Power distribution network resilience planning method, device and equipment considering extreme events

Abstract

The application provides a power distribution network elasticity planning method, device and equipment considering extreme events, and relates to the technical field of power distribution network planning. The method comprises the following steps: fitting an influence expectation of extreme events in a region where a target power distribution network is located based on historical data of the extreme events; wherein the extreme events include meteorological extreme events, geological extreme events and human-caused extreme events; taking the influence expectation and the topological structure of the target power distribution network as inputs, taking reinforcement decision and elasticity resource allocation as outputs, constructing a target function based on node reinforcement decision cost, elasticity resource allocation cost, disaster loss cost and operation cost, constructing a constraint condition based on upper and lower limits of an expectation probability, a power distribution network resilience limit and extreme event measures, and obtaining an elasticity planning model of the target power distribution network; taking minimization of the target function as an optimization target, solving the elasticity planning model, and obtaining an elasticity planning result of the target power distribution network. The application can solve the problem of inaccurate extreme event evaluation in traditional planning.

Description

Technical Field

[0001] This invention relates to the field of distribution network planning technology, and in particular to a method, apparatus and equipment for flexible distribution network planning that takes into account extreme events. Background Technology

[0002] Resilient planning for distribution networks refers to considering the ability to withstand and quickly recover from the impact of extreme events in power grid planning. However, current distribution network planning faces numerous difficulties and challenges when dealing with extreme weather disasters, geological disasters, and man-made sabotage events. Traditional disaster probability models based on long-term historical statistics are gradually becoming ineffective in the context of climate change, the frequency of extreme events is accelerating, and changes in electricity consumption patterns under the "dual carbon" target further increase uncertainty. Because the main equipment in distribution networks has a long lifespan, incorporating extreme events into multi-year plans would dramatically increase the number of planning schemes, leading to a significant increase in the difficulty of subsequent investment planning and construction decisions, as well as project applications.

[0003] Traditional power distribution network planning methods struggle to accurately simulate the specific impacts of different types of extreme disasters on the power distribution network. For example, traditional statistical models cannot reflect the regularity and randomness of the spatiotemporal evolution of meteorological disasters, and cannot accurately assess the losses and risks caused to the power grid by events such as floods and typhoons. This leads to insufficient consideration of disaster scenarios during the planning phase, resulting in planning schemes lacking reliable basis. Furthermore, traditional methods lack clear scenario guidance for pre-disaster deployment of resilient resources or reinforcement and upgrading. The lack of modeling tools that combine specific extreme event scenarios with resilient enhancement measures leads to a disconnect between resilient planning schemes and planning and construction strategies, making it difficult for power supply companies to quantify the impact of extreme events on planning and construction needs. Summary of the Invention

[0004] This invention provides a method, apparatus, and equipment for flexible distribution network planning that considers extreme events, in order to address the problem of insufficient consideration of extreme event scenarios in flexible distribution network planning.

[0005] In a first aspect, embodiments of the present invention provide a distribution network resilient planning method considering extreme events, comprising: Based on historical data of extreme events, the expected impact of extreme events in the target distribution network area is fitted; among them, extreme events include meteorological extreme events, geological extreme events, and human-induced extreme events; The topology affecting the desired and target distribution network is taken as input, and the hardening decision and flexible resource allocation are taken as output. The objective function is constructed based on the node hardening decision cost, flexible resource allocation cost, disaster loss cost and operating cost. The constraints are constructed based on the upper and lower limits of the expected probability, the distribution network resilience limit and extreme event measures, so as to obtain the flexible planning model of the target distribution network. By minimizing the objective function, the elastic programming model is solved to obtain the elastic programming results for the target distribution network.

[0006] In one possible implementation, the expected impact of extreme events in the target distribution network area is fitted based on historical extreme event data, including: Based on historical data of meteorological extreme events, a data-driven deterministic probability distribution is used to fit the expected impact of meteorological extreme events in the target distribution network area.

[0007] In one possible implementation, the expected impact of extreme events in the target distribution network area is fitted based on historical extreme event data, including: Based on historical data of geological extreme events, a fuzzy probability model is used to fit the expected impact of geological extreme events in the target distribution network area.

[0008] In one possible implementation, extreme events in the target distribution network area are fitted based on historical extreme event data, including: For each type of man-made extreme event, the impact of the failure scenario that causes the most severe loss is taken as the expected impact of that type of man-made extreme event.

[0009] In one possible implementation, the objective function is:

[0010] in, To strengthen decision-making With flexible resource allocation decisions The set, The set of feasible solutions. For nodes The cost of reinforcement measures For nodes The cost of elastic resources, This is an integer variable; a value of 1 indicates a node. Strengthening, 0 indicates a node Without reinforcement, This is an integer variable; a value of 1 indicates a node. Configure elastic resources; a value of 0 indicates a node. Without configuring elastic resources, For disaster scenarios, A collection of disaster scenarios. For a set of configurable nodes for elastic resources, For a set of nodes, For disaster The cost of losses incurred after implementing response and recovery measures in a given scenario. For the set of fault uncertainties, Cost per unit of energy output for new energy sources (superscript 0 means t=0, initial annual investment). For the node Scene Time period Renewable power generation capacity (such as wind power and solar power output). The length of a single discrete time step.

[0011] In one possible implementation, constraints are constructed based on extreme event measures, including: For each type of extreme event, the analytic hierarchy process and expert scoring method are used to classify disaster scenarios according to the event occurrence time, location, spatiotemporal evolution, pre-disaster resilience measures, pre-disaster reinforcement measures, and post-disaster measures. It is then used to determine whether measures need to be implemented when each type of extreme event occurs under each disaster scenario, resulting in a 0-1 matrix of extreme event measures.

[0012] In one possible implementation, after solving the elastic programming model with the objective function minimization as the optimization goal, the following is also included: If the flexible planning results of the target distribution network cannot meet the resource regulation needs, then the objective function is to minimize the sum of the costs of newly built substations, capacity expansion and renovation, and new transmission lines in the current year. With construction capacity constraints and reserve coefficient constraints as constraints, the strategy for the construction and expansion of power grid equipment in the target distribution network is solved and used as a supplementary regulation method.

[0013] Secondly, embodiments of the present invention provide a distribution network resilient planning device that considers extreme events, comprising: The fitting module is used to fit the expected impact of extreme events in the target distribution network area based on historical data of extreme events; among which, extreme events include meteorological extreme events, geological extreme events, and human-induced extreme events; The module is used to take the topology affecting the desired and target distribution network as input, and the hardening decision and flexible resource allocation as output. It constructs the objective function based on the node hardening decision cost, flexible resource allocation cost, disaster loss cost and operating cost, and constructs the constraints based on the upper and lower limits of the expected probability, the distribution network resilience limit and extreme event measures, so as to obtain the flexible planning model of the target distribution network. The solver module is used to solve the elastic programming model with the objective function as the optimization objective, and obtain the elastic programming results of the target distribution network.

[0014] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0015] The distribution network resilient planning method, apparatus, and equipment considering extreme events provided in this invention classify extreme event types and fit the expected impact of extreme events based on historical data, transforming elusive disaster risks into quantifiable and calculable mathematical parameters, thus providing a solid scientific basis for planning decisions. Combining the distribution network topology, it constructs an objective function with multiple cost elements and multi-dimensional constraints, innovatively incorporating node reinforcement costs, resilient resource allocation costs, potential disaster loss costs, and operating costs to form a scientific resilient planning model and optimize its solution. This model automatically seeks the optimal balance between investment costs and system resilience, effectively solving the problems of inaccurate extreme event assessments and lack of systematic solutions in traditional planning. This solution achieves precise quantification of the impact of extreme events, providing a reliable basis for planning decisions. Simultaneously, through multi-cost trade-offs and constraint control, it maximizes the reduction of total life-cycle costs while improving the distribution network's resilience and recovery speed in the face of extreme events, balancing resilient needs and economic efficiency, and providing strong protection for the safe and stable operation of the distribution network, demonstrating significant engineering application value. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the implementation of a distribution network resilient planning method considering extreme events, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of scene relationships provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a parallel fault provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a distribution network resilient planning device that takes extreme events into account, provided in an embodiment of the present invention. Figure 5 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0018] See Figure 1 The document illustrates a flowchart of the implementation of the distribution network resilient planning method considering extreme events provided by an embodiment of the present invention, which is described in detail below: Step 101: Based on historical data of extreme events, fit the expected impact of extreme events in the target distribution network area; where extreme events include meteorological extreme events, geological extreme events and human-induced extreme events.

[0019] In this embodiment, extreme events are categorized into three types: meteorological, geological, and human-induced. The expected impact essentially represents a quantitative estimate of the probability and severity of damage or faults to power distribution network components caused by a specific type of extreme event. The fitting process typically relies on statistical analysis and model building of historical observation data. For example, for meteorological events such as typhoons and floods, a mapping relationship between wind speed, rainfall, and line fault probability can be established through long-term meteorological records. This is because accurate expected values ​​are the input basis for subsequent optimization models, and their accuracy directly determines the effectiveness of the planning scheme. In practice, relevant historical data can be extracted from regional meteorological bureaus, seismic monitoring networks, and the power grid company's own fault record databases, and appropriate statistical models can be used for fitting. The beneficial effect of this approach is that it transforms the originally vague and qualitative disaster risks into specific and calculable mathematical parameters, providing reliable data support for subsequent model-based optimization decisions.

[0020] Step 102: The topology affecting the expected and target distribution network is taken as input, and the hardening decision and flexible resource allocation are taken as output. The objective function is constructed based on the node hardening decision cost, flexible resource allocation cost, disaster loss cost and operating cost. The constraints are constructed based on the upper and lower limits of the expected probability, the distribution network resilience limit and extreme event measures to obtain the flexible planning model of the target distribution network.

[0021] In this embodiment, the model input includes the aforementioned expected impact and the topology of the target distribution network. The topology describes the connection relationships of components such as substations, lines, switches, and load points in the power grid, typically represented by a set of nodes and branches and their correlation matrix. The model's design objective is to solve for a set of optimal reinforcement decisions and resilient resource allocation schemes while satisfying a series of constraints. Reinforcement decisions mainly refer to strengthening and upgrading existing power grid equipment, such as reinforcing towers and replacing insulators; resilient resource allocation refers to adding resources that can improve the grid's resilience, such as distributed energy storage, mobile generators, and microgrid black-start devices. The model output is the planning scheme that incorporates these specific decisions. The significance of constructing this model lies in transforming a complex engineering decision problem into a quantifiable and computable mathematical optimization problem.

[0022] The core of the resilient planning model is the objective function. The objective function aims to minimize the total expected cost of the distribution network over its entire lifecycle, and its mathematical expression includes several key cost items. The node reinforcement decision cost item calculates the investment costs for all selected lines or nodes to be reinforced. The resilient resource allocation cost item corresponds to the purchase and installation costs of new energy storage, emergency power supplies, and other resilient resources. The disaster loss cost item is an expected value that integrates various possible disaster scenarios and their probabilities, used to quantify the power outage losses the distribution network will still suffer during a disaster after implementing specific reinforcement and resource allocation schemes; it is typically related to load shedding, outage duration, and unit outage cost. The operating cost item considers the maintenance and wear costs of the configured equipment during normal operation. These cost items are chosen to constitute the objective function because they comprehensively cover the capital expenditures, operating expenditures, and risk costs of the planning scheme, enabling the optimization process to truly achieve a comprehensive optimization of economy and reliability. The acquisition of various cost parameters mainly relies on equipment ledgers, typical cost databases, historical operating data, and economic evaluation standards for power supply reliability.

[0023] Another core component of the model is the constraints. Besides common power flow constraints and operational upper and lower bound constraints, this invention particularly emphasizes three types of key constraints. The expected probability upper and lower bound constraints are mainly used to handle the fuzzy probabilities of geological disasters, ensuring that the optimization process considers all possible scenarios within the probability interval. Distribution network resilience limits set minimum resilience performance indicators for the planned power grid, such as the maximum allowable expected power shortage or the minimum rapid recovery capability requirement. Extreme event response constraints are implemented through a predefined 0-1 matrix. This matrix is ​​constructed as follows: for each specific extreme event, the analytic hierarchy process (AHP) and expert scoring are applied, considering multiple dimensions such as the event's time, spatial location, and spatiotemporal evolution characteristics, combined with pre-disaster resilient measures, reinforcement measures, and post-disaster recovery measures, to form an evaluation system. Experts, based on experience, judge whether a certain measure is effective or necessary under a specific disaster scenario, thus marking it as 1 or 0 in the corresponding position of the matrix. In the mathematical model, this 0-1 matrix is ​​transformed into logical constraints on decision variables. For example, if a certain measure in the matrix is ​​marked as 0 for typhoon disasters, then when optimizing the typhoon scenario, the decision variable corresponding to that measure will be forced to be set to 0, meaning investment is not allowed. The beneficial effect of constructing constraints in this way is that it incorporates the prior knowledge of domain experts, avoids wasting resources on unrealistic or ineffective combinations of measures in the optimization model, and ensures the technical rationality and relevance of the planning scheme.

[0024] Step 103: Solve the elastic programming model with the objective function minimization as the optimization objective to obtain the elastic programming results of the target distribution network.

[0025] In this embodiment, after model construction is completed, the optimization objective is to minimize the objective function. Suitable mathematical optimization algorithms, such as mixed-integer linear programming solvers or heuristic algorithms, are used to solve the flexible planning model. The solution process automatically finds the set of reinforcement and resource allocation decisions that minimize the total expected cost, while satisfying all constraints. The solution result is the preliminary flexible planning result for the target distribution network, which clarifies where, how, and how much flexible resources should be allocated or what kind of reinforcement should be implemented.

[0026] This invention, through classifying extreme event types and fitting the expected impact of extreme events based on historical data, transforms elusive disaster risks into quantifiable and calculable mathematical parameters, providing a solid scientific basis for planning decisions. It innovatively incorporates node reinforcement costs, flexible resource allocation costs, potential disaster loss costs, and operating costs into a scientific flexible planning model, automatically seeking the optimal balance between investment costs and system resilience. This effectively solves the problems of inaccurate extreme event assessments and lack of systematic solutions in traditional planning. This solution achieves precise quantification of the impact of extreme events, providing a reliable basis for planning decisions. Furthermore, through multi-cost trade-offs and constraint control, it minimizes the total lifecycle cost while improving the distribution network's resilience and recovery speed in the face of extreme events, balancing flexibility needs with economic efficiency. This provides strong protection for the safe and stable operation of the distribution network and has significant engineering application value.

[0027] In one possible implementation, the expected impact of extreme events in the target distribution network area is fitted based on historical extreme event data, including: Based on historical data of meteorological extreme events, a data-driven deterministic probability distribution is used to fit the expected impact of meteorological extreme events in the target distribution network area.

[0028] In this embodiment, the impact of extreme weather events is specifically characterized as follows: ; In the formula: Let S be the mathematical expectation under scenario S; Let S be the expected value of scenario S. ; For objects, the weather probability is data-driven in scenario S; When extreme events are categorized as meteorological, such as typhoons, ice storms, floods, or extreme cold / heat events, historical data is relatively abundant and exhibits regularity. Therefore, for meteorological extreme events, a data-driven deterministic probability distribution is preferred to fit their expected impact. This means utilizing a large amount of historical meteorological data and corresponding power grid fault data to describe the probability of equipment failure under specific meteorological conditions through a probability density function. For example, based on historical typhoon path and intensity data, combined with a geographic information system, the wind speed distribution at different locations during typhoon passage can be simulated, and the failure probability can be calculated based on the wind speed-line failure rate curve. This method is chosen because the development of meteorological events conforms to physical laws to a certain extent, and historical data can effectively reflect their statistical characteristics. Parameters are mainly obtained from publicly available meteorological databases and internal accident reports from power grid companies. This method can generate deterministic probability values, which are easy to use directly in optimization models, thereby improving the accuracy of meteorological disaster risk assessment and the relevance of planning schemes.

[0029] In one possible implementation, the expected impact of extreme events in the target distribution network area is fitted based on historical extreme event data, including: Based on historical data of geological extreme events, a fuzzy probability model is used to fit the expected impact of geological extreme events in the target distribution network area.

[0030] In this embodiment, the impact of geological extreme events is specifically characterized based on fuzzy set fitting that considers all failure scenarios: ; In the formula: This represents the lower bound of the expected probability of geological disasters in the local area. This represents the upper limit of the expected probability of geological disasters in the local area.

[0031] For geological extreme events such as earthquakes, landslides, or debris flows, their occurrence is characterized by greater suddenness and difficulty in precise prediction, and historical data is relatively scarce. Therefore, this invention employs a fuzzy probability model to fit the expected impact. The fuzzy probability model does not provide a single, exact probability value, but rather a probability interval or membership function to describe the fuzzy range of the likelihood of the disaster. For example, based on historical data of regional geological activity and expert experience, the probability of an earthquake can be set between a lower and upper limit. The fuzzy model is chosen because it better handles the inherent uncertainty and incomplete information of geological disasters. The parameters required for the model, such as the upper and lower limits of the probability interval, can be obtained through geological survey reports, historical disaster records, and expert assessment. The advantage of this approach is that it acknowledges and quantifies cognitive uncertainty, enabling the planning model to consider worst-case scenarios, thereby designing more robust power grid reinforcement and resource allocation schemes.

[0032] In one possible implementation, extreme events in the target distribution network area are fitted based on historical extreme event data, including: For each type of man-made extreme event, the impact of the failure scenario that causes the most severe loss is taken as the expected impact of that type of man-made extreme event.

[0033] In this embodiment, the distribution network extreme event resilient programming model takes into account the impact of the most severe failure scenario caused by human-induced extreme events, specifically characterized as follows: ; In the formula: for We expect to increase constraints.

[0034] When considering extreme human-caused events, such as sabotage or cyberattacks, these events are often highly random and covert, lacking consistent historical data suitable for statistical modeling. To address this, this invention employs a conservative yet pragmatic strategy: for each specific extreme human-caused event, one or more typical failure scenarios capable of causing the most severe damage are selected and defined, and the impact of these scenarios is directly used as the expected impact of the event. For example, for a cyberattack, a scenario could be assumed where a critical substation's control system is compromised, leading to a complete power outage. This selection is made because, given the inability to accurately predict probabilities, using the worst-case scenario as the basis for planning maximizes the safety of the power grid under extreme conditions. Scenario definitions are typically based on cybersecurity analysis reports, past security incident cases, and deductions by domain experts. While this approach may seem conservative, its beneficial effect lies in ensuring that the planning scheme possesses a basic defensive baseline and emergency preparedness in the face of deliberate sabotage.

[0035] In one possible implementation, the objective function is:

[0036] in, To strengthen decision-making With flexible resource allocation decisions The set, The set of feasible solutions. For nodes The cost of reinforcement measures For nodes The cost of elastic resources, This is an integer variable; a value of 1 indicates a node. Strengthening, 0 indicates a node Without reinforcement, This is an integer variable; a value of 1 indicates a node. Configure elastic resources; a value of 0 indicates a node. Without configuring elastic resources, For disaster scenarios, A collection of disaster scenarios. For a set of configurable nodes for elastic resources, For a set of nodes, For disaster The cost of losses incurred after implementing response and recovery measures in a given scenario. For the set of fault uncertainties, Cost per unit of energy output for new energy sources (superscript 0 means t=0, initial annual investment). For the node Scene Time period Renewable power generation capacity (such as wind power and solar power output). The length of a single discrete time step.

[0037] In this embodiment, the objective function aims to minimize the total expected cost of the distribution network over its entire lifecycle. Its mathematical expression includes several key cost items. The node hardening decision cost item calculates the investment costs for all selected lines or nodes to be hardened. The flexible resource allocation cost item corresponds to the purchase and installation costs of new flexible resources such as energy storage and emergency power supplies. The disaster loss cost item is an expected value that integrates various possible disaster scenarios and their probabilities. It quantifies the power outage losses the distribution network will still suffer during a disaster after implementing specific hardening and resource allocation schemes, and is typically related to load shedding, outage duration, and unit outage cost. The operating cost item considers the maintenance and wear costs of the configured equipment during normal operation. These cost items are chosen to constitute the objective function because they comprehensively cover the capital expenditures, operating expenditures, and risk costs of the planning scheme, enabling the optimization process to truly achieve a comprehensive optimization of economy and reliability. The acquisition of various cost parameters mainly relies on equipment ledgers, typical cost databases, historical operating data, and economic evaluation standards for power supply reliability.

[0038] In the objective function, The routine operating costs associated with the configured equipment components, = + + The unit is usually MW or kW, and the data comes from the scale of wind turbines, photovoltaics and energy storage of 10 kV and below in the region during the planning year. The length of a single discrete time step, in h.

[0039] In one possible implementation, constraints are constructed based on extreme event measures, including: For each type of extreme event, the analytic hierarchy process and expert scoring method are used to classify disaster scenarios according to the event occurrence time, location, spatiotemporal evolution, pre-disaster resilience measures, pre-disaster reinforcement measures, and post-disaster measures. It is then used to determine whether measures need to be implemented when each type of extreme event occurs under each disaster scenario, resulting in a 0-1 matrix of extreme event measures.

[0040] In this embodiment, based on the category and specific extreme events, the Analytic Hierarchy Process (AHP) and expert scoring method are applied to construct an extreme event 0-1 matrix according to the event occurrence time, occurrence location, spatiotemporal evolution, pre-disaster resilience measures, pre-disaster reinforcement measures, and post-disaster measures, as shown in Table 1 below.

[0041] Table 1

[0042] In one possible implementation, after solving the elastic programming model with the objective function minimization as the optimization goal, the following is also included: If the flexible planning results of the target distribution network cannot meet the resource regulation needs, then the objective function is to minimize the sum of the costs of newly built substations, capacity expansion and renovation, and new transmission lines in the current year. With construction capacity constraints and reserve coefficient constraints as constraints, the strategy for the construction and expansion of power grid equipment in the target distribution network is solved and used as a supplementary regulation method.

[0043] In this embodiment, after obtaining the preliminary flexible planning results, a situation may arise where, despite optimal reinforcement and flexible resource allocation, in certain extreme disaster scenarios or rapid load growth, these "soft" measures alone may still be insufficient to meet the power supply reliability requirements or resource regulation needs of all loads, resulting in capacity shortages. To address this, the present invention provides a supplementary regulation method. This method treats the expansion of the power grid physical structure, including the construction of new substations, the expansion and renovation of existing substations, and the construction of new transmission lines, as an independent optimization problem. Its objective function is to minimize the present value of the sum of the costs of new substations, expansion and renovation, and new transmission lines in each year of the planning period. Simultaneously, it is constrained by actual engineering conditions, such as the upper limit of the number of new substations added annually, the reasonable range of substation reserve coefficients, and the target total substation capacity to be achieved. This supplementary optimization process can be triggered when capacity shortages still occur during simulation operation. The solution results in a time-series strategy for the construction and expansion of power grid equipment, which, together with the aforementioned flexible planning scheme, constitutes a complete, hierarchical distribution network flexible enhancement planning system. This two-step strategy has the advantage of refining investment efficiency, prioritizing more cost-effective and flexible methods to solve problems, and only initiating larger-scale traditional power grid expansions when necessary, thereby achieving better overall economic efficiency.

[0044] Establish a distribution network planning and construction strategy that includes the allocation of resources for new substations, expansion and renovation of existing substations, and new transmission lines. Specific steps include: Construct the cost objective function: ; In the formula, To minimize costs, r is the discount rate. Let the total cost be for year i, where i = 1, 2, 3, ..., n; the decision cost is defined to include the cost of constructing new substations in that year. Expansion and renovation costs And the cost of building new transmission lines That is: ; The cost of building a new substation is: ; In the formula, The cost per substation includes the land cost, equipment cost, construction cost, and labor cost required for each newly built substation. The number of newly built substations. The cost of flexible resources; The cost of expanding or renovating a substation is as follows: ; ; ; In the formula, Costs for expanding or renovating unit capacity. To accommodate the expansion and renovation required that year, For the total new substation capacity in year i, For the construction capacity of a single substation, the actual construction scale of each 110kV substation is set at 2*50MVA; This refers to the total annual substation capacity calculated based on the per capita substation capacity curve. The cost of constructing new transmission lines is: ; ; In the formula, Cost per unit length of newly constructed transmission line Let the length of the newly added transmission line be the length in year i. Let be the length of the line in year i; This represents the number of substations in the starting year. Station-to-line ratio, The cost of flexible resources; Construct constraints: ; In the formula, Due to capacity constraints, This is the lower limit for the number of new substations added each year. The upper limit for the number of new substations added each year; ; ; In the formula, As a reserve coefficient constraint, The degree of decrease in the reserve coefficient in year i is the same. Let be the substation reserve factor in year i. This represents the percentage of expandable capacity available in the initial year. The total substation capacity in the starting year. To determine the maximum construction scale of substations, the maximum construction scale for each 110kV substation is set at 3*50MVA; ; In the formula, The total number of substations is constrained during the power grid construction period. The total number of substations in the target year.

[0045] Furthermore, according to one embodiment of the present invention, the distribution network planning and construction data includes: distribution equipment ledger data (capacity, parameters, topology, etc. of equipment such as substations, lines, distribution transformers, and distributed power sources), distribution network operation mode and load data (historical load curves, equipment failure rates, power outage durations, and other operational reliability parameters), existing planned project list (project names, types, planning and construction costs, progress, etc. in the three-year distribution network planning database and reserve database, and associated with equipment data to improve planning accuracy), and cost parameters (typical equipment costs, depreciation rates, transmission and distribution prices), etc.

[0046] In a specific embodiment, the model, scenario, and specific steps involved are as follows: (1) Wind turbine output model Nonparametric kernel density estimation is used in probability theory to estimate unknown density functions. It is a method of nonparametric testing that does not rely on the distribution of relevant data and makes no assumptions about the original data distribution. Due to the influence of multiple variable factors such as geographical environment, time, and climate, nonparametric kernel density estimation is more suitable for describing the probability distribution of wind speed. Therefore, this paper chooses nonparametric kernel density estimation to describe the random uncertainty behavior of wind speed v.

[0047] in, Let K be the probability density function, K be the kernel function (usually a Gaussian kernel), and h be a smoothing parameter where h > 0. (i=1,2,...n) are n sample points of independent and identically distributed F, obtained from SCADA or weather stations, representing historical wind speed sequences; The density location is taken from the objects in the planning area.

[0048] The following analysis examines the output power model of the wind turbine. The wind turbine first converts wind energy into mechanical energy; the rotating blades drive the motor, which then converts the mechanical energy into electrical energy. During long-term operation, if transient characteristics are ignored, the active power output of the wind turbine can be considered to be closely related to wind speed. The correspondence between the actual output active power and the cut-in wind speed, cut-out wind speed, and rated wind speed is as follows:

[0049] in, This represents the actual output active power of the wind turbine. This refers to the rated output active power of the fan. The wind speed during the current time period. The rated wind speed of the fan. To cut into wind speed, To cut off the wind speed.

[0050] (2) Photovoltaic power output model The Beta distribution represents the probability distribution of a single random variable. The Beta probability density function is used to describe the random and uncertain behavior of illumination. The Beta probability density function is shown in the following formula.

[0051]

[0052] in, Normalize the entire density function to an integral of 1. , These are the shape parameters of the Beta distribution. This represents the actual light intensity. The rated light intensity is used. The actual output power of a photovoltaic cell is related to the light intensity and its output power under rated conditions. The mathematical model is as follows:

[0053] in, This refers to the actual output power of the photovoltaic cell. This refers to the rated active power of a photovoltaic cell under standard test conditions.

[0054] (3) Energy storage device charging and discharging model The charging and discharging of an energy storage power station is affected by its current state of charge, as well as its rated capacity and rated power. Its SsoE is updated as shown in the formula.

[0055]

[0056] In the formula: This represents the change in energy storage power; a positive value indicates energy storage charging, and a negative value indicates energy storage discharging. Indicates the time interval between energy storage charge and discharge; This indicates the energy storage status during the t-th time period; as well as These represent the energy storage charging efficiency and the discharging efficiency, respectively.

[0057] further It is a set of nodes that can be reinforced for disaster prevention.

[0058] Disaster reinforcement schemes are typically implemented on the basis of the existing distribution network by constructing new reinforced feeders, strengthening interconnections, and adding grid-side mobile energy storage or diesel generators. Therefore, the resilience of the expanded distribution network can be represented by the combination of the resilience of the existing distribution network and the incremental resilience improvement brought by the new components:

[0059] In the formula, and These are the resilience indicators for the expanded distribution network and the existing distribution network, respectively. The increase in system resilience is due to the addition of new components in the expanded planning scheme. Depending on the calculation method, the resilience index can be EENS (Expected Power Loss) or PLC (Probability of Load Failure). In this project, it is the Probability of Load Failure. The above formula can also be used to measure the effectiveness of the expanded distribution network in improving resilience.

[0060] Assuming the expanded distribution network includes m new components, then the states containing only the new components are: Type 2, considering (normal / fault), is represented here as , This represents a system state that only includes newly added components.

[0061] Suppose set A has There are X elements in total (first the power grid minus the decommissioned ones), and set B contains X elements. There are Y elements in total (planning increments). The calculation process for ⊕ is defined as follows:

[0062] Based on the above definitions and derivations, it can be seen that the system state set of the existing power distribution network structure is... and the system state set of the extended distribution network structure The following relationship exists between them:

[0063] like Figure 2 As shown, in order to improve the efficiency of calculating the resilience index of the expanded distribution network, the expanded system state is now divided into the following two scenarios based on the fault conditions of the newly added components: Scenario 1: In this scenario, all newly added components in the system fail. The distribution network resilience index is defined as follows: ; Scenario 2: In this scenario, the system experiences failures only in some newly added components or only in components within the existing distribution network. The distribution network resilience index is defined as follows:

[0064] For Scenario 1, since all newly added components in the system fail, the distribution network structure is exactly the same as the original distribution network structure without expansion planning. Therefore, there is no load reduction in this scenario, and it has no impact on the resilience increment index.

[0065] For Scenario 2, since only some newly added components or only components in the original distribution network fail, the structure of the distribution network is different from the structure before the improvement. Scenario 2 can be further divided into two sub-scenarios based on the resilience increment index: In scenario 2.1, the fault status of the distribution network under this subclass affects the resilience increment.

[0066] In scenario 2.2, the distribution network fault state under this subclass has no impact on the resilience increment.

[0067] 1) Calculation of resilience index for scenario 1 In Scenario 1, all newly added components fail. In this case, the expanded distribution network is essentially identical to the original distribution network structure, containing neither any new components nor any failures of existing components. Therefore, the resilience index for Scenario 1 can be directly obtained from the resilience index of the existing distribution network. (The system state set of the existing distribution network structure is described below.) , 1 ≤ i ≤ X, where si ex represents the system state of the existing distribution network structure, and X represents the total number of system states of the existing distribution network structure. Assume This indicates the system state after the expansion plan, when all newly added components in the distribution network fail, and only includes the newly added components themselves.

[0068] therefore, The calculation can be performed using the following formula: (2.4) Where Ix and Px represent the probability of system state x occurring and the impact of a fault, respectively, N is the set of newly added components in the distribution network, and ue is the unavailability rate of the components. Furthermore, the resilience index of the original distribution network structure can be expressed in the following form: (2.5) Therefore, equation (2.4) can be simplified to: (2.6) From the above formula, we can see that... Regardless of the resilience increment index, no further detailed analysis is needed for the system state in Scenario 1. It can be easily calculated using the existing distribution network structure's resilience index, Rex.

[0069] 2) Calculation of resilience index for scenario 2 In Scenario 2, some of the newly added components may or may not malfunction. In this case, the expanded distribution network structure is divided into two parts: existing components and newly added components, such as... Figure 3 As shown, any fault state p in the expanded distribution network structure corresponds to a fault state q in the existing distribution network structure. In fault states p and q, the existing components that fail are identical; however, newly added components in fault state p may also fail. Therefore, we call fault states p and q "parallel faults." It is important to note that because newly added components may have different states, fault state q in the existing distribution network structure can correspond to multiple "parallel faults" in the expanded distribution network structure.

[0070] For parallel faults p and q, assuming the difference between the impact Ip of the expanded distribution network fault state p and the impact Iq of the existing distribution network fault state q is ΔIp, it can be calculated using the following formula: (2.7) Assumption , 1≤i≤M-1, where for The system status of all newly added components except sM ad. Therefore, the resilience index of Scenario 2. It can be calculated using the following formula: (2.8) Will Substituting the calculation formula into (2.8), we obtain the following form: (2.9) because It is the probability of system state p occurring, and therefore can be rewritten in the following form: (2.10) Wherein, Np normal is the set of newly added components in system state p that are in normal condition, and Np failed is the set of newly added components in system state p that are in fault condition.

[0071] Substituting the system state formula (2.10) into equation (2.9), we get: (2.11) From the above formulas, we can see that... It is divided into two parts: the first part is related to the resilience index of the existing distribution network structure, and the second part is the system resilience increment ΔR. Therefore, ΔR can be expressed as: (2.12) 3) Calculation of resilience index for network-side configuration planning scheme After analyzing scenarios 1 and 2, the extended distribution network structural resilience index The calculation formula is as follows: (2.13) because Therefore, the above formula can be rewritten in the following form: (2.14) Furthermore, due to: (2.15) Therefore, (2.13) can be further simplified to the following form: (2.16) Since ΔR can be calculated from (2.12), (2.16) can be further expressed as: (2.17) At this point, It has been proven that it can be decomposed into It consists of two parts: ΔR and ΔR. Because... The existing resilience index of the distribution network does not require additional calculation. Therefore, only ΔR needs to be calculated to obtain the result quickly and conveniently. In fact, when comparing and selecting different extended-plan distribution network structures, the resilience increment of the extended-plan distribution network structure relative to the existing distribution network structure is more helpful for comparative analysis than the overall index, and helps planners make decisions. Therefore, the method proposed in this invention has important practical application value.

[0073] In practical applications, the solution of the present invention can also selectively combine, for example, the distribution network investment capacity and distribution network investment needs at the regional power grid level to carry out distribution network planning and construction in special areas.

[0074] The investment capacity of the regional power grid-level distribution network can be assessed through, for example, the following: By combining net profit, depreciation, and maximum financing amount into a fitting prediction model, the following fitting prediction model for maximum investment capacity is obtained, which is used to calculate the investment capacity of a provincial region, where the smaller value between maximum financing amount 1 and maximum financing amount 2 is taken: ; In the formula, This is the fitted value of the net profit for the current period; To calculate the fitted value of accumulated depreciation based on the depreciation period; the maximum financing amount 1 is the credit limit obtained by "cash flow multiple (4 times)"; the maximum financing amount 2 is the new financing limit obtained by the asset-liability ratio constraint. The book value of the asset. Outstanding interest-bearing liabilities The maximum allowable debt-to-equity ratio, depending on... Adjustable parameters that vary; External credit enhancements that can be included in the credit line, etc. Adjustable parameters that vary.

[0075] For example, by calculating the city / county and investment capacity assessment, the fitted prediction model for the maximum investment capacity is shown below: ; In the formula, To strengthen decision-making With flexible resource allocation decisions The set of cities and counties with the largest investment capacity is constrained by the provincial-level largest investment capacity.

[0076] The investment demand for power distribution networks can be comprehensively judged using indicators such as the following: A grid-level distribution network investment demand assessment unit is constructed to perform multi-dimensional index fusion calculation of the project demand verification index, the project demand ranking index, and the saturation verification index to obtain the feasibility assessment index of the power grid investment project.

[0077] (1) Project requirement verification indicators The Field method is used to construct project demand verification indicators, which can be adjusted annually based on changes in macroeconomic policies and investment orientation. Based on network structure efficiency, load supply efficiency, and equipment technology efficiency indicators, it is determined whether the provided project list data examples are eligible for investment demand access. Table 2 shows the project demand verification indicators; projects meeting the following conditions are included in investment demand.

[0078] Table 2

[0079] (2) Project requirement ranking indicators Ascertain the current power grid situation in the power supply area, including the current grid structure, current load, power supply area, etc., and estimate the current power supply capacity P1 of the power grid.

[0080] Based on the power supply area planning, determine the 2025 load, the prospective annual load P2, and the five-year load growth rate of the power supply area, and calculate the growth year Y from the current load to the target load.

[0081] Based on the investment verification standard method (unit area cost verification method or unit load cost verification method), the target total investment T1 for the power supply area is calculated. The current power supply capacity achievement rate E (%) is determined by the ratio of the current power supply capacity to the target load level, and the current load achievement rate F (%) is determined by the ratio of the current load level to the target load level. The larger value G (G=F or E) of the power supply capacity achievement rate E (%) and the load achievement rate F (%) is taken. Multiplying (1-G) by the target total investment yields the shortfall investment demand T2.

[0082] Finally, divide the gap investment demand T2 by the growth year Y to obtain the regional average annual investment demand T3.

[0083] The average annual demand investment T3 = the target total investment in the power supply area T1 * (1-G) / the growth year Y from the current load to the prospective saturation load.

[0084] ①If the actual reported annual investment demand for the region is ≤T3, then it is reasonable, and the

Reasonable

[0085] ③ Based on the linkage evaluation of power supply capacity achievement rate and load achievement rate, if the power supply capacity achievement rate is greater than or equal to the load achievement rate, it means that the grid construction is ahead of load development, and the "ahead of the curve" label is output. ④ Based on the linkage evaluation of power supply capacity achievement rate and load achievement rate, if the power supply capacity achievement rate and load achievement rate are basically matched (within 10%), it means that the grid construction and load development are basically matched, and the [matching tag] is output: According to T3, the project demand verification indicators are divided into four quadrants. The [Matching Label] and [Reasonable Label] are given priority, followed by the [Advanced Label] and [Reasonable Label], then the [Matching Label] and [Unreasonable Label], and finally the [Advanced Label] and [Unreasonable Label]. At the same time, the levels are further divided according to the growth year Y from the current load growth to the projected saturation load.

[0086] (3) Project saturation verification index The investment power supply capacity is verified based on the project saturation verification index. When the grid power supply capacity P1 exceeds the saturation load and the investment saturation is greater than 1, the investment saturation analysis and verification are triggered, and projects in the project list are reduced. Among them: Investment saturation = Power supply capacity / Saturation load * 100%.

[0087] Based on the above, a project list is generated under the comprehensive game theory of the aforementioned distribution network investment capacity constraints and grid-level distribution network investment demand evaluation, and manual adjustments are supported: The results from all the modules are summarized to generate the final distribution network resilient planning investment strategy report and project list. Specifically, this includes: a project list arranged by priority (specifying project name, content, investment amount, and corresponding resilient measure categories, etc.), a project implementation timeline (annual distribution), and a comprehensive evaluation of investment benefits. When generating the plan, the output module comprehensively considers regional-level strategy (constraints provided by the investment capability module) and grid-level urgency (priorities provided by the sorting module), essentially achieving a comprehensive optimal game between regional and grid levels.

[0088] Furthermore, this module's interface supports manual adjustments: decision-makers can adjust the order or investment amount of individual projects based on the actual situation of the year, and the system will recalculate the plan indicators in real time to ensure the practicality and compliance of the final output with management requirements. The output module also provides interfaces with external systems for importing project lists and plans into the enterprise's project management or investment control platform, achieving a seamless connection from planning to investment execution.

[0089] As shown above, this invention classifies extreme events into three categories: meteorological, geological, and human-caused, and employs different mathematical models to characterize their probability of occurrence and impact for each category. Meteorological extreme events (such as typhoons and snowstorms) are characterized by their long duration, wide impact range, and relatively high predictability, possessing abundant historical data. Their probability of occurrence and impact on the distribution network can be simulated based on data-driven deterministic probability distributions. Geological events (such as earthquakes and landslides) are sudden and difficult to predict, with limited historical data. However, considering their general regularity, a fuzzy probability model (fuzzy set) is used to fit the impact of all possible fault scenarios, obtaining the expected probability range (lower / upper limit) of this type of disaster. Human-caused events (such as external damage and cyberattacks) are highly random and lack obvious patterns, making them difficult to predict in advance. Therefore, the modeling only considers the typical fault scenario causing the most severe consequences as a representative to assess its maximum potential impact. Through the above classification and modeling, the threat level of different extreme events to the distribution network can be more accurately characterized.

[0090] This invention, based on identified extreme event types and typical scenarios, employs the Analytic Hierarchy Process (AHP) and expert scoring to quantitatively evaluate disaster scenarios and resilient countermeasures from multiple dimensions. Evaluation dimensions include, but are not limited to: event occurrence time (season, time period, etc.), occurrence location (geographical distribution and impact on key nodes), spatiotemporal evolution characteristics (duration, spread range), and pre-disaster resilient measures, pre-disaster reinforcement measures, and post-disaster recovery measures. Based on whether each dimension's elements are effective under specific scenarios, a corresponding 0-1 matrix (i.e., decision matrix) is constructed. Matrix elements are binary, representing whether a certain measure needs to be implemented under a specific disaster scenario. For example, for a "typhoon" scenario, corresponding pre-load transfer plans (pre-disaster resilient measures, "1" indicates implementation), tower reinforcement plans (pre-disaster reinforcement, "1" indicates implementation), and emergency repair plans (post-disaster recovery measures) can be formulated, thus forming a resilient decision vector under the typhoon scenario. By constructing this matrix exhaustively covering major extreme event scenarios, a qualitative strategy combination library is provided for subsequent planning.

[0091] This invention integrates the aforementioned 0-1 resilient decision matrix into a multi-scenario planning model for distribution networks (a resilient planning model for distribution networks under extreme events), comprehensively considering the impact of various disaster scenarios on distribution network planning schemes. The model defines a set of disaster scenarios, each corresponding to an extreme event and a combination of corresponding resilient measures. By constraining the activation status of measures for the corresponding scenario in the 0-1 matrix, the simulation can demonstrate which resilient resources are deployed, which reinforcement measures are implemented, and how post-disaster recovery proceeds under that scenario. This planning model aims to improve the overall resilience of the distribution network, balancing multiple factors such as normal operating costs, disaster losses, and resilient enhancement investments. Through multi-scenario analysis, the impact of different resilient strategies on the reliability and economy of the distribution network under various disaster scenarios can be evaluated, thereby selecting planning schemes that exhibit high robustness under all possible events. This step essentially establishes an extreme event-driven resilient distribution network planning model.

[0092] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0093] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0094] Figure 4 A schematic diagram of a distribution network resilient planning device considering extreme events, provided in an embodiment of the present invention, is shown. For ease of explanation, only the parts relevant to the embodiment of the present invention are shown, and are described in detail below: like Figure 4As shown, the distribution network resilience planning device 4, which considers extreme events, includes: The fitting module 41 is used to fit the expected impact of extreme events in the target distribution network area based on historical data of extreme events; among which, extreme events include meteorological extreme events, geological extreme events and human-induced extreme events; Module 42 is used to take the topology of the expected and target distribution network as input, and the hardening decision and flexible resource allocation as output. It constructs an objective function based on the node hardening decision cost, flexible resource allocation cost, disaster loss cost and operating cost, and constructs constraints based on the upper and lower limits of the expected probability, the distribution network resilience limit and extreme event measures, to obtain the flexible planning model of the target distribution network. The solver module 43 is used to solve the elastic programming model with the objective function minimization as the optimization objective, and obtain the elastic programming results of the target distribution network.

[0095] In one possible implementation, the fitting module 41 is specifically used for: Based on historical data of meteorological extreme events, a data-driven deterministic probability distribution is used to fit the expected impact of meteorological extreme events in the target distribution network area.

[0096] In one possible implementation, the fitting module 41 is specifically used for: Based on historical data of geological extreme events, a fuzzy probability model is used to fit the expected impact of geological extreme events in the target distribution network area.

[0097] In one possible implementation, the fitting module 41 is specifically used for: For each type of man-made extreme event, the impact of the failure scenario that causes the most severe loss is taken as the expected impact of that type of man-made extreme event.

[0098] In one possible implementation, the objective function is:

[0099] in, To strengthen decision-making With flexible resource allocation decisions The set, The set of feasible solutions. For nodes The cost of reinforcement measures For nodes The cost of elastic resources, This is an integer variable; a value of 1 indicates a node. Strengthening, 0 indicates a node Without reinforcement, This is an integer variable; a value of 1 indicates a node. Configure elastic resources; a value of 0 indicates a node. Without configuring elastic resources, For disaster scenarios, A collection of disaster scenarios. For a set of configurable nodes for elastic resources, For a set of nodes, For disaster The cost of losses incurred after implementing response and recovery measures in a given scenario. For the set of fault uncertainties, Cost per unit of energy output for new energy sources (superscript 0 means t=0, initial annual investment). For the node Scene Time period Renewable power generation capacity (such as wind power and solar power output). The length of a single discrete time step.

[0100] In one possible implementation, building module 42 is specifically used for: For each type of extreme event, the analytic hierarchy process and expert scoring method are used to classify disaster scenarios according to the event occurrence time, location, spatiotemporal evolution, pre-disaster resilience measures, pre-disaster reinforcement measures, and post-disaster measures. It is then used to determine whether measures need to be implemented when each type of extreme event occurs under each disaster scenario, resulting in a 0-1 matrix of extreme event measures.

[0101] In one possible implementation, the solver module 43 is also used for: After solving the elastic programming model with the objective function as the optimization objective, if the elastic programming result of the target distribution network cannot meet the resource regulation needs, then the objective function is to minimize the sum of the cost of newly built substations, the cost of capacity expansion and renovation, and the cost of newly built transmission lines in the current year. With construction capacity constraints and reserve coefficient constraints as constraints, the strategy for the construction and expansion of power grid equipment in the target distribution network is solved and used as a supplementary regulation method.

[0102] This invention, through classifying extreme event types and fitting the expected impact of extreme events based on historical data, transforms elusive disaster risks into quantifiable and calculable mathematical parameters, providing a solid scientific basis for planning decisions. It innovatively incorporates node reinforcement costs, flexible resource allocation costs, potential disaster loss costs, and operating costs into a scientific flexible planning model, automatically seeking the optimal balance between investment costs and system resilience. This effectively solves the problems of inaccurate extreme event assessments and lack of systematic solutions in traditional planning. This solution achieves precise quantification of the impact of extreme events, providing a reliable basis for planning decisions. Furthermore, through multi-cost trade-offs and constraint control, it minimizes the total lifecycle cost while improving the distribution network's resilience and recovery speed in the face of extreme events, balancing flexibility needs with economic efficiency. This provides strong protection for the safe and stable operation of the distribution network and has significant engineering application value.

[0103] Figure 5 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 5 As shown, the electronic device 5 of this embodiment includes a processor 50 and a memory 51. The memory 51 stores a computer program 52. When the processor 50 executes the computer program 52, it implements the steps in the various method embodiments described above. Alternatively, when the processor 50 executes the computer program 52, it implements the functions of each module / unit in the various device embodiments described above.

[0104] For example, computer program 52 may be divided into one or more modules / units, which are stored in memory 51 and executed by processor 50 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 52 in electronic device 5.

[0105] Electronic device 5 may include, but is not limited to, processor 50 and memory 51. Those skilled in the art will understand that... Figure 5 This is merely an example of electronic device 5 and does not constitute a limitation on electronic device 5. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 5 may also include input / output devices, network access devices, buses, etc.

[0106] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0107] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0108] The above-described 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 the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A distribution network resilient planning method considering extreme events, characterized in that, include: Based on historical data of extreme events, the expected impact of extreme events in the target distribution network area is fitted; among them, extreme events include meteorological extreme events, geological extreme events, and human-induced extreme events; Using the expected impact and the topology of the target distribution network as inputs, and reinforcement decisions and flexible resource allocation as outputs, an objective function is constructed based on the node reinforcement decision cost, flexible resource allocation cost, disaster loss cost, and operating cost. Constraints are constructed based on the upper and lower limits of expected probability, distribution network resilience limits, and extreme event measures to obtain the flexible planning model of the target distribution network. The elastic programming model is solved with the objective function minimization as the optimization objective, and the elastic programming results of the target distribution network are obtained.

2. The distribution network resilient planning method considering extreme events according to claim 1, characterized in that, The method of fitting the expected impact of extreme events in the target distribution network area based on historical extreme event data includes: Based on historical data of meteorological extreme events, a data-driven deterministic probability distribution is used to fit the expected impact of meteorological extreme events in the target distribution network area.

3. The distribution network resilient planning method considering extreme events according to claim 1, characterized in that, The method of fitting the expected impact of extreme events in the target distribution network area based on historical extreme event data includes: Based on historical data of geological extreme events, a fuzzy probability model is used to fit the expected impact of geological extreme events in the target distribution network area.

4. The distribution network resilient planning method considering extreme events according to claim 1, characterized in that, The fitting of extreme events in the target distribution network area based on historical extreme event data includes: For each type of man-made extreme event, the impact of the failure scenario that causes the most severe loss is taken as the expected impact of that type of man-made extreme event.

5. The distribution network resilient planning method considering extreme events according to claim 1, characterized in that, The objective function is: in, To strengthen decision-making With flexible resource allocation decisions The set, The set of feasible solutions. For nodes The cost of reinforcement measures For nodes The cost of elastic resources, This is an integer variable; a value of 1 indicates a node. Strengthening, 0 indicates a node Without reinforcement, This is an integer variable; a value of 1 indicates a node. Configure elastic resources; a value of 0 indicates a node. Without configuring elastic resources, For disaster scenarios, A collection of disaster scenarios. For a set of configurable nodes for elastic resources, For a set of nodes, For disaster The cost of losses incurred after implementing response and recovery measures in a given scenario. For the set of fault uncertainties, Cost per unit of energy output for new energy sources (superscript 0 means t=0, initial annual investment). For the node Scene Time period Renewable power generation capacity (such as wind power and solar power output). The length of a single discrete time step.

6. The distribution network resilient planning method considering extreme events according to claim 1, characterized in that, Constraints are constructed based on measures for extreme events, including: For each type of extreme event, the analytic hierarchy process and expert scoring method are used to classify disaster scenarios according to the event occurrence time, location, spatiotemporal evolution, pre-disaster resilience measures, pre-disaster reinforcement measures, and post-disaster measures. It is then used to determine whether measures need to be implemented when each type of extreme event occurs under each disaster scenario, resulting in a 0-1 matrix of extreme event measures.

7. The distribution network resilient planning method considering extreme events according to any one of claims 1 to 6, characterized in that, After solving the elastic programming model with the objective function as the optimization objective, the method further includes: If the elastic planning results of the target distribution network cannot meet the resource regulation needs, then the objective function is to minimize the sum of the costs of newly built substations, capacity expansion and renovation, and new transmission lines in the current year. With construction capacity constraints and reserve coefficient constraints as constraints, the strategy for the construction and expansion of power grid equipment in the target distribution network is solved and used as a supplementary regulation method.

8. A distribution network resilient planning device considering extreme events, characterized in that, include: The fitting module is used to fit the expected impact of extreme events in the target distribution network area based on historical data of extreme events; among which, extreme events include meteorological extreme events, geological extreme events, and human-induced extreme events; The construction module is used to take the expected impact and the topology of the target distribution network as inputs, and the hardening decision and flexible resource allocation as outputs. It constructs an objective function based on the node hardening decision cost, flexible resource allocation cost, disaster loss cost and operating cost, and constructs constraints based on the upper and lower limits of expected probability, distribution network resilience limit and extreme event measures to obtain the flexible planning model of the target distribution network. The solution module is used to solve the elastic planning model with the objective function minimization as the optimization objective, and obtain the elastic planning result of the target distribution network.

9. The distribution network flexible planning device considering extreme events according to claim 8, characterized in that, The fitting module is specifically used for: Based on historical data of meteorological extreme events, a data-driven deterministic probability distribution is used to fit the expected impact of meteorological extreme events in the target distribution network area.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 7.

Images