Distribution network large-area fault auxiliary decision generation method and electronic equipment

Abstract

The invention discloses a method for generating a large-area fault auxiliary decision of a distribution network and electronic equipment. The method comprises the following steps: acquiring fault information by utilizing a multi-source heterogeneous mining technology; based on the fault information, utilizing a genetic algorithm to generate a power restoration auxiliary decision; acquiring bus load information by using a multi-source heterogeneous mining technology; based on the power supply recovery auxiliary decision and the bus load information, generating a bus power failure batch transfer decision by using a machine learning algorithm; and generating a comprehensive aid decision based on at least one of the power supply restoration aid decision and the bus power failure batch transfer decision, wherein the comprehensive aid decision is used for being displayed to the target user to assist the target user in making a decision. According to the invention, the technical problem of low reliability and applicability of auxiliary decision making of large-area faults of the distribution network is solved.

Description

Technical Field

[0001] The present invention relates to the field of auxiliary decision generation, and in particular to a method and electronic device for generating auxiliary decisions for large-area faults in a distribution network. Background Art

[0002] The analysis of auxiliary decision-making for large-scale power outages in distribution networks involves multiple technical aspects. In terms of the generation of transfer strategies, a prediction-based intelligent distribution network load transfer strategy generation method is proposed. This method first obtains the cross-sectional data of the system operation from the distribution monitoring system, combines the system's primary model information and equipment operation status, and calculates the topological structure of the system operation through the topological service to provide a basis for load transfer. Then, the system prediction results are obtained from the prediction system, and the topological structure and prediction results of the system operation are input into the load transfer expert system. According to the weight configuration of the transfer, the transfer strategy is calculated and generated. Finally, the transfer strategy is stored in the database or displayed on the interface to provide a decision-making basis for the operator.

[0003] In terms of planned maintenance strategy generation, a distribution network maintenance strategy method based on ant colony algorithm is proposed. This method uses ant colony algorithm to solve the problem of distribution equipment status maintenance model. The traditional maintenance model based on minimum cost often leads to too long remaining time. In order to solve this problem, the optimization model is improved and the constraint condition of equipment operation reliability is introduced to limit the number of cycles and cycle duration of the maintenance plan. By analyzing the fault data of actual distribution transformers, an optimal maintenance plan model that can comprehensively consider the number of cycles and total cost is obtained. The model has high reliability.

[0004] A method for generating load transfer strategies for smart distribution networks based on prediction is proposed. This method relies on accurate prediction results to generate transfer strategies, but the prediction results may be affected by many factors, such as weather, emergencies, etc., resulting in inaccurate prediction results, thereby affecting the effectiveness of the transfer strategy; secondly, this method may have shortcomings when processing multi-source heterogeneous data and cannot fully mine the valuable information in the data, resulting in inaccurate fault perception.

[0005] A distribution network maintenance strategy method based on ant colony algorithm is proposed. Although the ant colony algorithm can find a better solution in some cases, it still has problems such as slow convergence and easy to fall into local optimality, which may affect the efficiency and quality of the maintenance strategy generation; secondly, this method relies on the fault data of actual distribution transformers for analysis. If the data is not comprehensive or accurate enough, it may affect the reliability and applicability of the model.

[0006] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0007] The embodiments of the present invention provide a method for generating auxiliary decision-making for large-area faults in a distribution network and an electronic device, so as to at least solve the technical problem that the auxiliary decision-making for large-area faults in a distribution network has low reliability and applicability.

[0008] According to one aspect of an embodiment of the present invention, a method for generating auxiliary decisions for large-scale faults in a distribution network is provided, comprising: obtaining fault information using multi-source heterogeneous mining technology, the fault information including abnormal grid data of a grid area where a large-scale fault power outage occurs; generating a power restoration auxiliary decision based on the fault information using a genetic algorithm, the power restoration auxiliary decision including a strategy for transferring power with a specific switch combination, switch sequence and line path; obtaining bus load information using multi-source heterogeneous mining technology, the bus load information including bus load and transfer capacity; generating a bus power outage batch transfer decision based on the power restoration auxiliary decision and the bus load information using a machine learning algorithm, the bus power outage batch transfer decision including a transfer strategy for transferring the load from a power outage bus to other normally operating buses; generating a comprehensive auxiliary decision based on at least one of the power restoration auxiliary decision and the bus power outage batch transfer decision, the comprehensive auxiliary decision being used to display to a target user to assist the target user in making a decision.

[0009] Furthermore, after generating a bus power outage batch transfer decision based on the power restoration auxiliary decision and bus load information using a machine learning algorithm, the method also includes: obtaining maintenance information, the maintenance information includes equipment affected by the power outage; generating a preliminary maintenance strategy based on the maintenance information, the preliminary maintenance strategy includes the scope and sequence of equipment maintenance during the power transfer process; using a particle swarm optimization algorithm to optimize the preliminary maintenance strategy to obtain a planned maintenance strategy.

[0010] Furthermore, generating a comprehensive auxiliary decision based on at least one of the power supply restoration auxiliary decision and the bus power outage batch transfer decision also includes: generating a comprehensive auxiliary decision based on the power supply restoration auxiliary decision, the bus power outage batch transfer decision and the planned maintenance strategy.

[0011] Furthermore, after generating a bus power outage batch transfer decision using a machine learning algorithm, the method also includes: extracting features of strategy information features of the power supply restoration auxiliary decision and the bus power outage batch transfer decision to obtain classification feature results, the strategy information features including one of the following: plan preparation time, plan preparation personnel; based on the classification feature results, both the power supply restoration auxiliary decision and the bus power outage batch transfer decision are classified and managed.

[0012] Furthermore, after generating the comprehensive auxiliary decision, the method includes: sending the comprehensive auxiliary decision to the main network OMS system through the distribution network OMS system; using a preset interactive interface to verify information between the distribution network OMS system and the main network OMS system, and the verified information includes the name of the remote control switch in the power grid and the location of the remote control switch.

[0013] Furthermore, based on the fault information, a genetic algorithm is used to generate an auxiliary decision for power restoration, including: generating a set of initial power transfer strategies based on the fault information, the fault information including at least one of the following: current mutation information, voltage fluctuation pattern; using the initial power transfer strategy as the initial population of the genetic algorithm; defining the fitness function of the genetic algorithm; performing genetic operations on the initial population, the genetic operations including at least one of the following: selection, crossover, mutation; determining the optimal solution of the genetic algorithm based on the fitness function, and determining the optimal solution as the auxiliary decision for power restoration.

[0014] Furthermore, the fitness function of the genetic algorithm is defined, including: determining an optimization target, the optimization target including at least one of the following: minimizing the scope of power outage, and shortening the power restoration time; and defining the fitness function based on the optimization target.

[0015] Furthermore, based on the power restoration auxiliary decision and bus load information, a machine learning algorithm is used to generate a bus power outage batch transfer decision, including: using the bus load information as a training set; using the training set to train the selected machine learning algorithm to adjust the parameters of the model established by the machine learning algorithm to obtain an evaluation model; and using the evaluation model to generate a bus power outage batch transfer decision.

[0016] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present invention is executed when the program is running.

[0017] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0018] In the embodiment of the present invention, a method of using multi-source heterogeneous mining technology to obtain fault information and bus load information is adopted, and a genetic algorithm is used to generate power restoration auxiliary decision-making and a machine learning algorithm is used to generate bus power outage batch transfer decision, thereby achieving the purpose of realizing fast and accurate fault perception and realizing fast response and efficient auxiliary decision-making through advanced algorithms, thereby achieving the technical effect of being able to quickly respond and provide scientific and reasonable auxiliary decisions when large-scale faults and power outages occur, ensuring the reliability of transfer and the accuracy of maintenance, and thus solving the technical problem of low reliability and applicability of auxiliary decision-making for large-scale faults in distribution networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 It is a hardware structure block diagram of a computer terminal of a method for generating auxiliary decision of large-area fault of distribution network according to one optional embodiment of the present invention;

[0021] Figure 2 is a flow chart of a method for generating auxiliary decision-making for large-area faults in a distribution network according to one optional embodiment of the present invention;

[0022] Figure 3 is a schematic diagram of a method for generating an auxiliary decision for power restoration according to one of the optional embodiments of the present invention;

[0023] Figure 4 is a flow chart of a machine learning algorithm according to one of the optional embodiments of the present invention;

[0024] Figure 5 is a schematic diagram of a framework of a supply transfer strategy management system according to one of the optional embodiments of the present invention;

[0025] Figure 6 is a schematic diagram of a flow chart of a planned maintenance strategy analysis according to one optional embodiment of the present invention;

[0026] Figure 7 is a schematic diagram of a flow chart of a main distribution network remote control switch control analysis according to one optional embodiment of the present invention;

[0027] Figure 8 It is a schematic diagram of the overall technical architecture according to one of the optional embodiments of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] According to an embodiment of the present invention, a method embodiment of a method for generating auxiliary decision-making for large-area faults in a distribution network is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0031] The method embodiment can be executed in an electronic device including a memory and a processor in a computer or a similar computing device. For example, the electronic device running on the computer, such as Figure 1 As shown, the electronic device of the computer may include one or more processors 102 (the processor may include but is not limited to a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processing (DSP) chip, a microprocessor (MCU), a field programmable logic device (FPGA), a neural network processor (NPU), a tensor processing unit (TPU), an artificial intelligence (AI) type processor, etc.) and a memory 104 for storing data. Optionally, the electronic device of the above-mentioned computer may also include a transmission device 106 for communication functions, an input and output device 108, and a display device 110. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the electronic device of the computer. For example, the electronic device of the computer may include more or fewer components than the above structure description, or have a configuration different from the above structure description.

[0032] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the generation method of the large-scale fault auxiliary decision of the distribution network in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the generation method of the large-scale fault auxiliary decision of the distribution network is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0033] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0034] The display device 110 may be, for example, a touch screen type liquid crystal display (LCD) and a touch display (also referred to as a "touch screen" or a "touch display screen"). The liquid crystal display may enable a user to interact with a user interface of the mobile terminal. In some embodiments, the mobile terminal has a graphical user interface (GUI), and a user may interact with the GUI by finger contact and / or gestures on a touch-sensitive surface, wherein the human-computer interaction functions here may optionally include the following interactions: creating web pages, drawing, word processing, making electronic documents, games, video conferencing, instant messaging, sending and receiving emails, call interfaces, playing digital videos, playing digital music and / or web browsing, etc. The executable instructions for executing the above human-computer interaction functions are configured / stored in a computer program product or a readable storage medium executable by one or more processors.

[0035] With the continuous development and expansion of the power system, the distribution network is facing an increasingly complex operating environment. In actual operation, large-scale power outages due to distribution network faults occur from time to time. How to quickly and accurately conduct auxiliary decision-making analysis to improve the efficiency of power transfer and reduce the scope of planned maintenance has become an urgent problem to be solved. At present, traditional fault handling methods often rely on manual experience and rules, which are powerless in the face of complex and changeable power grid structures. The equipment information interaction verification between the distribution network and the main network is not efficient enough, which affects the timeliness and accuracy of fault handling, resulting in a slow decision-making process and difficulty in achieving global optimization. At the same time, traditional means lack flexibility and intelligence, and cannot adapt to real-time changing power grid conditions, resulting in low fault recovery efficiency. In order to overcome these problems, it is necessary to introduce more advanced technical means, such as hybrid optimization algorithms, which combine the advantages of multiple optimization technologies and can generate more scientific and reasonable auxiliary decision-making solutions, and propose more accurate auxiliary decisions for large-scale power outages due to distribution network faults.

[0036] Principle and application of hybrid optimization algorithm: Hybrid optimization algorithm combines the advantages of multiple algorithms to solve complex optimization problems, especially in nonlinear, multi-variable, multi-objective or uncertain scenarios. It usually includes but is not limited to genetic algorithms, particle swarm optimization algorithms, machine learning algorithms and traditional optimization algorithms to improve the algorithm's search efficiency and global optimization capabilities and overcome the limitations of a single algorithm.

[0037] Multi-source heterogeneous data mining technology: Multi-source heterogeneous data mining technology focuses on processing data from different sources with different formats and structures. The core of the technology lies in data preprocessing, feature extraction and selection, and then using appropriate mining algorithms (such as classification, clustering, association rule learning, etc.) to analyze the data and extract valuable information or knowledge. In the power system, it can identify fault modes, predict demand changes, etc. from multiple data sources such as smart meter data, protection relay signals, and distributed energy output information.

[0038] Genetic Algorithm (GA): Genetic Algorithm is a heuristic optimization algorithm inspired by natural selection and genetic mechanisms in the process of biological evolution. In the algorithm, the solution to the problem is encoded as chromosomes, which evolve through selection, crossover and mutation operations, and each generation produces offspring that are more adaptable to the environment. Through iterative optimization, the algorithm can find the approximate optimal solution or the optimal solution to the problem. In the distribution network, genetic algorithms can be used to optimize the transfer strategy and find the best path that can minimize the scope of power outages and restoration time.

[0039] Machine learning algorithms: Machine learning is a branch of artificial intelligence that focuses on building models to allow computers to automatically learn from data and make predictions or decisions. It covers a variety of algorithms, such as support vector machines (SVMs), decision trees, neural networks, etc. In distribution networks, models are trained to predict load changes, the reliability of transfer strategies, and assess potential environmental impacts, thereby generating more intelligent and optimized strategies.

[0040] Particle Swarm Optimization (PSO): Particle Swarm Optimization is a global optimization algorithm inspired by the foraging behavior of bird flocks. In the algorithm, each solution is regarded as a particle in the search space. These particles iteratively update their positions and velocities to eventually find the optimal solution or a near-optimal solution. In power systems, particle swarm optimization can be used to generate planned maintenance strategies, and find the optimal maintenance schedule by considering multiple objectives such as maintenance time, cost, and impact on users.

[0041] Figure 2 is a method according to an embodiment of the present invention, such as Figure 2 As shown, the method comprises the following steps:

[0042] Step S10, using multi-source heterogeneous mining technology to obtain fault information, the fault information includes power grid abnormal data of the power grid area where a large-scale fault power outage occurs;

[0043] Step S20, based on the fault information, using a genetic algorithm to generate a power restoration auxiliary decision, the power restoration auxiliary decision including a strategy for switching power with a specific switch combination, switch sequence and line path;

[0044] Step S30, using multi-source heterogeneous mining technology to obtain bus load information, the bus load information includes bus load and transfer capacity;

[0045] Step S40, based on the power restoration auxiliary decision and bus load information, a bus power outage batch transfer decision is generated using a machine learning algorithm, and the bus power outage batch transfer decision includes a transfer strategy for transferring the load from the power outage bus to other normally operating buses;

[0046] Step S50, generating a comprehensive auxiliary decision based on at least one of the power restoration auxiliary decision and the bus power outage batch transfer decision, the comprehensive auxiliary decision being used to display to the target user to assist the target user in making a decision.

[0047] In the embodiment of the present invention, a method of using multi-source heterogeneous mining technology to obtain fault information and bus load information is adopted, and a genetic algorithm is used to generate power restoration auxiliary decision-making and a machine learning algorithm is used to generate bus power outage batch transfer decision, thereby achieving the purpose of realizing fast and accurate fault perception and realizing fast response and efficient auxiliary decision-making through advanced algorithms, thereby achieving the technical effect of being able to quickly respond and provide scientific and reasonable auxiliary decisions when large-scale faults and power outages occur, ensuring the reliability of transfer and the accuracy of maintenance, and thus solving the technical problem of low reliability and applicability of auxiliary decision-making for large-scale faults in distribution networks.

[0048] The present invention focuses on solving many key technical problems in the scenario of large-scale fault and power outage in the distribution network. In the actual operation of the power system, when a large-scale power outage in the distribution network is caused by natural disasters, equipment failures, etc., the traditional response method has extremely low decision-making efficiency when facing such situations, and it is difficult to produce optimized large-scale power outage auxiliary decisions, bus power outage batch transfer strategies and planned maintenance strategies. At the same time, there are obvious deficiencies in the value mining of multi-source heterogeneous data, and it is impossible to achieve fast and accurate fault perception. Furthermore, the management of the current transfer strategy is not convenient and efficient enough, which is not conducive to operators to query and select. In addition, there is a lack of effective means for information interaction and verification between the distribution network and the main network, making it difficult to maximize the transfer efficiency and minimize the scope of planned maintenance. With the help of a hybrid optimization algorithm, the present invention integrates advanced means such as multi-source heterogeneous data mining technology, genetic algorithm, machine learning algorithm and particle swarm optimization algorithm to generate corresponding strategies and perform information interaction and verification. It is widely used in various complex distribution network environments. When a large-scale power outage occurs, it can respond quickly and provide scientific and reasonable auxiliary decision-making to ensure the reliability of power transfer and the accuracy of maintenance, provide convenience for operators to query and select power transfer strategies, strengthen the coordinated operation of distribution networks and main networks, and effectively improve the stability and reliability of the power system.

[0049] The purpose of the present invention is to provide a distribution network large-scale fault power outage auxiliary decision analysis method based on a hybrid optimization algorithm, so as to solve the problems of low decision-making efficiency, difficulty in generating optimization strategies, insufficient multi-source heterogeneous data mining, inconvenient management of transfer strategies, and lack of effective means for information interaction and verification between the distribution network and the main network in the traditional response methods when the distribution network has a large-scale fault power outage. In view of the problem of low decision-making efficiency of traditional methods in the case of large-scale power outages, the present invention uses advanced algorithms and technical means to achieve rapid response and efficient decision-making. A hybrid optimization algorithm is used to generate a scientific and reasonable transfer strategy to minimize the scope and time of the power outage. An effective information interaction and verification mechanism is established to strengthen the coordinated operation between the distribution network and the main network, and improve the overall stability and reliability of the power system.

[0050] In summary, the present invention uses a variety of advanced algorithms and technologies to achieve rapid fault perception, generate reliable power transfer strategies and accurate planned maintenance strategies, facilitate operator query and selection, strengthen the coordinated operation of the distribution network and the main network, improve the stability and reliability of the power system, and ensure that decisions can be made quickly and effectively and power supply can be restored when a large-scale distribution network failure occurs.

[0051] The present invention provides a distribution network large-scale fault power outage auxiliary decision analysis method based on a hybrid optimization algorithm. The method generates a large-scale power outage auxiliary decision, a bus power outage batch transfer strategy and a planned maintenance strategy according to the hybrid optimization algorithm, transmits them to the distribution network OMS system through a Webservice interface, and then transmits them to the distribution network OCS system for aggregation to obtain an optimized comprehensive decision. The method is finally formed through the interactive verification of equipment information between the distribution network OCS system and the main network OCS system.

[0052] The design of the technical solution, first of all, uses multi-source heterogeneous data mining technology to quickly perceive faults and uses genetic algorithms to generate auxiliary decisions. The preliminary auxiliary decision for large-scale fault power outages is generated; secondly, based on the preliminary auxiliary decision, the machine learning algorithm is used to generate a busbar power outage batch transfer strategy, which needs to consider load information and transfer capacity to ensure the reliability of transfer. This step also requires the management of the transfer strategy obtained after analysis to facilitate operator query and selection; then, the equipment that needs to be repaired for this fault is counted, and the particle swarm optimization algorithm is used to form a planned maintenance strategy. Finally, these strategies are interacted with the distribution network OMS system through the Webservice interface, and the distribution network OMS system and the distribution network OCS system interact with each other, and the distribution network OCS system summarizes and generates a comprehensive strategy. The comprehensive strategy generated by the distribution network OCS system is transmitted to the main network OCS system through the Webservice interface, and the information is interactively verified, and multiple optimized transfer strategies are selected for operation and maintenance personnel to choose, so as to maximize the transfer efficiency and minimize the scope of planned maintenance.

[0053] The main goal of large-scale power outage auxiliary decision-making (also known as power restoration auxiliary decision-making) is to generate effective decisions and achieve rapid power restoration when a large-scale power outage occurs. This application uses multi-source heterogeneous data mining technology to quickly and effectively confirm fault-related information, including specific location, cause of occurrence, etc.; then uses genetic algorithms to find the best power restoration auxiliary decision. Figure 3 The specific implementation steps are as follows:

[0054] (1) Fault perception. Fault perception is to detect and respond to faults in a timely manner, avoid fault expansion, prevent potential safety risks, and improve the reliability of the entire power system. In the case of large-scale power outages caused by distribution network faults, multi-source heterogeneous data mining technology plays a vital role. By deeply mining data from different data sources and different structures, the occurrence of faults can be quickly and accurately perceived. This provides a key information basis for subsequent auxiliary decision-making, and can capture fault signals at the first time, buying valuable time for timely response measures.

[0055] (2) Auxiliary decision generation. After obtaining the specific information of the fault, and based on the fault information, a genetic algorithm is used to generate auxiliary decisions for restoring power supply. Genetic algorithm is a powerful optimization tool that simulates natural selection and genetic mechanisms to solve complex optimization problems. In this application, a genetic algorithm is used to find the optimal power transfer strategy to minimize the impact of power outages and quickly restore power supply. Based on the results of fault perception, the genetic algorithm initially generates a set of feasible power transfer strategies as the initial population of the genetic algorithm; then the advantages and disadvantages of this set of power transfer strategies are evaluated, including considering the reliability of the power transfer and the stability of the system, in order to search for a better power transfer strategy; according to the characteristics of the genetic algorithm, the genetic operation and fitness evaluation are repeated until the optimal power transfer strategy that meets the requirements is found.

[0056] When a large-scale power outage occurs in the distribution network, the initial auxiliary decision refers to a set of emergency response strategies that are quickly generated using optimization algorithms (such as genetic algorithms) after the fault perception stage, based on the initial understanding of the fault and the current state of the power grid. The purpose of this strategy is to restore power supply as quickly as possible and reduce the impact of the power outage on users and the power grid. The generation of initial auxiliary decisions usually includes the following steps:

[0057] 1. Fault perception and identification: Through multi-source heterogeneous data mining technology, the location, type and impact range of the fault can be identified from the data collected by various sensors, smart meters, monitoring systems, etc.

[0058] 2. State assessment: Assess the current state of the grid, including key parameters such as load, voltage, and available resources (such as backup power sources and transfer lines) to determine which parts can be safely used for transfer.

[0059] 3. Strategy generation: Generate one or more power transfer strategies using optimization techniques such as genetic algorithms. Genetic algorithms continuously evolve the solution set by simulating natural selection processes such as crossover, mutation, and selection, and look for the best strategy that can minimize the scope and duration of power outages.

[0060] 4. Preliminary decision: Based on the above steps, a preliminary auxiliary decision is formed, including the transfer path, switch operation sequence, possible transfer capacity, etc., to provide a basis for subsequent refined control and adjustment.

[0061] The generation of preliminary auxiliary decisions is the first step in the fault recovery process, which requires providing a rough solution within a limited time for subsequent detailed analysis and execution.

[0062] Planned maintenance strategy refers to the maintenance and repair activities planned in advance in the distribution network, which is used to preventively check and repair equipment to improve the reliability and efficiency of the power grid and reduce the occurrence of sudden failures. In the present invention, the generation of planned maintenance strategy is based on the evaluation of faulty equipment and in-depth analysis of the power grid operation status. The main steps include:

[0063] 1. Equipment impact analysis: Determine which equipment or lines are affected by the fault, and assess their degree of damage and impact on grid operation.

[0064] 2. Maintenance resource planning: Consider the available human and material resources, as well as the maintenance priority of the equipment, and plan a reasonable maintenance sequence and time window.

[0065] 3. Strategy optimization: Use optimization techniques such as particle swarm optimization algorithm to adjust the maintenance plan to find the maintenance strategy with the least impact on power grid operation and the lowest cost while meeting the maintenance needs.

[0066] 4. Strategy execution and adjustment: Implement the optimized planned maintenance strategy in the power grid, and make dynamic adjustments based on the real-time operating status of the power grid and changes in the external environment (such as weather and load demand) to ensure the smooth progress of maintenance activities and the stable operation of the power grid.

[0067] The purpose of the planned maintenance strategy is to improve the availability of equipment and the overall performance of the system through scientific planning and optimization under the premise of ensuring the safety and stable operation of the power grid. The combination of these two concepts, namely, preliminary auxiliary decision-making and planned maintenance strategy, constitutes the comprehensive solution of the present invention to deal with large-scale power outages caused by distribution network failures. It not only focuses on emergency recovery after failures, but also focuses on preventive maintenance and overhaul to comprehensively improve the operating efficiency and power supply quality of the power system.

[0068] Furthermore, after generating a bus power outage batch transfer decision based on the power restoration auxiliary decision and bus load information using a machine learning algorithm, the method also includes: obtaining maintenance information, the maintenance information includes equipment affected by the power outage; generating a preliminary maintenance strategy based on the maintenance information, the preliminary maintenance strategy includes the scope and sequence of equipment maintenance during the power transfer process; using a particle swarm optimization algorithm to optimize the preliminary maintenance strategy to obtain a planned maintenance strategy.

[0069] After analyzing the busbar power outage batch transfer strategy for the initially generated strategy, it is also necessary to consider the equipment and lines that need to be repaired during the fault, so as to reduce the scope of maintenance, save human resources, and improve the efficiency of power restoration. For planned maintenance, it is first necessary to analyze the scope of the fault. Determine which equipment is affected by the fault and to what extent. This helps to determine the focus and priority of maintenance, arrange maintenance resources reasonably, and improve maintenance efficiency; then, combined with the analysis of the transfer strategy, consider how to transfer power during the maintenance process to ensure that the power supply to users is not overly affected. Analyze the feasibility and risks of transfer, formulate a reasonable transfer plan, and ensure the stable operation of the power system during the maintenance period; finally, optimize the planned maintenance strategy through the particle swarm optimization algorithm. The algorithm can find the optimal solution among many maintenance plans, taking into account the influence of various factors, such as maintenance time, cost, impact on users, etc., and ultimately provide scientific and reasonable auxiliary decision-making, providing strong support for operation and maintenance personnel. The implementation method of planned maintenance strategy analysis is as follows Figure 6 shown.

[0070] Furthermore, generating a comprehensive auxiliary decision based on at least one of the power supply restoration auxiliary decision and the bus power outage batch transfer decision also includes: generating a comprehensive auxiliary decision based on the power supply restoration auxiliary decision, the bus power outage batch transfer decision and the planned maintenance strategy.

[0071] The busbar power outage batch transfer strategy is a strategy to transfer the load from the outage busbar to other normal operating buses in order to ensure the continuity of user power supply and the stability of power grid operation when a certain section of the busbar (i.e., a group of connection points in the substation for collecting and distributing electricity) needs to be shut down due to a fault, maintenance or other reasons in the operation of the power distribution network of the power system. This strategy usually involves multiple steps: 1. Load statistics and analysis: First, it is necessary to count all load points on the outage busbar, including various types of users and power equipment, and analyze their load properties and importance levels. For example, industrial loads, residential loads or loads of critical infrastructure (such as hospitals, emergency services). 2. Transfer path identification: Next, based on the topological structure and operating status of the power grid, identify the paths that can be used for transfer. This includes finding connecting lines between adjacent buses or substations, and evaluating the carrying capacity of the lines and the stability of the power grid. 3. Strategy planning: After identifying possible transfer paths, it is necessary to plan specific transfer strategies. This may include determining the amount of load to be transferred, the time window for transfer, the priority of transfer, etc. Strategy planning needs to take into account the physical constraints of the power grid, such as the maximum current limit of the line, and the balance of power supply and demand to avoid overload or voltage anomalies during the transfer process. 4. Batch transfer: Batch transfer means transferring the load from the blackout bus to the transferable bus not one by one, but in one or more batches. This can reduce the number of operations, improve the efficiency of transfer, and reduce the instantaneous impact on the power grid. 5. Strategy optimization and evaluation: The transfer strategy needs to be optimized through an optimization algorithm to find the optimal solution that meets all constraints (such as line capacity limitations, voltage stability, load importance, etc.). At the same time, the strategy needs to be evaluated to ensure its feasibility and minimize the impact on the operation of the power grid.

[0072] Furthermore, after generating a bus power outage batch transfer decision using a machine learning algorithm, the method also includes: extracting features of strategy information features of the power supply restoration auxiliary decision and the bus power outage batch transfer decision to obtain classification feature results, the strategy information features including one of the following: plan preparation time, plan preparation personnel; based on the classification feature results, both the power supply restoration auxiliary decision and the bus power outage batch transfer decision are classified and managed.

[0073] Effectively manage the obtained transfer strategies to facilitate operators' query and selection. By establishing a complete transfer strategy management system, various transfer plans can be classified, stored and displayed. Operators can quickly select the most appropriate transfer strategy according to the actual situation based on the characteristics of strategy information, such as the time and personnel of the plan preparation. This not only improves the convenience of operation, but also ensures the efficient execution of the transfer strategy. The designed transfer strategy management system framework is as follows: Figure 5 shown.

[0074] Furthermore, after generating the comprehensive auxiliary decision, the method includes: sending the comprehensive auxiliary decision to the main network OMS system through the distribution network OMS system; using a preset interactive interface to verify information between the distribution network OMS system and the main network OMS system, and the verified information includes the name of the remote control switch in the power grid and the location of the remote control switch.

[0075] The main distribution network remote control switch control analysis and implementation method is as follows Figure 7 As shown. After the preliminary auxiliary decisions are analyzed and decided in sequence, the information interaction and verification between the distribution network system and the main network system are involved, including the large-scale power outage auxiliary decisions generated by each link, the bus power outage batch transfer strategy and the planned maintenance strategy are summarized. This application uses the Webservice interface to summarize these strategies and analysis information to the distribution network OMS system, which is then transmitted from the distribution network OMS system to the distribution network OCS system for summary and generation of a comprehensive strategy. The Webservice interface is then used to implement information interaction and verification between the distribution network OCS system and the main network OCS system, including the name and position of the remote control switch in the verification information. After the operator selects the comprehensive strategy to be executed, the main network OCS system performs sequence control through the verified switch opening strategy to achieve power restoration after a fault.

[0076] Furthermore, based on the fault information, a genetic algorithm is used to generate an auxiliary decision for power restoration, including: generating a set of initial power transfer strategies based on the fault information, the fault information including at least one of the following: current mutation information, voltage fluctuation pattern; using the initial power transfer strategy as the initial population of the genetic algorithm; defining the fitness function of the genetic algorithm; performing genetic operations on the initial population, the genetic operations including at least one of the following: selection, crossover, mutation; determining the optimal solution of the genetic algorithm based on the fitness function, and determining the optimal solution as the auxiliary decision for power restoration.

[0077] Furthermore, the fitness function of the genetic algorithm is defined, including: determining an optimization target, the optimization target including at least one of the following: minimizing the scope of power outage, and shortening the power restoration time; and defining the fitness function based on the optimization target.

[0078] The initially generated auxiliary decision does not take into account the load information and transfer capacity of the busbar, so it is necessary to use machine learning algorithms to further consider these two factors in order to obtain a comprehensive transfer strategy. At the same time, the generated strategy needs to be managed to help operators query according to feature information. The specific implementation steps are as follows:

[0079] (1) Load information statistics and transfer capacity analysis. Understanding the load situation on the bus can help predict the load shock that may occur during the power restoration process, so as to take measures to avoid excessive pressure on the power grid; transfer capacity refers to the additional power supply capacity that can be provided by other power sources or paths in a certain area, ensuring that there is enough transfer capacity to quickly switch to the backup path when problems occur in the main power supply path to ensure uninterrupted power supply. Using the above-mentioned multi-source heterogeneous data mining technology, the load information and transfer capacity information are statistically analyzed and imported into the machine learning algorithm for model training. Multiple objectives are considered for optimization, such as minimizing power outages and reducing environmental impacts, etc., to ensure that the generated strategy meets various physical and technical constraints of power grid operation, and then dynamically adjust the strategy based on real-time feedback and new information to adapt to changing situations. The machine learning algorithm process is shown in Figure 3.

[0080] Furthermore, based on the power restoration auxiliary decision and bus load information, a machine learning algorithm is used to generate a bus power outage batch transfer decision, including: using the bus load information as a training set; using the training set to train the selected machine learning algorithm to adjust the parameters of the model established by the machine learning algorithm to obtain an evaluation model; and using the evaluation model to generate a bus power outage batch transfer decision.

[0081] The core of the present invention is to use a hybrid optimization algorithm, combined with advanced technologies such as genetic algorithm, machine learning algorithm and particle swarm optimization algorithm, to deeply mine and analyze a large amount of data in the distribution network system. First, through multi-source heterogeneous data mining technology, real-time operation data, including equipment status, load information, topology structure, etc., are obtained from the distribution monitoring system to provide comprehensive data support for subsequent decision-making. Then, the genetic algorithm is used to generate preliminary preliminary auxiliary decisions, and the machine learning algorithm is used to comprehensively consider the load information and the transfer capacity limit to calculate and generate the optimal busbar power outage batch transfer strategy. At the same time, the particle swarm optimization algorithm is used to formulate the optimal planned maintenance strategy to minimize the maintenance scope and shorten the power outage time. In order to achieve efficient collaborative operation between the distribution network and the main network, the present invention realizes information exchange between the distribution network OMS system and the OCS system, as well as data verification between the distribution network OCS system and the main network OCS system through the Webservice interface. This information exchange and verification mechanism ensures the effective implementation and real-time update of the strategy, so that the distribution network system can better adapt to the real-time changes in power grid conditions. The multiple optimization comprehensive strategies finally formed are displayed through the interface or stored in the database. The operation and maintenance personnel can quickly select the most appropriate strategy according to the actual situation to achieve the goal of maximizing the transfer efficiency and minimizing the planned maintenance scope. Figure 8 shown.

[0082] In an optional embodiment, according to the above technical solution, an example is given:

[0083] Suppose a large-scale power outage occurs in a certain area of ​​a city.

[0084] (1) After the fault signal is detected, the data from different sensors and monitoring devices are analyzed through multi-source heterogeneous data mining technology. For example, the current data obtained from the smart meter shows that the current in a certain area suddenly drops from the normal 100A to nearly 0A; the voltage monitoring device shows that the voltage in the area drops sharply from 220V. It is determined that a fault has occurred in the area, and it is preliminarily judged that the fault location may be in a specific substation or line section.

[0085] (2) Based on the fault information, a genetic algorithm is used to generate auxiliary decisions for power restoration. First, a set of feasible transfer strategies is initially generated as the initial population of the genetic algorithm. For example, consider transferring power from another nearby substation, where the maximum carrying current of the transfer line is 150A and the voltage is 220V. Then the advantages and disadvantages of this set of transfer strategies are evaluated, including the reliability of the transfer and the stability of the system. After multiple genetic operations and fitness evaluations, the optimal transfer strategy that meets the requirements is finally found, such as determining to transfer power with a specific switch combination and line path to ensure that the current is stable at around 120A and the voltage is maintained at 220V during the transfer process.

[0086] (3) Using multi-source heterogeneous data mining technology to conduct statistical analysis on the load information on the bus in the area, it is found that the current bus load is 80kW, while the transfer capacity is 100kW. This information is imported into the machine learning algorithm for model training, and optimization is performed considering goals such as minimizing power outages and reducing environmental impact. The generated transfer strategy may be to gradually increase the transfer current from the initial 50A to 80A within a specific time period to avoid load shock. At the same time, the transfer strategy is managed, and the operator can query and select according to characteristic information such as the plan preparation time.

[0087] (4) Analyze the scope of the fault and determine the equipment and lines affected by the fault. For example, it is found that the insulation damage of a certain section of the line caused the fault, affecting the power supply of multiple users in the surrounding area. Combined with the analysis of the power transfer strategy, consider how to transfer power during the maintenance process. Use the particle swarm optimization algorithm to formulate a reasonable planned maintenance strategy. For example, determine that maintenance will be carried out within the next 2 hours, during which time a specific power transfer path is used to ensure power supply to users, keep the maintenance cost within a certain range, and minimize the impact on users.

[0088] (5) The strategies and analysis information generated in each link are aggregated to the distribution network OMS system, and then transmitted to the distribution network OCS system to generate a comprehensive strategy. The Webservice interface is used to implement information interaction and verification between the distribution network OCS system and the main network OCS system to ensure that the name, location and other information of the remote control switch are accurate. After the operator chooses to execute the comprehensive strategy, the main network OCS system performs sequence control through the verified switch opening strategy to achieve fault power restoration. For example, a specific main network switch is controlled to switch from the closed state to the open state, the current is disconnected from the line where the switch is located, and then the current is redistributed through the transfer line to achieve power restoration in the fault area.

[0089] In summary, the present invention significantly improves the auxiliary decision-making ability and efficiency when a large-scale distribution network failure occurs due to a power outage through advanced hybrid optimization algorithms and information technology means. It not only solves the deficiencies in the prior art, but also enhances the flexibility and intelligence level of the system and improves the automation level of distribution network fault processing. Through the application of the present invention, decisions can be made quickly and effectively and power supply can be restored when a large-scale distribution network failure occurs, thereby ensuring the stable operation of the power system and the power supply reliability of users.

[0090] The present invention has the following innovative features:

[0091] 1. Use advanced data mining technology to process heterogeneous data from different sources to achieve fast and accurate fault perception.

[0092] 2. Take advantage of the hybrid optimization algorithm to optimize the transfer and maintenance strategies.

[0093] 3. Establish a mechanism to ensure accurate exchange and verification of information between the distribution network and the main network.

[0094] The present invention provides the following method:

[0095] 1. A method for auxiliary decision-making analysis of large-scale power outages caused by distribution network faults based on a hybrid optimization algorithm.

[0096] 2. A method to achieve fast and accurate fault perception using multi-source heterogeneous data mining technology.

[0097] 3. A method for generating power transfer strategies by combining genetic algorithm, machine learning algorithm and particle swarm optimization algorithm.

[0098] 4. Use the Webservice interface to implement effective information interaction verification between the distribution network and the main network.

[0099] The present invention provides a comprehensive and efficient method through its unique technical features and the comprehensive application of multiple advanced algorithms. In terms of data mining, the prediction-based method may only rely on the cross-sectional data and primary model information of the system operation. Although the method based on the ant colony algorithm introduces the equipment operation reliability as a constraint condition, it may have limitations in processing multi-source heterogeneous data. The present invention can analyze fault information more comprehensively and achieve fast and accurate fault perception by integrating multi-source heterogeneous data mining technology; in terms of the application of hybrid optimization algorithms, the prediction-based method mainly relies on the weight configuration of the expert system, which may not be flexible enough in the complex and changeable actual operating environment. Although the method based on the ant colony algorithm optimizes the maintenance cycle, the ant colony algorithm itself may have the problem of slow convergence speed and easy to fall into local optimality. The present invention adopts hybrid optimization methods such as genetic algorithm, machine learning algorithm and particle swarm optimization algorithm. The combination of these algorithms can better balance the search efficiency and the quality of the solution, and avoid falling into the local optimal solution. In terms of information interaction and verification mechanism, the existing methods have not formed a specific mechanism, while the present invention emphasizes the information interaction verification between the distribution network and the main network, and adopts effective means to ensure the coordinated operation between the two, which helps to maximize the transfer efficiency and minimize the scope of planned maintenance.

[0100] An embodiment of the present application further provides an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of the present invention when running.

[0101] An embodiment of the present application further provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0102] An embodiment of the present application further provides a computer program product, including a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0103] An embodiment of the present application further provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0104] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0105] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0106] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0107] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0108] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0109] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for generating auxiliary decision-making for large-scale distribution network faults, characterized in that: include: Using multi-source heterogeneous mining technology to obtain fault information, the fault information includes abnormal power grid data in the power grid area where a large-scale power outage occurs; Based on the fault information, a genetic algorithm is used to generate an auxiliary decision for power restoration, wherein the auxiliary decision for power restoration includes a strategy for switching power with a specific switch combination, switch sequence, and line path; Using multi-source heterogeneous mining technology to obtain bus load information, the bus load information includes bus load and transfer capacity; Based on the power restoration auxiliary decision and the bus load information, a bus power outage batch transfer decision is generated using a machine learning algorithm, wherein the bus power outage batch transfer decision includes a transfer strategy for transferring the load from the power outage bus to other normally operating buses; A comprehensive auxiliary decision is generated based on at least one of the power restoration auxiliary decision and the bus power outage batch transfer decision, and the comprehensive auxiliary decision is used to display to the target user to assist the target user in making a decision.

2. The generation method according to claim 1, characterized in that: After generating a bus power outage batch transfer decision based on the power restoration auxiliary decision and the bus load information using a machine learning algorithm, the method further includes: Acquiring maintenance information, wherein the maintenance information includes equipment affected by the power outage; generating a preliminary maintenance strategy based on the maintenance information, wherein the preliminary maintenance strategy includes the scope and sequence of equipment maintenance during the power transfer process; The preliminary maintenance strategy is optimized using a particle swarm optimization algorithm to obtain a planned maintenance strategy.

3. The generation method according to claim 2, characterized in that: Generating a comprehensive auxiliary decision based on at least one of the power restoration auxiliary decision and the busbar power outage batch transfer decision also includes: A comprehensive auxiliary decision is generated based on the power restoration auxiliary decision, the bus power outage batch transfer decision and the planned maintenance strategy.

4. The generation method according to claim 1, characterized in that: After generating a busbar power outage batch transfer decision using a machine learning algorithm, the method further includes: Extracting features of the strategy information of the power restoration auxiliary decision and the busbar power outage batch transfer decision to obtain classification feature results, wherein the strategy information features include one of the following: plan preparation time and plan preparation personnel; Based on the classification feature results, the power restoration auxiliary decision and the bus power outage batch transfer decision are both classified and managed.

5. The generation method according to claim 1, characterized in that: After generating the comprehensive auxiliary decision, the method includes: The comprehensive auxiliary decision is sent to the main network OMS system through the distribution network OMS system; The distribution network OMS system and the main network OMS system are verified by using a preset interactive interface, and the verified information includes the name of the remote control switch in the power grid and the location of the remote control switch.

6. The generation method according to claim 1, characterized in that: Based on the fault information, a genetic algorithm is used to generate an auxiliary decision for restoring power supply, including: Generate a set of initial power transfer strategies based on the fault information, wherein the fault information includes at least one of the following: current mutation information and voltage fluctuation mode; Using the initial transfer strategy as the initial population of the genetic algorithm; Defining a fitness function of the genetic algorithm; Performing a genetic operation on the initial population, wherein the genetic operation includes at least one of the following: selection, crossover, and mutation; An optimal solution of the genetic algorithm is determined based on the fitness function, and the optimal solution is determined as the power restoration auxiliary decision.

7. The generation method according to claim 6, characterized in that: The fitness function of the genetic algorithm is defined, including: Determining an optimization goal, wherein the optimization goal includes at least one of the following: minimizing the scope of power outage and shortening the time to restore power supply; The fitness function is defined based on the optimization objective.

8. The generation method according to claim 1, characterized in that: Based on the power restoration auxiliary decision and the bus load information, a bus power outage batch transfer decision is generated using a machine learning algorithm, including: Using the bus load information as a training set; Using the training set to train the selected machine learning algorithm to adjust parameters of the model established by the machine learning algorithm to obtain an evaluation model; The evaluation model is used to generate the busbar power outage batch transfer decision.

9. An electronic device, characterized in that: include: A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 8 when running.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 8.

Images