback of BAS local search direction is introduced into the velocity update of PSO particles. When updating, particles will combine global information with local information, making the search more efficient.

[0078] BAS step size Dynamic update is performed based on the global search information of PSO. The formula is as follows: + ; in: is the step length attenuation coefficient; is the PSO global feedback coefficient; it is used to adjust the feedback influence of the PSO optimal solution on the BAS local jump step size; is the current global optimal solution position of PSO (i.e. the global optimal position in the particle swarm); is the current local optimal solution position of BAS.

[0079] The collaborative mechanism driving function is designed as follows: ; in: : The collaborative mechanism drives the function value; : PSO optimization objective function; : BAS optimization objective function; ; ; , : No. The voltage and current measured at a certain moment or sampling point; , :According to the constructed electrical equivalent model, assuming that the fault occurs at The voltage and current calculated below; : Dynamic fusion coefficient, controlling the weight of search strategy; Fusion coefficient The dynamic adjustment formula is: ; in: : Controls the rate factor of switching during the search phase; : The current iteration number of the PSO algorithm; : Maximum number of iterations of the PSO algorithm.

[0080] Example 3 A distribution network fault location device based on the BAS-PSO algorithm is provided, comprising: A data acquisition module is used to obtain a time series characteristic sequence of the distribution network, where the time series characteristic sequence includes voltage, current and zero sequence components; A neural network analysis module is used to input the time series feature sequence into a pre-trained deep neural network model and output a fault probability distribution result; and screen candidate fault sections according to the fault probability distribution result; The fault location module is used to initialize the search area with the candidate fault section and adopt the BAS-PSO algorithm to obtain the predicted fault location and corresponding fault probability according to the temporal feature sequence of the candidate fault section.

[0081] Example 4 A computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the distribution network fault location method based on the BAS-PSO algorithm of embodiment 1 or embodiment 2 is implemented.

[0082] Example 5 A computer-readable storage medium is provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the distribution network fault location method based on the BAS-PSO algorithm of embodiment 1 or embodiment 2 is implemented.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A distribution network fault location method based on the BAS-PSO algorithm is characterized by: The following steps are involved: Obtain the time series feature sequence of the edge nodes of the distribution network; Input the time series feature sequence into a pre-trained deep neural network model embedded in the edge node, and output a fault probability distribution result; filter candidate fault sections according to the fault probability distribution result; Taking the candidate fault section as the initial search area, the BAS-PSO algorithm is used to obtain the predicted fault location and the corresponding fault probability according to the temporal feature sequence of the candidate fault section.

2. The distribution network fault location method based on the BAS-PSO algorithm according to claim 1 is characterized in that: The time sequence characteristic sequence includes voltage, current and zero sequence components; The pre-trained deep neural network model is a CNN neural network model or an LSTM neural network model.

3. The distribution network fault location method based on the BAS-PSO algorithm according to claim 1 is characterized in that: The BAS-PSO algorithm uses the candidate fault segment as the initial search area and obtains the predicted fault location and corresponding fault probability according to the temporal feature sequence of the candidate fault segment, including the following steps: In the initial stage of fault location, the PSO algorithm is used to search globally to obtain the fault range, and then the BAS algorithm is used for local optimization to obtain the fault point from the fault range; Obtain the electrical parameters between the fault point and the left and right converter stations, and establish an electrical equivalent model based on the electrical parameters; The error of the electrical equivalent model is minimized, and the predicted fault location and corresponding fault probability are obtained based on the error minimization result.

4. The distribution network fault location method based on the BAS-PSO algorithm according to claim 3 is characterized in that: In the initial stage of fault location, the PSO global search is used to obtain the fault range, and then the BAS local optimization is used to obtain the fault point within the fault range, including the following steps: In each round of iteration, the inertia weight and learning factor of PSO, as well as the step size and perturbation direction of BAS are dynamically adjusted according to the current search results of PSO and BAS.

5. The distribution network fault location method based on the BAS-PSO algorithm according to claim 4 is characterized in that: In each round of iteration, the particle velocity of the PSO algorithm in the next iteration is updated according to the disturbance direction in the current iteration of the BAS algorithm; the step size of the BAS algorithm in the next iteration is updated according to the current global optimal position obtained by the PSO algorithm and the current global optimal position obtained by the BAS algorithm.

6. The distribution network fault location method based on the BAS-PSO algorithm according to claim 1 is characterized in that: The BAS-PSO algorithm uses the candidate fault segment as the initial search area and obtains the predicted fault location and the corresponding fault probability step according to the temporal feature sequence of the candidate fault segment. Distributed energy and backup power supplies in the distribution network are dispatched according to the predicted fault location and corresponding fault probability.

7. The distribution network fault location method based on the BAS-PSO algorithm according to claim 6 is characterized in that: The step of scheduling the distributed energy resources and backup power supplies of the distribution network according to the predicted fault location and the corresponding fault probability includes: Taking the actual power supply power of distributed energy, the power demand of distribution network load and the transmission loss of the power supply path as decision variables, a power deviation loss objective function is constructed. With the goal of minimizing the sum of power deviation and transmission loss, a scheduling plan for distributed energy and backup power supply is obtained.

8. A distribution network fault location device based on the BAS-PSO algorithm is characterized by: include: Data acquisition module, used to obtain the time series characteristic sequence of the edge nodes of the distribution network; A neural network analysis module is used to input the time series feature sequence into a pre-trained deep neural network model embedded in the edge node, output a fault probability distribution result, and screen candidate fault sections based on the fault probability distribution result; The fault location module is used to initialize the search area with the candidate fault section and adopt the BAS-PSO algorithm to obtain the predicted fault location and corresponding fault probability according to the temporal feature sequence of the candidate fault section.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the processor is configured to execute a computer program stored in the memory to implement the distribution network fault location method based on the BAS-PSO algorithm as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the distribution network fault location method based on the BAS-PSO algorithm according to any one of claims 1 to 7 is implemented.

Images