Multi-source cooperative power distribution network fault self-healing and recovery optimization method, system and device and storage medium

Abstract

The invention discloses a multi-source cooperative power distribution network fault self-recovery and recovery optimization method, system and device and a storage medium, and relates to the technical field of power system power distribution networks, and the method comprises the steps: collecting the electrical quantity and parameters of each node in real time, and obtaining system state information through data processing; detecting faults by using a multi-criterion algorithm, and determining types and severity; determining a fault section according to a result and minimizing isolation; a collaborative recovery strategy is formulated by integrating multiple sources, and network reconstruction and switching operation optimization are carried out; and finally evaluating a recovery effect, and combining with a historical data feedback adjustment strategy to improve the recovery efficiency. The fault recovery capability of the power distribution network can be remarkably improved, the fault isolation is fast, the load recovery rate is high, the recovery time is short, the economic loss is reduced, the self-healing of different fault scenes is strong, and the unification of technology and economic benefits is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system distribution network, and in particular to a multi-source coordinated distribution network fault self-healing and recovery optimization method, system, device and storage medium. BACKGROUND

[0002] With the continuous expansion of the distribution network scale, the load is also more and more complex, and the traditional fault processing method has been difficult to meet the requirements of modern power system on power supply reliability. The existing distribution network fault self-healing technology lacks consideration of the cooperation of multiple power sources, mainly relies on a single recovery strategy, and usually adopts a preset switch operation sequence, which is difficult to adapt to complex fault scenarios and dynamic system state.

[0003] Most of the existing fault location is based on a single information source, and the single signal source refers to the signals of the fault recorder or the protection device, which makes the fault location vulnerable to measurement errors and communication failures. In the fault recovery process, the traditional method often only considers the recovery of the main network power supply, ignores the potential of the cooperation of multiple signal sources such as distributed power supply, energy storage system and demand response, and lacks effective coping strategies when dealing with complex faults and cascading faults; when selecting an optimization algorithm, heuristic algorithms or simple greedy strategies are mostly selected, which is difficult to find a global optimal solution within a limited time; the uncertainty factors such as load prediction error, device state change and communication delay are not fully considered, the calculation complexity is high, and it is difficult to meet the requirements of real-time decision-making. SUMMARY

[0004] In view of the above problems, the present application provides a multi-source coordinated distribution network fault self-healing and recovery optimization method, system, device and storage medium.

[0005] Therefore, the technical problem solved by the present application is that the single-source recovery capability is limited, the coordination mechanism is imperfect, and the optimization effect is poor in the distribution network fault self-healing.

[0006] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides a multi-source coordinated distribution network fault self-healing and recovery optimization method, comprising: Real-time acquisition of electrical quantities and device operating parameters of each node, and data quality inspection and multi-information source fusion to obtain system state information; Using a multi-criterion fusion fault detection algorithm to analyze the system state information, identifying the fault type and evaluating the fault severity to obtain a fault detection result; Based on the fault detection result, the fault section is determined in combination with the network topology, and the minimum isolation strategy is executed to remove the fault equipment and section; After implementing the minimum isolation strategy, a multi-source collaborative recovery strategy is formulated by comprehensively considering the main grid power supply, distributed power sources, energy storage systems, and demand response. Network reconstruction is performed based on a multi-source collaborative recovery strategy, which changes the network topology through switching operations and optimizes the switching operation sequence. The recovery effect is evaluated, and the multi-source collaborative recovery strategy is optimized and adjusted based on historical data feedback.

[0007] As a preferred scheme for a multi-source collaborative distribution network fault self-healing and recovery optimization method, wherein: The system status information obtained by real-time acquisition of electrical quantities and equipment operating parameters at each node, followed by data quality checks and multi-source fusion, includes: Establish a status monitoring system covering the entire distribution network to collect electrical quantities, switch status, and equipment operating parameters of each node in real time, and optimize the collection frequency according to the data type; design a multi-source information fusion mechanism and establish a communication redundancy mechanism; in the process of data collection and fusion, consider multi-source modeling related information, covering the data collection and fusion of distributed power sources, energy storage systems, and demand response models.

[0008] As a preferred scheme for a multi-source collaborative distribution network fault self-healing and recovery optimization method, wherein: The fault detection algorithm, which utilizes multi-criteria fusion to analyze system state information, identifies fault types, and assesses fault severity, yields fault detection results including: A fault detection algorithm based on multi-criteria fusion is used to comprehensively analyze various fault characteristics, design a fault type identification mechanism to distinguish different fault types, and establish a fault severity assessment model. A fault misjudgment handling mechanism is designed to reconfirm suspicious fault information, and a fault information reporting and sharing mechanism is established.

[0009] As a preferred scheme for a multi-source collaborative distribution network fault self-healing and recovery optimization method, wherein: The fault detection algorithm, which utilizes multi-criteria fusion to analyze system state information, identify fault types, and assess fault severity to obtain fault detection results, also includes: A recovery path optimization model and a collaborative optimization algorithm are introduced. The recovery objective function comprehensively considers the power outage load, switching operation cost and distributed power source startup cost. Power balance and voltage constraints are set, and an improved collaborative optimization algorithm and fitness function are adopted.

[0010] As a preferred scheme for a multi-source collaborative distribution network fault self-healing and recovery optimization method, wherein: The process of determining faulty segments based on fault detection results and network topology, and then implementing a minimum isolation strategy to isolate faulty devices and segments, includes: The distribution network topology is represented using graph theory, and node state vectors and branch state matrices are defined. A real-time decision-making mechanism and fault isolation strategy are introduced. The real-time decision-making adopts a rolling time-domain optimization approach, which comprehensively considers the cost function of future system states and control inputs to determine the optimal control strategy. The weights are dynamically adjusted and updated in real time according to time intervals. Isolation commands are generated based on the comparison between fault current and current threshold. The minimum isolation region is determined by finding the smallest set of candidate isolation regions that meet the conditions.

[0011] As a preferred scheme for a multi-source collaborative distribution network fault self-healing and recovery optimization method, wherein: After implementing the minimum isolation strategy, and comprehensively considering the main grid power supply, distributed power sources, energy storage systems, and demand response, a multi-source collaborative recovery strategy is formulated, including: Considering load restoration priorities, the importance of a load is assessed by comprehensively considering its power level, reliability requirement level, economic impact level, and social impact level. The restoration sequence is optimized based on the load importance to determine the restoration order with the goal of maximizing the overall benefits generated by all loads under the restoration decision.

[0012] The beneficial effects of this preferred technical solution are as follows: by considering load restoration priority and comprehensively evaluating the importance of loads from multiple factors, it can ensure that important loads are restored first during the restoration process, thus guaranteeing the power needs of key users; by optimizing the restoration sequence with the goal of maximizing comprehensive benefits, the overall efficiency and effectiveness of distribution network restoration can be improved, and the rational allocation of resources can be achieved.

[0013] As a preferred scheme for a multi-source collaborative distribution network fault self-healing and recovery optimization method, wherein: The network reconstruction based on the multi-source collaborative recovery strategy, which changes the network topology through switching operations and optimizes the switching operation sequence, includes: During network reconstruction, the spanning tree with the smallest sum of edge weights is found in the candidate set of spanning trees to determine the optimal network topology. The optimization of the switch operation sequence aims to minimize the total operation time and the difference in operation time intervals. The multi-objective coordination optimization adopts a multi-objective function vector, which comprehensively considers multiple objective functions. By determining the Pareto optimal solution, it is determined that there is no other solution that is better than the current solution in all objectives, thereby selecting a better operation scheme.

[0014] The beneficial effects of this preferred technical solution are as follows: by finding the spanning tree with the minimum sum of edge weights to determine the network topology, the reconstructed network structure can be more reasonable, reducing operating costs; the optimization objective of the switching operation sequence can reduce the switching operation time and unstable factors in the operation process, improving operation efficiency and system stability; the multi-objective coordination optimization and Pareto optimal solution determination method can balance and coordinate among multiple objectives, select the operation scheme with better comprehensive performance, and improve the overall effect of distribution network fault recovery.

[0015] Secondly, the present invention provides a multi-source collaborative distribution network fault self-healing and recovery optimization system, comprising: The data acquisition and fusion module is used to collect electrical quantities and equipment operating parameters of each node in real time, and to perform data quality inspection and multi-source fusion to obtain system status information; The fault detection and analysis module is used to analyze system status information, identify fault types and assess fault severity using a fault detection algorithm that integrates multiple criteria, and obtain fault detection results. The fault isolation execution module is used to determine the faulty segment based on the fault detection results and the network topology, and to execute the minimum isolation strategy to isolate the faulty device and segment. The recovery strategy formulation module is used to formulate a multi-source collaborative recovery strategy after executing the minimum isolation strategy, taking into account the main grid power supply, distributed power sources, energy storage systems and demand response. The network reconstruction optimization module is used to perform network reconstruction based on a multi-source collaborative recovery strategy, change the network topology through switching operations, and optimize the switching operation sequence. The effect evaluation and adjustment module is used to evaluate the recovery effect and optimize and adjust the multi-source collaborative recovery strategy based on historical data feedback.

[0016] Thirdly, the present invention provides a computer device, comprising: Memory and processor; 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 multi-source collaborative distribution network fault self-healing and recovery optimization method are implemented.

[0017] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement steps of a multi-source collaborative distribution network fault self-healing and recovery optimization method.

[0018] The beneficial effects of this invention are as follows: The multi-source collaborative mechanism of this invention fully leverages the characteristics and advantages of various power sources. Through the coordinated cooperation of the main grid, distributed power sources, energy storage systems, and demand response, it significantly improves the system's fault recovery capability. Compared with traditional single recovery methods, multi-source collaboration can restore more load in a shorter time, greatly improving power supply reliability. The intelligent fault location and identification technology based on multi-information fusion and machine learning algorithms achieves rapid and accurate fault location, can handle complex fault scenarios, reduces fault handling time, and lays the foundation for rapid recovery. The fault location accuracy exceeds 95%, and the response time is shortened to the second level. The formulated optimized recovery strategy considers multiple constraints and multiple objectives. Through an improved particle swarm optimization algorithm, it achieves global optimization of the recovery scheme, maximizing the load recovery rate and minimizing economic losses while ensuring system safety, thus achieving a balance between technical effectiveness and economic benefits. The real-time decision-making mechanism adopts rolling time-domain optimization and dynamic weight adjustment, which can adapt to the dynamic changes in system state, improve the real-time performance and adaptability of decision-making, provide effective coping strategies for complex and ever-changing fault scenarios, and significantly improve the intelligence level and self-healing capability of the distribution network. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is an overall flowchart of a multi-source collaborative distribution network fault self-healing and recovery optimization method provided by the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0022] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a multi-source collaborative distribution network fault self-healing and recovery optimization method, including: S1: Real-time acquisition of electrical quantities and equipment operating parameters of each node, and data quality inspection and fusion of multiple information sources to obtain system status information; S2: Analyze system state information using a multi-criteria fusion fault detection algorithm, identify fault types and assess fault severity, and obtain fault detection results; S3: Based on the fault detection results and the network topology, determine the faulty segment and execute the minimum isolation strategy to remove the faulty device and segment; S4: After implementing the minimum isolation strategy, a multi-source collaborative recovery strategy is formulated by comprehensively considering the main grid power supply, distributed power sources, energy storage systems and demand response. S5: Perform network reconstruction based on a multi-source collaborative recovery strategy, change the network topology through switching operations, and optimize the switching operation sequence; S6: Evaluate the recovery effect and optimize and adjust the multi-source collaborative recovery strategy based on historical data feedback.

[0023] It should be noted that through steps S1-S6, a complete, intelligent and efficient distribution network fault handling and recovery system was constructed, realizing the process from data collection and monitoring before the fault, to accurate detection and isolation during the fault, and then to multi-source collaborative recovery and strategy optimization after the fault, which significantly improved the reliability and stability of the distribution network power supply.

[0024] Example 2, refer to Figure 1 As an embodiment of the present invention, based on the previous embodiment, a multi-source collaborative distribution network fault self-healing and recovery optimization method is provided, including: In this embodiment, the electrical quantities and equipment operating parameters of each node are collected in real time in step S1 above, and data quality checks and multi-source information fusion are performed to obtain system status information including: Establish a status monitoring system covering the entire distribution network to collect real-time information on electrical quantities such as voltage, current, and power at each node, as well as switch status and equipment operating parameters. The acquisition frequency is optimized based on the data type: millisecond-level acquisition for protection signals, second-level for SCADA (Supervisory Control and Data Acquisition) data, and minute-level for load data. Establish a data quality verification mechanism to ensure the accuracy and completeness of the collected data.

[0025] A multi-information source fusion mechanism is designed, comprehensively utilizing various information sources such as protection devices, fault recorders, smart switches, and user repair reports. A communication redundancy mechanism is established to ensure that critical information can still be obtained when some communication links fail. A real-time database is built to support the rapid storage and retrieval of massive amounts of data.

[0026] When performing data acquisition and fusion, it is also necessary to consider information related to multi-source modeling, including the distributed power generation output model: In the formula, Let i be the output of the i-th distributed power source at time t. To maximize output, This is the available output power.

[0027] The energy storage system model is as follows: In the formula, Let i be the state of charge of the i-th energy storage system at time t. , These represent charge and discharge efficiencies, , These are the charging and discharging powers, This is the rated capacity.

[0028] The demand response model is as follows: In the formula, For the load after demand response, As the baseline load, This is the demand response elasticity coefficient. This serves as an incentive signal for electricity prices.

[0029] It should be noted that the data collection and fusion of these models helps to obtain more comprehensive system status information.

[0030] In another possible implementation, a Bayesian network-based fusion method can be used for multi-source data fusion. First, probabilistic models are performed on the data from each source to determine the prior probability of each source's data. Then, based on Bayes' theorem, the posterior probability is updated using real-time acquired data, thereby achieving multi-source fusion. For example, for electrical quantity data and equipment operating parameter data, separate probabilistic models are established, and when new data arrives, the fused result is calculated using Bayes' formula.

[0031] In another possible implementation, convolutional neural networks (CNNs) from deep learning can be used for multi-source fusion. Data from different sources is preprocessed and then used as input to the CNN. The CNN automatically extracts features from the data using convolutional layers, pooling layers, and other structures, and then fuses these features in a fully connected layer to obtain the final fusion result. For example, node electrical quantity data and switch status data can be normalized and then input into the CNN for fusion.

[0032] In this embodiment, the fault detection algorithm using multi-criteria fusion in step S2 above analyzes the system state information, identifies the fault type, and assesses the fault severity to obtain the fault detection results, including: A multi-criteria fusion fault detection algorithm is employed to comprehensively analyze various fault characteristics such as sudden current changes, voltage drops, and power anomalies. A fault type identification mechanism is designed to distinguish between different types of faults, including short-circuit faults, open-circuit faults, and equipment faults. A fault severity assessment model is established to provide a basis for subsequent recovery strategies.

[0033] Utilize machine learning algorithms to improve the accuracy and speed of fault identification, and establish a historical fault case database and expert knowledge base. Design a fault misjudgment handling mechanism to conduct secondary verification of suspicious fault information. Establish a fault information reporting and sharing mechanism to ensure that relevant personnel receive fault information in a timely manner.

[0034] Specifically, a fault detection algorithm using multi-criteria fusion is used to analyze system state information, identify fault types, and assess fault severity, as expressed as: In the formula, Let N be the probability of failure, and N be the number of information sources. Let k be the weight of the information source. Let be the fault criterion function for the k-th information source, which is the measurement data of the k-th information source.

[0035] In another possible implementation, the multi-criteria fusion fault detection algorithm can also choose a neural network-decision tree fusion algorithm. First, a neural network is used to perform preliminary processing and feature extraction on the fault features, and the output of the neural network is used as the input to the decision tree. The decision tree classifies the faults based on these inputs, determining the fault type and severity. For example, a multilayer perceptron neural network can be used to process fault features such as current and voltage, and then the processing results are input into the decision tree for fault classification.

[0036] In another possible implementation, a multi-criteria fusion algorithm combining fuzzy logic and evidence theory can be used. First, fuzzy logic is used to fuzzify each fault criterion, obtaining the fuzzy membership degree of each criterion. Then, these fuzzy membership degrees are used as basic probability assignments in evidence theory and fused using the Dempster-Shafer synthesis rule to obtain the final fault detection result. For example, overcurrent and overvoltage criteria can be fuzzified before being fused using evidence theory.

[0037] Fault section location is achieved using the current amplitude comparison method: In the formula, This is the faulty section. Let i be the set of adjacent nodes of node i. This is the current before the fault. This represents the current after the fault.

[0038] Fault type identification uses support vector machines: In the formula, y represents the fault type determination result. For Lagrange multipliers, For training sample labels, is the kernel function, b is the bias term, and n is the number of support vectors.

[0039] To achieve more accurate fault detection and handling, a recovery path optimization model and a collaborative optimization algorithm are introduced. The recovery objective function is designed as follows: In the formula, J is the objective function for recovery. For the collection of power outage loads, Let K be the power of the i-th power-off load, and K be the set of switching operations. Let J be the cost of the k-th switching operation, and J be the set of distributed power sources to be started. Let the startup cost of the j-th distributed power source be... , , These are the weighting coefficients.

[0040] In another possible implementation, a particle swarm optimization algorithm can be used to optimize the recovery path. A swarm of particles is initialized, with each particle representing a possible recovery path. Each particle has its own position and velocity, which are updated based on its historical best position and the global best position. Through continuous iteration, the particles gradually converge towards the optimal recovery path. For example, the particle swarm size can be set to 50, and the maximum number of iterations to 100, continuously updating the particle velocities and positions to find the optimal recovery path.

[0041] In another possible implementation, dynamic programming can be used to optimize the recovery path. The recovery process is divided into multiple stages, each with different states and decisions. State transition equations and objective functions are defined, and starting from the last stage, the optimal decision for each stage is calculated recursively in reverse, ultimately yielding the optimal path for the entire recovery process. For example, the recovery process of a distribution network can be divided into multiple stages based on time or switching operation steps, and the optimal recovery path can be solved using dynamic programming.

[0042] The power balance constraint is: In the formula, A set of nodes supplying power to the main grid. Main grid power supply capacity It is a collection of distributed power nodes. It is a collection of energy storage nodes. For energy storage discharge power, To restore the power supply load, This represents the load power.

[0043] The voltage constraint is: In the formula, , These are the lower and upper voltage limits, respectively.

[0044] The collaborative optimization algorithm employs an improved particle swarm optimization algorithm: In the formula, Let be the velocity of the i-th particle in the t-th generation in the d-th dimension. Let w be the position and w be the inertia weight. , For acceleration coefficient, , It is a random number. For the optimal position of an individual, This is the globally optimal position.

[0045] The fitness function is improved to: In the formula, For the fitness function, As a recovery rate indicator, As an economic indicator, For safety indicators, , , These are the weight parameters.

[0046] In this embodiment, step S3 above, based on the fault detection results and combined with the network topology, determines the faulty segment and executes the minimum isolation strategy to isolate the faulty device and segment, including: Based on network topology and fault detection results, the faulty section is quickly identified, and multiple methods such as current distribution analysis and impedance ranging are used to improve positioning accuracy. A minimum isolation strategy is designed to minimize the power outage area while ensuring safety. Rapid isolation operations are then performed to disconnect the faulty equipment and section.

[0047] Establish a safety inspection mechanism for isolation operations to prevent misoperation and escalation of the impact of failures. Design a system safety assessment after isolation to ensure the stable operation of the remaining systems. Establish an emergency power supply plan for the isolation area to provide temporary power to critical loads.

[0048] Specifically, the distribution network topology is represented using graph theory: In the formula, For distribution network diagram, For a set of nodes, Let be the set of edges. This is the weight matrix.

[0049] The node state vector is defined as: In the formula, Let i be the state vector of the i-th node. Active power Reactive power The node voltage amplitude, The voltage phase angle, This serves as an identifier for the node type.

[0050] The branch state matrix is ​​represented as follows: In the formula, Let represent the branch state from node i to node j. For branch current, The branch switch is in the current state. For branch impedance, The health status of the branch road.

[0051] To achieve more accurate and real-time fault isolation, a real-time decision-making mechanism and fault isolation strategy are introduced. Real-time decision-making employs rolling time-domain optimization: In the formula, To predict the time domain, Let the state cost function be... To control the cost function, This is the predicted state value at time t+k. To control the input, To control the increment.

[0052] The dynamic weights are adjusted as follows: In the formula, Let be the dynamic weight at time t. As the benchmark weight, The attenuation coefficient is... For time intervals.

[0053] The fault isolation criterion is: In the formula, For isolation instructions, For fault current, This is the current threshold.

[0054] The minimum isolation area is determined as follows: In the formula, Let A be the minimum isolation region, A be the candidate set of isolation regions, |A| be the region size, and f be the fault point. For the regional boundary, For network boundaries.

[0055] In this embodiment, after implementing the minimum isolation strategy in step S4 above, a multi-source collaborative recovery strategy is formulated by comprehensively considering the main grid power supply, distributed power sources, energy storage systems, and demand response, including: A recovery strategy is formulated based on the current system status and available resources, comprehensively considering various means such as mains power supply, distributed power sources, energy storage systems, and demand response. A multi-objective optimization algorithm is employed to balance recovery speed, economic cost, and system security. A phased recovery plan is designed, prioritizing the recovery of critical loads and facilities.

[0056] Establish a resource scheduling mechanism to coordinate the output of various power sources and the charging and discharging of energy storage systems. Design load management strategies to alleviate supply and demand imbalances through demand response and load shifting. Establish a recovery effectiveness evaluation mechanism to adjust recovery strategies in a timely manner.

[0057] When developing a multi-source collaborative recovery strategy, load recovery priorities need to be considered. Load importance is assessed as follows: In the formula, Let i be the importance of the i-th load. For load power level, For reliability requirement level, Based on economic impact level, Based on the level of social impact, , , , These are the weighting coefficients.

[0058] The recovery sequence is optimized as follows: In the formula, Where T is the load quantity and T is the time period. To restore the decision variables.

[0059] In this embodiment, step S5 above, which involves performing network reconstruction based on a multi-source collaborative recovery strategy, changing the network topology through switching operations, and optimizing the switching operation sequence, includes: Network reconstruction is performed based on a recovery strategy, changing the network topology through switching operations. Graph theory algorithms are used to optimize the switching operation sequence, reducing the number of operations and time. An operation safety check mechanism is established to ensure the safety of each operation.

[0060] Design an operational coordination mechanism to avoid conflicts and interference between multiple operations. Establish an emergency handling mechanism for operational failures and adjust operational plans promptly. Monitor system status changes during operation to ensure that the operational results meet expectations.

[0061] The following algorithms and optimization methods are used for network reconstruction and switching operation sequence optimization: The minimum spanning tree algorithm is used for network reconstruction. In the formula, Let T be the minimum spanning tree and T be the candidate set of spanning trees. Let be the weight of edge (i,j).

[0062] The switching operation sequence is optimized as follows: In the formula, K represents the number of switching operations. Let k be the time of the operation. For time smoothing weights.

[0063] In another possible implementation, the annealing algorithm can be simulated to optimize the switching operation sequence. An initial temperature and annealing rate are set, and a random switching operation sequence is generated as the initial solution. The objective function value of this sequence (such as the total operation time and the difference between operation time intervals) is calculated, and then the sequence is randomly perturbed to obtain a new solution. The objective function values ​​of the new solution and the old solution are compared; if the new solution is better, it is accepted; if the new solution is worse, it is accepted with a certain probability. As the temperature gradually decreases, the optimal switching operation sequence is eventually obtained. For example, with an initial temperature of 100°C and an annealing rate of 0.95, the switching operation sequence is iteratively optimized.

[0064] In another possible implementation, the switching operation sequence can be optimized using a tabu search algorithm. A tabu list is defined, initially empty. An initial switching operation sequence is randomly generated, and its objective function value is calculated. A better solution is searched within the neighborhood. If a better solution is found that is not in the tabu list, it is accepted, and the tabu list is updated. If a solution is in the tabu list but meets certain untabulation conditions, it is also accepted. By continuously searching and updating the tabu list, the optimal switching operation sequence is found. For example, the tabu list length is set to 10, and solutions within the neighborhood are searched each time, with the tabu list updated according to rules.

[0065] Multi-objective coordinated optimization employs a multi-objective function vector: In the formula, For a multi-objective function vector, Let there be i objective functions, and m be the number of objective functions.

[0066] The Pareto optimal solution is determined as follows: In the formula, express Dominate .

[0067] In this embodiment, step S6 above, which evaluates the recovery effect and optimizes and adjusts the multi-source collaborative recovery strategy based on historical data feedback, includes: Establish a recovery effectiveness evaluation system to assess recovery effectiveness from multiple dimensions, including load recovery rate, recovery time, and economic cost. Analyze problems and shortcomings in the recovery process and summarize lessons learned. Establish a continuous optimization mechanism for recovery strategies, improving algorithms and strategies based on evaluation results.

[0068] Establish a fault case library and knowledge base to provide reference for future fault handling. Design a self-learning mechanism to enable the system to continuously learn and improve from historical experience. Establish a post-recovery system monitoring mechanism to ensure stable system operation.

[0069] When evaluating and optimizing the multi-source collaborative recovery strategy, it is necessary to comprehensively consider the recovery path optimization model, collaborative optimization algorithm, load recovery priority, and other factors introduced earlier. Evaluation should be conducted from multiple dimensions, including load recovery rate, economic cost, and system security, by comparing the actual recovery situation with the model's expectations. Simultaneously, historical data should be used to continuously adjust model parameters and weighting coefficients, such as those in the recovery objective function. , , In collaborative optimization algorithms , , To improve the accuracy and adaptability of multi-source collaborative recovery strategies, etc.

[0070] Example 3: The above is an illustrative scheme of a multi-source collaborative distribution network fault self-healing and recovery optimization method according to this embodiment. It should be noted that the technical solution of a multi-source collaborative distribution network fault self-healing and recovery optimization system and the technical solution of the above-described multi-source collaborative distribution network fault self-healing and recovery optimization method belong to the same concept. Details not described in detail in the technical solution of the multi-source collaborative distribution network fault self-healing and recovery optimization system in this embodiment can be found in the description of the above-described multi-source collaborative distribution network fault self-healing and recovery optimization method.

[0071] This embodiment also provides a multi-source collaborative distribution network fault self-healing and recovery optimization system, including: The data acquisition and fusion module is used to collect electrical quantities and equipment operating parameters of each node in real time, and to perform data quality inspection and multi-source fusion to obtain system status information; The fault detection and analysis module is used to analyze system status information, identify fault types and assess fault severity using a fault detection algorithm that integrates multiple criteria, and obtain fault detection results. The fault isolation execution module is used to determine the faulty segment based on the fault detection results and the network topology, and to execute the minimum isolation strategy to isolate the faulty device and segment. The recovery strategy formulation module is used to formulate a multi-source collaborative recovery strategy after executing the minimum isolation strategy, taking into account the main grid power supply, distributed power sources, energy storage systems and demand response. The network reconstruction optimization module is used to perform network reconstruction based on a multi-source collaborative recovery strategy, change the network topology through switching operations, and optimize the switching operation sequence. The effect evaluation and adjustment module is used to evaluate the recovery effect and optimize and adjust the multi-source collaborative recovery strategy based on historical data feedback.

[0072] This embodiment also provides an electronic device applicable to a multi-source collaborative distribution network fault self-healing and recovery optimization method, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement a multi-source collaborative distribution network fault self-healing and recovery optimization method as proposed in the above embodiments.

[0073] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements a multi-source collaborative distribution network fault self-healing and recovery optimization method as proposed in the above embodiments.

[0074] The storage medium proposed in this embodiment belongs to the same inventive concept as the multi-source collaborative distribution network fault self-healing and recovery optimization method proposed in the above embodiment. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0075] Example 4, referring to Tables 1-3, is an embodiment of the present invention, providing a multi-source collaborative distribution network fault self-healing and recovery optimization method. To verify the beneficial effects of the present invention, a simulation experiment is conducted for scientific demonstration.

[0076] This embodiment uses MATLAB to simulate and verify the IEEE 33-node distribution network system, and extends the analysis by incorporating actual distribution network data from a specific region. The simulation system includes six distributed power sources with a total capacity of 1.8MW, three energy storage systems with a total capacity of 0.9MWh, and a controllable load ratio of 25%. The simulation considers various fault scenarios, including single-point faults, multi-point faults, and equipment failures.

[0077] Table 1: Performance Comparison of Different Fault Self-Healing Methods

[0078] Table 2: Multi-source synergistic contribution analysis

[0079] Table 3: Self-healing effect under different fault scenarios

[0080] Simulation results show that the method of this invention significantly outperforms traditional methods in all performance indicators. Fault isolation time is reduced to 4.3 seconds, load recovery rate is increased to 94.6%, recovery time is reduced to 8.9 minutes, and economic losses are reduced by 75%. In the multi-source collaborative mechanism, main grid recovery provides primary support, while distributed power sources and energy storage systems provide rapid response, resulting in the highest cost-effectiveness ratio for demand response. The system exhibits good self-healing capabilities under different fault scenarios, verifying the effectiveness and practicality of the method.

[0081] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-source collaborative method for optimizing fault self-healing and recovery in distribution networks, characterized in that, include: Real-time collection of electrical quantities and equipment operating parameters at each node, followed by data quality verification and fusion of multiple information sources to obtain system status information; A fault detection algorithm based on multi-criteria fusion is used to analyze system state information, identify fault types and assess fault severity, and obtain fault detection results. Based on the fault detection results and the network topology, the faulty segment is determined, and the minimum isolation strategy is implemented to remove the faulty device and segment. After implementing the minimum isolation strategy, a multi-source collaborative recovery strategy is formulated by comprehensively considering the main grid power supply, distributed power sources, energy storage systems, and demand response. Network reconstruction is performed based on a multi-source collaborative recovery strategy, which changes the network topology through switching operations and optimizes the switching operation sequence. The recovery effect is evaluated, and the multi-source collaborative recovery strategy is optimized and adjusted based on historical data feedback.

2. The multi-source collaborative distribution network fault self-healing and recovery optimization method as described in claim 1, characterized in that, The system status information obtained by real-time acquisition of electrical quantities and equipment operating parameters at each node, followed by data quality checks and multi-source fusion, includes: Establish a status monitoring system covering the entire distribution network to collect electrical quantities, switch status, and equipment operating parameters of each node in real time, and optimize the collection frequency according to the data type; design a multi-source information fusion mechanism and establish a communication redundancy mechanism; in the process of data collection and fusion, consider multi-source modeling related information, covering the data collection and fusion of distributed power sources, energy storage systems, and demand response models.

3. The multi-source collaborative distribution network fault self-healing and recovery optimization method as described in claim 2, characterized in that, The fault detection algorithm, which utilizes multi-criteria fusion to analyze system state information, identifies fault types, and assesses fault severity, yields fault detection results including: A fault detection algorithm based on multi-criteria fusion is used to comprehensively analyze various fault characteristics, design a fault type identification mechanism to distinguish different fault types, and establish a fault severity assessment model. A fault misjudgment handling mechanism is designed to reconfirm suspicious fault information, and a fault information reporting and sharing mechanism is established.

4. The multi-source collaborative distribution network fault self-healing and recovery optimization method as described in claim 3, characterized in that, The fault detection algorithm, which utilizes multi-criteria fusion to analyze system state information, identify fault types, and assess fault severity to obtain fault detection results, also includes: A recovery path optimization model and a collaborative optimization algorithm are introduced. The recovery objective function comprehensively considers the power outage load, switching operation cost and distributed power source startup cost. Power balance and voltage constraints are set, and an improved collaborative optimization algorithm and fitness function are adopted.

5. The multi-source collaborative distribution network fault self-healing and recovery optimization method as described in claim 4, characterized in that, The process of determining faulty segments based on fault detection results and network topology, and then implementing a minimum isolation strategy to isolate faulty devices and segments, includes: The distribution network topology is represented using graph theory, and node state vectors and branch state matrices are defined. A real-time decision-making mechanism and fault isolation strategy are introduced. The real-time decision-making adopts a rolling time-domain optimization approach, which comprehensively considers the cost function of future system states and control inputs to determine the optimal control strategy. The weights are dynamically adjusted and updated in real time according to time intervals. Isolation commands are generated based on the comparison between fault current and current threshold. The minimum isolation region is determined by finding the smallest set of candidate isolation regions that meet the conditions.

6. The multi-source collaborative distribution network fault self-healing and recovery optimization method as described in claim 5, characterized in that, After implementing the minimum isolation strategy, and comprehensively considering the main grid power supply, distributed power sources, energy storage systems, and demand response, a multi-source collaborative recovery strategy is formulated, including: Considering load restoration priorities, the importance of a load is assessed by comprehensively considering its power level, reliability requirement level, economic impact level, and social impact level. The restoration sequence is optimized based on the load importance to determine the restoration order with the goal of maximizing the overall benefits generated by all loads under the restoration decision.

7. The multi-source collaborative distribution network fault self-healing and recovery optimization method as described in claim 6, characterized in that, The network reconstruction based on the multi-source collaborative recovery strategy, which changes the network topology through switching operations and optimizes the switching operation sequence, includes: During network reconstruction, the spanning tree with the smallest sum of edge weights is found in the candidate set of spanning trees to determine the optimal network topology. The optimization of the switch operation sequence aims to minimize the total operation time and the difference in operation time intervals. The multi-objective coordination optimization adopts a multi-objective function vector, which comprehensively considers multiple objective functions. By determining the Pareto optimal solution, it is determined that there is no other solution that is better than the current solution in all objectives, thereby selecting a better operation scheme.

8. A multi-source collaborative distribution network fault self-healing and recovery optimization system, using the method described in any one of claims 1 to 7, characterized in that, include: The data acquisition and fusion module is used to collect electrical quantities and equipment operating parameters of each node in real time, and to perform data quality inspection and multi-source fusion to obtain system status information; The fault detection and analysis module is used to analyze system status information, identify fault types and assess fault severity using a fault detection algorithm that integrates multiple criteria, and obtain fault detection results. The fault isolation execution module is used to determine the faulty segment based on the fault detection results and the network topology, and to execute the minimum isolation strategy to isolate the faulty device and segment. The recovery strategy formulation module is used to formulate a multi-source collaborative recovery strategy after executing the minimum isolation strategy, taking into account the main grid power supply, distributed power sources, energy storage systems and demand response. The network reconstruction optimization module is used to perform network reconstruction based on a multi-source collaborative recovery strategy, change the network topology through switching operations, and optimize the switching operation sequence. The effect evaluation and adjustment module is used to evaluate the recovery effect and optimize and adjust the multi-source collaborative recovery strategy based on historical data feedback.

9. An electronic device, characterized in that, include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores computer-executable instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images