Power distribution network collaborative power recovery method and system based on post-disaster

Abstract

The invention discloses a power distribution network cooperative power restoration method and system based on post-disaster, and the method comprises the steps: obtaining real-time supply and demand condition data and network topology constraint information, analyzing and extracting a key load node set and a power flow distribution boundary condition, so as to construct a resource distribution model, fusing power flow analysis and a multi-target optimization method to obtain an optimal distribution strategy; and determining a temporary power supply network configuration and capacity distribution scheme by combining temporary equipment access parameters, and further planning a deployment path of the temporary power supply configuration so as to deploy a temporary power supply network to perform collaborative power restoration on the power distribution network. The method can quickly evaluate the supply and demand conditions in the post-disaster environment, optimize resource allocation, ensure key load power supply, improve the recovery efficiency and stability of the power distribution network, and provide an effective solution for quick recovery of a post-disaster power system.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method and system for coordinated power restoration of a distribution network after a disaster. Background Art

[0002] As the lifeblood of modern society, the stability and resilience of the power system under extreme disasters are directly related to social security and sustainable economic development. Especially in areas with complex geographical environments and limited resources such as Haichengshan Island, the distribution network must not only cope with the physical damage caused by natural disasters, but also meet the urgent needs of post-disaster power supply. However, extreme disasters often cause large-scale damage to power infrastructure, and the recovery method of the distribution network is particularly critical in terms of efficiency and adaptability. How to quickly restore power supply in the case of scarce resources and ensure the basic operation of society has become a key area of ​​research.

[0003] At present, the recovery of distribution networks after disasters mainly relies on preset static scheduling schemes or single-objective optimization strategies. However, these methods are difficult to adapt to dynamically changing post-disaster scenarios. For example, static schemes are difficult to adapt to real-time supply and demand fluctuations, while single-objective optimization tends to ignore load differences or network topology constraints, resulting in insufficient power supply to key users or uneven resource allocation. In addition, the lack of systematic coordination and management of the access of mobile power sources and emergency equipment has exacerbated the instability of temporary power supply networks. This significantly restricts the rational allocation of power resources and system recovery efficiency after disasters. Therefore, how to dynamically optimize the allocation ratio of power resources based on real-time supply and demand conditions and network topology constraints after extreme disasters, while achieving critical load protection, maximization of social benefits, and stable operation of temporary power supply networks, has become a key issue in the collaborative recovery method of the Haichengshan Island distribution network after disasters. Summary of the invention

[0004] The present invention provides a method and system for coordinated power restoration of a distribution network after a disaster. Based on real-time supply and demand conditions and network topology constraints, the method dynamically optimizes the allocation ratio of power resources, and aims at key load protection, maximizing social benefits, and stable operation of a temporary power supply network. The method can quickly evaluate the supply and demand conditions in a post-disaster environment, optimize resource allocation, ensure power supply to key loads, and improve the efficiency and stability of distribution network restoration, thus providing an effective solution for the rapid restoration of the power system after a disaster.

[0005] In order to achieve the above object, an embodiment of the present invention provides a method for coordinated power restoration of a distribution network after a disaster, comprising: Acquire real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, extract and analyze the real-time supply and demand data, and determine the current load rating parameters of the distribution network; wherein the real-time supply and demand data includes load demand data and power supply data; According to the load rating parameters, load historical data and user types, the priority weight of each load node is calculated to determine the key load node set of the distribution network; according to the network topology information, the feasible path of the power flow of the distribution network is calculated, and the power flow distribution boundary conditions of the distribution network are determined; Constructing a resource allocation model according to the flow distribution boundary conditions and the key load node set, and using the resource allocation model to determine a first resource allocation strategy through the real-time supply and demand data; A multi-objective optimization model is constructed according to resource allocation efficiency and social benefits, as well as the boundary conditions of the power flow distribution; the first resource allocation strategy is optimized using the multi-objective optimization model to obtain a second resource allocation strategy; wherein the multi-objective optimization model includes a comprehensive objective function and objective constraints; Determining a temporary power supply configuration of the distribution network according to the power flow distribution data in the second resource allocation strategy, the priority weight and the access parameters of the temporary equipment; Determine the power supply redundancy of each key load node according to the resource allocation ratio in the second resource allocation strategy and the output power of the temporary power supply configuration, and obtain the capacity allocation strategy of the distribution network; According to the power flow distribution data and voltage stability constraint parameters in the second resource allocation strategy, as well as the information of the temporary power supply configuration, a deployment path of the temporary power supply configuration is planned to coordinate power restoration of the distribution network.

[0006] As an improvement of the above solution, after planning the deployment path of the temporary power supply configuration, the method further includes: According to the capacity allocation plan and the temporary power supply configuration, the current load demand data is analyzed to obtain key load guarantee rate and network stability index, and the key load guarantee rate and network stability index are analyzed by simulation algorithm to obtain the restoration efficiency evaluation result; According to the recovery efficiency evaluation result, updating the load rating parameter and the resource allocation ratio to adjust the resource allocation model and generate a third resource allocation strategy; According to the third resource allocation strategy, a deployment path of the adjusted temporary power supply configuration is planned to coordinate power restoration of the distribution network.

[0007] As an improvement of the above scheme, the real-time supply and demand data and network topology information of the post-disaster distribution network in the target area are obtained, the real-time supply and demand data are extracted and analyzed, and the current load rating parameters of the distribution network are determined, including: Obtain the real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, use the time series analysis method to analyze the change trend of the real-time supply and demand data, and obtain the load fluctuation characteristics and the corresponding power supply status; Calculating the matching deviation between the load fluctuation characteristics and the corresponding power supply state, determining the current supply and demand imbalance area and degree of the distribution network, so as to obtain the current supply and demand matching deviation distribution of the distribution network; According to the supply-demand matching deviation distribution, a current load rating parameter of the distribution network is determined.

[0008] As an improvement of the above scheme, the priority weight of each load node is calculated according to the load rating parameters, load historical data and user type to determine the key load node set of the distribution network; the feasible path of the power flow of the distribution network is calculated according to the network topology information, and the flow distribution boundary conditions of the distribution network are determined, including: Calculate the priority weight of each load node according to the load rating parameters, load historical data and user type; All load nodes corresponding to the priority weight being greater than a preset load weight threshold are taken as a set of key load nodes of the distribution network; Calculating a feasible path of the power flow of the distribution network through a topology analysis algorithm according to the line connection status and damaged node location information in the network topology information; According to the feasible path, the boundary constraint conditions of the power flow are determined, and according to the boundary constraint conditions and the line connection status, the boundary conditions of the power flow distribution of the distribution network are obtained.

[0009] As an improvement of the above solution, the resource allocation model is constructed according to the flow distribution boundary conditions and the key load node set, and the resource allocation model is adopted to determine the first resource allocation strategy through the real-time supply and demand data, including: Constructing a resource allocation model according to the flow distribution boundary conditions and the key load node set; wherein the resource allocation model includes an allocation objective function and allocation constraints; The real-time supply and demand data is input into the resource allocation model, and according to the real-time supply and demand data, the allocation objective function and the allocation constraints, an optimization algorithm is used to iteratively adjust the resource allocation ratio between each load node until the resource allocation model converges, thereby obtaining a first resource allocation strategy.

[0010] As an improvement of the above scheme, a multi-objective optimization model is constructed according to the resource allocation efficiency and social benefits, as well as the flow distribution boundary conditions; and the first resource allocation strategy is optimized by using the multi-objective optimization model to obtain a second resource allocation strategy, including: Constructing a comprehensive objective function of a multi-objective optimization model according to resource allocation efficiency and social benefits, and establishing objective constraint conditions of the multi-objective optimization model according to the flow distribution boundary conditions and actual operation constraints; According to the real-time supply and demand data, the comprehensive objective function and the objective constraints, an optimization algorithm is used to iteratively adjust the first resource allocation strategy until the multi-objective optimization model converges to obtain a second resource allocation strategy.

[0011] As an improvement of the above scheme, the current load demand data is analyzed according to the capacity allocation scheme and the temporary power supply configuration to obtain key load guarantee rate and network stability index, and the key load guarantee rate and network stability index are analyzed by simulation algorithm to obtain the recovery efficiency evaluation result, including: Calculate each critical load guarantee rate according to the load demand data by using the allocated capacity of the capacity allocation strategy and the power supply capacity of the temporary power supply configuration; According to the load demand data, the network stability index is evaluated through the network topology and power transmission conditions of the temporary power supply configuration; The critical load guarantee rate and the network stability index are analyzed by a simulation algorithm to obtain a restoration efficiency evaluation result.

[0012] As an improvement of the above solution, the load rating parameter and the resource allocation ratio are updated according to the recovery efficiency evaluation result to adjust the resource allocation model and generate a third resource allocation strategy, including: Obtaining deviation feedback data according to the recovery efficiency evaluation result, analyzing the deviation feedback data, and obtaining distribution characteristics of the deviation feedback data; Update the load rating parameter according to the distribution characteristics, optimize the resource allocation ratio according to the updated load rating parameter, and obtain an optimized allocation parameter; The resource allocation model is iteratively adjusted according to the optimized allocation parameters to generate a third resource allocation strategy.

[0013] In order to achieve the above object, an embodiment of the present invention provides a coordinated power restoration system for a distribution network after a disaster, comprising: A load rating parameter determination module is used to obtain real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, extract and analyze the real-time supply and demand data, and determine the current load rating parameters of the distribution network; wherein the real-time supply and demand data includes load demand data and power supply data; A distribution boundary condition determination module is used to calculate the priority weight of each load node according to the load rating parameters, load historical data and user type, so as to determine the key load node set of the distribution network; calculate the feasible path of the power flow of the distribution network according to the network topology information, and determine the power flow distribution boundary conditions of the distribution network; A first allocation strategy determination module is used to construct a resource allocation model according to the power flow distribution boundary conditions and the key load node set, and adopt the resource allocation model to determine a first resource allocation strategy through the real-time supply and demand data; A second allocation strategy determination module is used to construct a multi-objective optimization model according to resource allocation efficiency and social benefits, as well as the boundary conditions of the power flow distribution; the multi-objective optimization model is used to optimize the first resource allocation strategy to obtain a second resource allocation strategy; wherein the multi-objective optimization model includes a comprehensive objective function and objective constraints; A temporary power supply configuration determination module, configured to determine the temporary power supply configuration of the distribution network according to the power flow distribution data in the second resource allocation strategy, the priority weight and the access parameters of the temporary equipment; A capacity allocation strategy determination module, configured to determine the power supply redundancy of each key load node according to the resource allocation ratio in the second resource allocation strategy and the output power of the temporary power supply configuration, and obtain the capacity allocation strategy of the distribution network; The first deployment path planning module is used to plan the deployment path of the temporary power supply configuration according to the power flow distribution data and voltage stability constraint parameters in the second resource allocation strategy and the information of the temporary power supply configuration, so as to coordinate the restoration of power to the distribution network.

[0014] As an improvement of the above solution, the distribution network coordinated power restoration system after a disaster further includes: An efficiency evaluation result acquisition module is used to analyze the current load demand data according to the capacity allocation plan and the temporary power supply configuration to obtain key load guarantee rate and network stability index, and analyze the key load guarantee rate and network stability index through a simulation algorithm to obtain a recovery efficiency evaluation result; a third allocation strategy generating module, configured to update the load rating parameter and the resource allocation ratio according to the recovery efficiency evaluation result, so as to adjust the resource allocation model and generate a third resource allocation strategy; The second deployment path planning module is used to plan the deployment path of the adjusted temporary power supply configuration according to the third resource allocation strategy to coordinate power restoration of the distribution network.

[0015] Compared with the prior art, the embodiment of the present invention discloses a method and system for coordinated power restoration of distribution networks after disasters. By acquiring real-time supply and demand status data and network topology constraint information, analyzing and extracting key load node sets and flow distribution boundary conditions, a resource allocation model is constructed, and an optimized allocation strategy is obtained by integrating power flow analysis and multi-objective optimization methods. Combined with temporary equipment access parameters, the temporary power supply network configuration and capacity allocation plan are determined, and then the deployment path of the temporary power supply configuration is planned to deploy a temporary power supply network to coordinate power restoration of the distribution network. It can quickly evaluate the supply and demand status in a post-disaster environment, optimize resource allocation, ensure power supply to key loads, improve the efficiency and stability of distribution network restoration, and provide an effective solution for the rapid recovery of power systems after disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of a method for coordinated power restoration of a distribution network after a disaster provided by an embodiment of the present invention; Figure 2 It is a structural diagram of a distribution network collaborative power restoration system based on post-disaster power distribution provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0018] It should be noted that the terms "comprises" and "specifically" and any variations of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0019] See also Figure 1 , Figure 1 1 is a flow chart of a method for coordinated power restoration of a distribution network after a disaster provided by an embodiment of the present invention. The method for coordinated power restoration of a distribution network after a disaster includes: S1, obtaining real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, extracting and analyzing the real-time supply and demand data, and determining the current load rating parameters of the distribution network; wherein the real-time supply and demand data includes load demand data and power supply data; S2, calculating the priority weight of each load node according to the load rating parameters, load historical data and user type to determine the key load node set of the distribution network; calculating the feasible path of the power flow of the distribution network according to the network topology information, and determining the power flow distribution boundary conditions of the distribution network; S3, constructing a resource allocation model according to the power flow distribution boundary conditions and the set of key load nodes, and using the resource allocation model to determine a first resource allocation strategy through the real-time supply and demand data; S4, constructing a multi-objective optimization model according to the resource allocation efficiency and social benefits, and the flow distribution boundary conditions; optimizing the first resource allocation strategy using the multi-objective optimization model to obtain a second resource allocation strategy; wherein the multi-objective optimization model includes a comprehensive objective function and objective constraints; S5, determining a temporary power supply configuration of the distribution network according to the power flow distribution data in the second resource allocation strategy, the priority weight and the access parameters of the temporary equipment; S6, determining the power supply redundancy of each key load node according to the resource allocation ratio in the second resource allocation strategy and the output power of the temporary power supply configuration, and obtaining the capacity allocation strategy of the distribution network; S7, planning a deployment path of the temporary power supply configuration according to the power flow distribution data and voltage stability constraint parameters in the second resource allocation strategy, and the information of the temporary power supply configuration, so as to coordinate power restoration of the distribution network.

[0020] Exemplarily, the method for coordinated power restoration of a distribution network after a disaster described in an embodiment of the present invention is implemented by a distribution network power restoration server, and the distribution network power restoration server can exchange information with target users. The distribution network repowering server obtains real-time supply and demand data (load demand data and power supply data) through sensors and smart meters, and obtains network topology information (line connection status of the distribution network and damaged node location) through the distribution network geographic information system; uses time series analysis method to extract and analyze the real-time supply and demand data to determine the current load rating parameters of the distribution network; calculates the priority weight of each load node according to the load rating parameters, load history data and user type (such as industrial users, commercial users, residential users), and takes all load nodes corresponding to the priority weight greater than the preset load weight threshold as the key load node set of the distribution network. It can be understood that the load weight threshold can be set as needed; calculates the feasible path of the power flow of the distribution network according to the network topology information, and determines the flow distribution boundary conditions of the distribution network; determines the allocation objective function and allocation constraints according to the flow distribution boundary conditions and the key load node set to construct a resource allocation model; iteratively adjusts the parameters of the resource allocation model through real-time supply and demand data until When the resource allocation model converges, a first resource allocation strategy is obtained; a comprehensive objective function of a multi-objective optimization model is constructed according to resource allocation efficiency and social benefits, and the objective constraint of the multi-objective optimization model is established according to the flow distribution boundary conditions and actual operation constraints; according to the real-time supply and demand data, the comprehensive objective function and the objective constraint, the first resource allocation strategy is iteratively adjusted by an optimization algorithm until the multi-objective optimization model converges, and a second resource allocation strategy is obtained; according to the power flow distribution data, the priority weight and the access parameters of the temporary equipment in the second resource allocation strategy, the temporary power supply configuration of the distribution network is determined; according to the resource allocation ratio in the second resource allocation strategy and the output power of the temporary power supply configuration, the power supply redundancy of each key load node is determined to obtain the capacity allocation strategy of the distribution network; according to the power flow distribution data and the voltage stability constraint parameters in the second resource allocation strategy, and the information of the temporary power supply configuration, the deployment path of the temporary power supply configuration is planned to coordinate the power restoration of the distribution network. The embodiment of the present invention can quickly evaluate the supply and demand situation in a post-disaster environment, optimize resource allocation, ensure the power supply of key loads, improve the recovery efficiency and stability of the distribution network, and provide an effective solution for the rapid recovery of the power system after the disaster.

[0021] Further, after planning the deployment path of the temporary power supply configuration, the method further includes: S8, analyzing the current load demand data according to the capacity allocation plan and the temporary power supply configuration to obtain key load guarantee rate and network stability index, and analyzing the key load guarantee rate and network stability index through a simulation algorithm to obtain a restoration efficiency evaluation result; S9, updating the load rating parameter and the resource allocation ratio according to the recovery efficiency evaluation result to adjust the resource allocation model and generate a third resource allocation strategy; S10: According to the third resource allocation strategy, planning a deployment path of the adjusted temporary power supply configuration to coordinate power restoration of the distribution network.

[0022] Specifically, the step S1 includes: S11, obtaining real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, using a time series analysis method to perform a change trend analysis on the real-time supply and demand data, and obtaining load fluctuation characteristics and corresponding power supply status; S12, calculating the matching deviation between the load fluctuation characteristics and the corresponding power supply state, determining the current supply and demand imbalance area and degree of the distribution network, so as to obtain the current supply and demand matching deviation distribution of the distribution network; S13: Determine the current load rating parameter of the distribution network according to the supply-demand matching deviation distribution.

[0023] Specifically, in step S11, real-time supply and demand data are obtained through sensors and smart meters, load demand data and power supply data are extracted, and an initial data set is obtained. The initial data set is processed using a time series analysis method to analyze the changing trend of load demand during the fluctuation time, and load fluctuation characteristics and corresponding power supply status are obtained.

[0024] Specifically, in step S12, according to the comparison between the load fluctuation characteristics and the power supply status, the matching deviation between the load fluctuation characteristics and the corresponding power supply status is calculated, and the imbalance area and the degree of imbalance are determined. The matching deviation is processed by a clustering algorithm to generate a matching deviation distribution map to obtain the spatial distribution of the supply and demand imbalance. If the degree of imbalance exceeds the preset threshold, the frequency of obtaining real-time supply and demand data is adjusted according to the deviation distribution in the imbalance area to obtain high-frequency real-time supply and demand data. According to the high-frequency real-time supply and demand data update time series analysis, it is determined whether the load demand change tends to be stable, and the supply and demand matching optimization result is obtained. The dynamic change trend of the final imbalance area is determined by the spatial distribution of the deviation distribution and the supply and demand matching optimization result.

[0025] For example, in the real-time monitoring of the distribution network after the disaster, the load demand and power supply data are first collected in real time through sensors and smart meters deployed at key nodes. For example, the load demand recorded by the smart meters in a certain area within 15 minutes is 1200kW, 1250kW, and 1300kW, respectively, while the power supply data is 1100kW, 1150kW, and 1200kW. These data are modeled and predicted using time series analysis methods, such as the autoregressive integrated moving average model (ARIMA). Assume that the prediction results of the ARIMA model show that the load demand will increase to 1350kW in the next 30 minutes, while the power supply can only be maintained at 1220kW. By calculating the supply and demand deviation, it is found that the current supply and demand imbalance area is the downstream area of ​​a certain substation, and the deviation value is 130kW. Using a spatial interpolation algorithm (such as the Kriging interpolation method), the deviation value is distributed to the entire distribution network to generate a supply and demand matching deviation distribution map. The deviation distribution diagram shows that the deviation value of the downstream area of ​​a substation is 130kW, while the deviation values ​​of other areas fluctuate within the range of ±50kW. Based on these analysis results, the power dispatch strategy can be adjusted in real time, such as by starting the backup power supply or adjusting the load distribution, to alleviate the imbalance between supply and demand.

[0026] Specifically, in step S13, the supply-demand matching deviation is used as the current load rating parameter of the distribution network.

[0027] Specifically, the step S2 includes: S21, calculating the priority weight of each load node according to the load rating parameter, load historical data and user type; S22, taking all load nodes corresponding to the priority weight being greater than a preset load weight threshold as a set of key load nodes of the distribution network; S23, calculating a feasible path of the power flow of the distribution network through a topology analysis algorithm according to the line connection status and damaged node location information in the network topology information; S24, determining the boundary constraint conditions of the power flow according to the feasible path, and obtaining the power flow distribution boundary conditions of the distribution network according to the boundary constraint conditions and the line connection status.

[0028] Exemplarily, the priority weight of each load node is calculated using load rating parameters, load historical data and user type to obtain a preliminary weight distribution. If the preliminary weight distribution exceeds a preset threshold, it is marked as a critical load, and the preliminary weight distribution is processed by a clustering algorithm to obtain a set of critical load nodes. The operating status data of the load nodes are extracted according to the set of critical load nodes, and the priority weights are updated using time series analysis to obtain a dynamic weight distribution. By comparing the dynamic weight distribution with the supply and demand matching deviation distribution, it is determined whether the load node is in an unbalanced state to obtain an unbalanced node identifier. The data collection frequency is adjusted for the unbalanced node identifier to obtain high-frequency operating status data and obtain an updated load node status set. The priority weight is recalculated based on the updated load node status set, and the weight change trend is verified by comparing historical data to obtain an optimized set of critical load nodes.

[0029] In the distribution network, the priority weight of each load node can be calculated by combining historical data and user types through the analytic hierarchy process (AHP). For example, the historical load data of a certain area shows that the annual electricity consumption of industrial load nodes is 5000MWh, commercial load nodes is 3000MWh, and residential load nodes is 2000MWh. Through the AHP algorithm, the weights of industrial, commercial and residential loads are set to 5, 3 and 2 respectively. According to the user type, the power outage loss factor is introduced, and the loss factors of industrial, commercial and residential loads are 10, 5 and 1 respectively. Through weighted calculation, the priority weight of industrial load nodes is 5×10=50, that of commercial load nodes is 3×5=15, and that of residential load nodes is 2×1=2. If the preset threshold is 3, the industrial load nodes and commercial load nodes are marked as critical loads, forming a set of critical load nodes. In order to optimize the priority weight calculation, the fuzzy comprehensive evaluation method can be introduced to correct the weights by combining multi-dimensional indicators such as the geographical location of the load nodes, power supply reliability and user importance. For example, if an industrial load node is located in the core area of ​​a city and has high power supply reliability requirements, the fuzzy comprehensive evaluation method can be used to increase its weight from 5 to 6. Finally, the key load node set can be applied to optimize the power dispatching strategy to ensure the power supply reliability of the core load.

[0030] Exemplarily, the network topology data stored in the distribution network geographic information system is obtained, the line connection status and the damaged node location information are extracted, and the initial topology description data is obtained. The initial topology description data is processed by a topology analysis algorithm, and the feasible path set of the power flow is calculated to obtain the path distribution data. The boundary constraints of the power flow are extracted through the path distribution data, and the power flow distribution boundary data is obtained in combination with the line connection status. If the power flow distribution boundary data exceeds the preset threshold, the damaged node is marked as a high-risk node to obtain a high-risk node set. According to the high-risk node set, the real-time status data of the corresponding node is extracted, and the path distribution data is updated by time series analysis to obtain dynamic path distribution data. By comparing the dynamic path distribution data with the initial topology description data, it is determined whether the damaged node affects the power flow stability, and the stability identification data is obtained. The power flow calculation parameters are adjusted according to the stability identification data, and the dynamic path distribution data is processed by the support vector machine algorithm to obtain the optimized power flow distribution boundary data.

[0031] For example, the GIS data of a distribution network area shows that the total length of the line is 150 kilometers, of which the length of the damaged line is 10 kilometers, mainly concentrated between nodes A and B. The network topology is analyzed using the depth-first search algorithm (DFS), and it is found that a feasible power flow path is formed between nodes C, D and E, with a maximum transmission capacity of 50MW. Through the power flow calculation model, combined with the line impedance parameters, the power flow distribution from node C to node E is calculated to be 35MW, and the power flow distribution from node D to node E is 15MW, which meets the line capacity constraint. The minimum spanning tree algorithm (Prim algorithm) in graph theory is used to optimize the network, generate a new path, include node F in the power flow distribution range, and increase the power flow distribution from node F to node E to 20MW. By introducing a linear programming model, combined with the line load rate and node voltage stability, the boundary conditions of the power flow distribution are optimized to ensure that the power flow distribution from node A to node B does not exceed 30MW, and the power flow distribution from node C to node E is stable within 40MW. Finally, the optimized power flow distribution boundary conditions are input into the power dispatching system to provide data support for subsequent power dispatching decisions.

[0032] Specifically, the step S3 includes: S31, constructing a resource allocation model according to the power flow distribution boundary conditions and the key load node set; wherein the resource allocation model includes an allocation objective function and allocation constraints; S32, inputting the real-time supply and demand data into the resource allocation model, and adopting an optimization algorithm to iteratively adjust the resource allocation ratio between each load node according to the real-time supply and demand data, the allocation objective function and the allocation constraint conditions, until the resource allocation model converges, thereby obtaining a first resource allocation strategy.

[0033] For example, the flow distribution boundary conditions and the key load node set are used as the core input parameters of the model. The flow distribution boundary conditions include information such as the feasible paths of the power flow, the transmission capacity limits of each path, and the voltage stability constraints. These conditions limit the transmission range and capacity of power in the distribution network. The key load node set clarifies the load nodes that need to be focused on in the resource allocation process, and its priority weight reflects the importance of different nodes. The objective function of the resource allocation model is set according to actual needs, usually with the goal of maximizing the power supply guarantee rate of key loads, maximizing social benefits, and minimizing power transmission losses. On the premise of meeting the power supply needs of key loads, the operating efficiency and economic benefits of the entire distribution network are improved as much as possible, and unnecessary energy waste is reduced. In addition to the constraints in the flow distribution boundary conditions, other constraints need to be considered, such as total power supply constraints, power limits of each line, equipment capacity limits, etc. Ensure that in the resource allocation process, there will be no unreasonable situations such as power supply exceeding actual capacity or equipment overload. Input the real-time collected supply and demand data into the constructed resource allocation model. These data include the real-time load demand of each load node and the total amount and distribution of the current power supply, so that the model can perform resource allocation calculations based on the actual power supply and demand conditions. The model uses optimization algorithms such as the gradient descent method for iterative calculations. In each iteration, the distribution ratio of power resources between each load node is continuously adjusted according to the current resource allocation situation and the objective function and constraints of the model. At the initial allocation, a preliminary allocation may be made according to certain experience or simple rules, and then iterative optimization is gradually performed to make the allocation result more in line with actual needs and target requirements. After multiple iterative calculations, when the model converges to a solution that meets the requirements, the first resource allocation strategy is obtained. This strategy clarifies how to allocate limited power resources to each load node under the current real-time supply and demand data and network conditions, especially the power supply guarantee measures for key load nodes, such as the specific allocation strategies of power quantity and power supply path allocated to each key load node.

[0034] For example, in a regional distribution network, real-time supply and demand data include the load demand and voltage status of 10 key load nodes with a total load of 100MW. Through monitoring data, it is found that the load of a certain node suddenly increases to 20MW, which exceeds expectations, and the real-time supply and demand data needs to be adjusted.

[0035] In one possible implementation, the system shows that the total load is adjusted to 110MW according to the node priority and load changes, and the newly added load is allocated to the backup path. When extracting the power distribution information from the adjusted real-time supply and demand data, combined with preset boundary conditions such as the line capacity upper limit of 50MW and the voltage stability requirements, the feasible path can be effectively screened. For example, the power distribution information of a path from node G to node H shows that the transmission power is 30MW, which does not exceed the boundary conditions. It can be understood that the method optimizes the power transmission efficiency by gradually adjusting the resource allocation ratio. For example, the initial allocation ratio may be 60% for node G and 40% for node H, which is adjusted to 50% and 50% after iteration, so that the line load is more balanced and the proportional distribution data tends to be stable. Preferably, if the load rate of node G is 80%, the voltage deviation is 5%, and the state value exceeds the preset threshold of 70%, it is marked as an abnormal node. After the abnormal node set is updated, the real-time supply and demand data is adjusted accordingly, and the restriction conditions of the abnormal nodes are added. In one embodiment, the support vector machine algorithm can be used to classify abnormal nodes and normal nodes. After optimizing the generation of real-time supply and demand data, the resource allocation strategy shows that the load of node G needs to be reduced to 15MW to ensure system stability. The flow from node G to node H is adjusted to 25MW, and the backup path shares 10MW.

[0036] It should be noted that the method predicts future trends through historical flow data and generates a first resource allocation strategy. Specifically, if historical data shows that the load of node H increases by 5MW over time, the strategy will reserve additional capacity to ensure long-term reliable operation.

[0037] In one possible implementation, if the capacity of node G decreases due to equipment aging, the system will issue an early warning through abnormal node marking, and the optimized strategy will tilt resources to other paths. This method not only improves the adaptability of power flow distribution, but also reduces the risk of overload of key load nodes. Preferably, the iterative adjustment of the constraint path set can also provide a real-time basis for power dispatching to ensure the efficiency and safety of resource allocation. It can be understood that the combination of the gradient descent method and the support vector machine algorithm makes the resource allocation strategy more accurate. For example, in a certain iterative calculation, the path capacity utilization rate increased from 70% to 85%, and the number of abnormal nodes decreased by 30%. In one embodiment, linear regression analysis can also reveal the seasonal changes in power flow distribution and provide a reference for long-term planning. This multi-level analysis and optimization significantly enhances the operating stability and resource utilization efficiency of the distribution network.

[0038] Specifically, the step S4 includes: S41, constructing a comprehensive objective function of a multi-objective optimization model according to resource allocation efficiency and social benefits, and establishing objective constraint conditions of the multi-objective optimization model according to the flow distribution boundary conditions and actual operation constraints; S42, according to the real-time supply and demand data, the comprehensive objective function and the objective constraint conditions, adopting an optimization algorithm to iteratively adjust the first resource allocation strategy until the multi-objective optimization model converges to obtain a second resource allocation strategy.

[0039] For example, the comprehensive objective function is set by comprehensively considering resource allocation efficiency, social benefits and flow distribution boundary conditions. In terms of resource allocation efficiency, the goal is to maximize the effective use of power resources, reduce energy waste and improve overall power supply efficiency; in terms of social benefits, focus on the security of key loads, ensure the power supply of important users, and incorporate the maximization of key load security rate into the objective function; combined with the flow distribution boundary conditions, maintain the stable operation of the power system, meet the voltage stability constraints and line capacity restrictions, such as minimizing voltage fluctuations and line overload risks as one of the goals. In addition to the constraints contained in the flow distribution boundary conditions (such as line transmission capacity restrictions, node voltage stability range, etc.), other actual operation constraints need to be considered. The total power supply constraint ensures that the total amount of power allocated does not exceed the system's power generation and power supply capacity; the equipment capacity constraint ensures that the load of each device (such as transformers, generators, etc.) does not exceed its rated capacity; and the load demand constraint meets the basic power demand of each load node. The above objective functions and constraints are integrated to construct a multi-objective optimization model. In actual construction, linear weighting method, goal programming method and other methods may be used to comprehensively process multiple objectives, so that the model can seek a balance between multiple objectives and obtain a relatively optimal solution. The first resource allocation strategy is used as the initial solution of the multi-objective optimization model, and real-time resource allocation efficiency data, social benefit evaluation indicators and relevant information of flow distribution boundary conditions are input. These data provide the basis and basis for optimization of the model, so that the model can adjust the initial strategy according to actual conditions. The multi-objective optimization model is solved using appropriate optimization algorithms, such as genetic algorithms and particle swarm optimization algorithms. These algorithms simulate natural evolution or group intelligence behavior, and continuously search and improve the first resource allocation strategy under the premise of meeting constraints, and gradually find a better resource allocation strategy. During the iteration process, the algorithm will continuously adjust the resource allocation parameters, such as the power supply of each load node and the power supply path, according to the objective function of the model, to improve the resource allocation efficiency and social benefits, while ensuring the stable operation of the power system. When the optimization algorithm converges to a solution that meets the preset conditions (such as the objective function value no longer has obvious improvement, the number of iterations reaches the upper limit, etc.), this solution is the optimized second resource allocation strategy. Compared with the first resource allocation strategy, this strategy has better performance in terms of resource allocation efficiency, social benefits and power system stability. It can allocate power resources more reasonably and meet the recovery needs of distribution networks after disasters.

[0040] For example, combining power flow analysis with multi-objective optimization, the voltage stability constraint is first extracted from the boundary conditions of the power flow distribution, and the node voltage fluctuation range is set to ±5% to ensure system stability. By collecting current power grid data, the total load demand of a regional power grid is 150MW, of which the load demands of key nodes A, B, and C are 50MW, 40MW, and 30MW respectively, and the remaining 30MW is borne by other nodes. The available resources on the power generation side include 100MW output of thermal power units, 40MW output of hydropower units, and 10MW output of energy storage systems. A linear programming algorithm is used to minimize carbon emissions and maximize social benefits as the objective function, combined with voltage stability constraints and line transmission capacity restrictions, to perform optimization calculations. The initial optimization results show that the thermal power units are allocated with an output of 80MW, the hydropower units are allocated with an output of 40MW, the energy storage system is allocated with an output of 10MW, and the remaining 20MW load is met by increasing the output of the thermal power units. In the further optimization process, multi-objective weight coefficients were introduced, and the resource allocation ratio was adjusted in combination with the power supply priorities of nodes A, B, and C, so that the load satisfaction rate of node A was increased to 98%, and the load satisfaction rates of nodes B and C reached 95% and 92%, respectively. Finally, through multiple rounds of optimization calculations, an optimized resource allocation plan was generated, in which the output of thermal power units was adjusted to 90MW, the output of hydropower units remained at 40MW, and the output of the energy storage system was adjusted to 10MW, ensuring the power supply reliability of key nodes and the overall stability of the system, while achieving the goals of minimizing carbon emissions and maximizing social benefits.

[0041] Specifically, in step S5, characteristic values ​​are obtained through power flow distribution data, and the priority of key load nodes is processed by support vector machine algorithm to obtain load classification data. Stability-related indicators are extracted from the load classification data, and the temporary equipment adjustment requirements are determined in combination with the equipment location information to obtain equipment adjustment data. The network topology is updated according to the equipment adjustment data, and the joint adjustment parameters are processed by linear programming algorithm to obtain topology adjustment data. The trend of power flow distribution change is obtained for the topology adjustment data. If the change trend exceeds the preset threshold, it is marked as an abnormal state to obtain abnormal marking data. The stability correction requirements are extracted from the abnormal marking data, and the index extraction results are adjusted by time series analysis to obtain correction index data. The allocation scheme is updated by correcting the index data to determine whether the configuration determination conditions are met, and the network configuration after joint adjustment is obtained. By obtaining temporary equipment parameter data, the equipment type, power capacity and voltage level information are extracted from it to generate an equipment feature matrix. According to the geographical coordinates of the mobile power supply and the emergency equipment, the distance to the existing power grid node is calculated, and the nearest node is selected as a potential access point. The network topology is analyzed by graph theory algorithm, key nodes and weak links are identified, and a network vulnerability assessment model is established. If the access point load rate exceeds the preset threshold, the device allocation plan is adjusted to select the suboptimal access point until the load rate requirement is met. The power flow calculation method is used to simulate the network operation status after the temporary equipment is connected, and the node voltage and line power are calculated. Based on the power flow calculation results, it is determined whether there is an overload or undervoltage situation. If so, it is marked as an unstable state. According to the stability judgment results and equipment characteristics, the temporary power supply network configuration is optimized.

[0042] For example, based on the optimized power flow distribution eigenvalues, the eigenvalue decomposition algorithm is used to extract the power flow stability index of key nodes, where the voltage eigenvalues ​​of nodes D, E, and F are 92, 88, and 85, respectively, indicating that node D has the best stability. Combined with the priority weights of key load nodes, the priorities of nodes D, E, and F are 6, 3, and 1, respectively. The overall power flow stability of the system is calculated to be 78 by the weighted average method, which is lower than the set threshold of 85. In order to achieve network topology optimization, temporary equipment location information is introduced, and two temporary transformers are deployed near nodes E and F, respectively, and their outputs are adjusted to 20MW and 15MW. The line connection is replanned through the graph theory algorithm, and the line capacity between nodes E and D is increased to 50MW, and the line capacity between nodes F and D is increased to 40MW. Using the improved particle swarm optimization algorithm, the network configuration is optimized with the goal of minimizing line loss and maximizing power flow stability, and finally the adjusted network topology is obtained, where the voltage eigenvalue of node E is increased to 91, the voltage eigenvalue of node F is increased to 89, and the overall power flow stability of the system reaches 86, which meets the design requirements.

[0043] For example, the capacity and location information is extracted from the mobile power and emergency equipment database, and the data mining technology is used to screen out equipment with a capacity greater than 30MW and no more than 5 kilometers away from the target area. The capacity of mobile power supply A is 35MW and is located 3 kilometers northeast of node G, and the capacity of mobile power supply B is 40MW and is located 4 kilometers southeast of node H. The device access nodes are matched by network topology constraints, and the Dijkstra algorithm is used to calculate the shortest path from the device to the target node, and it is determined that mobile power supply A is connected to node G and mobile power supply B is connected to node H. Combined with the power flow stability analysis, the power flow calculation is performed using the Newton-Raphson method, and the voltage characteristic value of node G is 89 and the voltage characteristic value of node H is 90, both of which meet the requirement of the stability threshold of 85. According to the equipment output parameters, the output of mobile power supply A is set to 30MW and the output of mobile power supply B is set to 35MW. The line load distribution is optimized by the linear programming algorithm, and the line capacity between node G and node H is increased to 45MW, and the line capacity between node G and node D is increased to 50MW. Based on the power flow stability index, the Monte Carlo simulation method was used to evaluate the reliability of the temporary power supply network, and the overall stability of the system was obtained to be 87, which met the design requirements.

[0044] Specifically, in step S6, the mapping relationship between the resource allocation ratio and the output power is obtained through the optimization strategy, and the linear regression algorithm is used to determine the correlation strength to obtain the power allocation benchmark of the temporary equipment. The priority weight data is extracted from the key load nodes, and the data is matched with the emergency equipment capacity information after weight sorting to determine the capacity demand distribution of the load nodes. According to the capacity demand distribution and the power allocation benchmark of the temporary equipment, the power supply redundancy of each load node is calculated to obtain the redundancy evaluation result. If the redundancy evaluation result is lower than the preset threshold, the resource allocation ratio is adjusted, and the emergency equipment capacity is re-matched to obtain an updated allocation plan. The power coverage of the key load nodes is calculated through the updated allocation plan, and the Monte Carlo algorithm is used to simulate various allocation scenarios to determine the optimal power allocation combination. According to the optimal power allocation combination and the priority weight of the load node, the operating parameters of the temporary equipment are adjusted to obtain the final capacity allocation strategy.

[0045] It is worth noting that in the second resource allocation strategy, by associating the resource allocation ratio with the output power of temporary equipment, key parameters are first extracted from the priority weights of key load nodes and matched with the emergency equipment capacity data.

[0046] For example, the priority weight of the critical load node X is 8, the capacity of the emergency equipment C is 25MW, and the capacity of the emergency equipment D is 30MW. The power supply redundancy of node X is calculated by the weighted average algorithm to be 2, which meets the requirement of minimum redundancy 0. The multi-objective optimization algorithm is adopted, with the minimization of power supply cost as the objective function, combined with the equipment capacity and load demand constraints, to determine the output power of equipment C to be 20MW, and the output power of equipment D to be 25MW, to ensure the power supply reliability of the critical load nodes. In order to improve the accuracy of the allocation scheme, the hierarchical analysis method is introduced to classify the critical load nodes, and the nodes X, Y, and Z are divided into high, medium, and low priorities, with weights of 8, 6, and 4, respectively. Combined with the emergency equipment capacity data, the power supply demand of each node is predicted by the linear regression model. The predicted demand of node Y is 15MW, and the predicted demand of node Z is 10MW. According to the prediction results, the output power of equipment C and D is dynamically adjusted to ensure that the power supply redundancy of the high-priority node X is always higher than 2, and the redundancy of the medium and low priority nodes is maintained above 0 and 8, respectively. Finally, the output distribution of equipment was optimized through genetic algorithms, and the output of equipment C was adjusted to 18MW and the output of equipment D was adjusted to 22MW. Combined with the line capacity limitations, the shortest path algorithm was used to optimize the power transmission path to ensure the overall stability and economy of the power supply network.

[0047] Specifically, in step S7, the topological structure information of the temporary power supply network is obtained, including the number of nodes, line connection relationship and parameters, and a network model is constructed. The voltage distribution and line flow of each node are calculated according to the network model, and the power flow equation is solved by the Newton-Raphson method to obtain the initial power flow distribution result. The sensitivity analysis method is used to calculate the degree of influence of each node on the system stability, and identify key nodes and weak lines. For key nodes and weak lines, voltage stability constraints and power balance constraints are set to construct an optimization model. The optimization model is solved by a genetic algorithm to obtain the optimal access location and capacity configuration scheme of temporary equipment that meets the constraints. According to the optimal configuration scheme, the static voltage stability margin index of the network is calculated to determine whether the system stability meets the requirements. If not, the constraints are adjusted and the optimization solution is re-performed. Based on the final temporary equipment configuration scheme, the deployment path of the temporary equipment is planned, and the installation sequence and connection method are determined.

[0048] By analyzing the power flow distribution characteristic values ​​of the temporary power supply network and combining the voltage stability constraint parameters, the stability margin of the network operation is determined. For example, the power flow characteristic value of node A is 2, node B is 8, and node C is 5. Based on these data, the Newton-Raphson algorithm is used to calculate the voltage deviation of each node, and the voltage deviation of node A is 03, node B is 05, and node C is 02, which meets the voltage stability constraint conditions.

[0049] Combined with the power flow stability index, the sensitivity analysis method is used to evaluate the impact of temporary equipment access nodes on network stability. For example, the sensitivity index of node D is 12, node E is 08, and node F is 15. Through comparative analysis, node F is determined to be the optimal access point. On this basis, the Dijkstra algorithm in graph theory is used to optimize the deployment path of temporary equipment with the goal of minimizing line loss. For example, the shortest path from node F to node A is path 1, with a loss of 5MW, the shortest path to node B is path 2, with a loss of 3MW, and the shortest path to node C is path 3, with a loss of 4MW. Finally, path 2 is selected as the main deployment path to ensure the stability and economy of network operation.

[0050] Specifically, the step S8 includes: S81, calculating each critical load guarantee rate according to the load demand data, through the allocated capacity of the capacity allocation strategy and the power supply capacity of the temporary power supply configuration; S82, evaluating a network stability index according to the load demand data, through the network topology and power transmission conditions of the temporary power supply configuration; S83, analyzing the key load guarantee rate and the network stability index through a simulation algorithm to obtain a recovery efficiency evaluation result.

[0051] Exemplarily, based on the load demand data, a data acquisition tool is used to extract the load change characteristics of each node to obtain the load distribution state. According to the load distribution state, each key load guarantee rate is calculated through the allocation capacity of the capacity allocation strategy and the power supply capacity of the temporary power supply configuration; the key load guarantee rate is compared with a preset threshold value, and if it is lower than the threshold value, the capacity allocation strategy is adjusted to obtain an updated allocation parameter. Through the updated allocation parameters, the topology adjustment information of the temporary power supply network configuration is obtained to determine the change trend of network stability. According to the change trend, a simulation algorithm is used to calculate the dynamic distribution characteristics of the network stability index to determine whether the system meets the power balance requirements. If the power balance requirements are not met, the network configuration is optimized through a linear programming algorithm to obtain an adjusted recovery efficiency parameter. According to the adjusted recovery efficiency parameter, the stable state of the system operation is determined to obtain a recovery efficiency evaluation result.

[0052] For example, firstly, the updated load demand data is extracted from the real-time supply and demand situation, for example, the load demand of node X is 120MW, node Y is 80MW, and node Z is 150MW. Through the Monte Carlo simulation method, combined with the load fluctuation characteristics, 1000 random scenarios are generated to analyze the critical load guarantee rate. For example, at a confidence level of 95%, the critical load guarantee rate reaches 95%. The particle swarm optimization algorithm is used to optimize the calculation with the network stability index as the objective function. If the optimized network stability index is 92, it meets the preset threshold of 90. Based on the simulation results, the support vector machine model is used to evaluate the system recovery efficiency. The system recovery efficiency predicted by the model is 93%, which is only 2% compared with the actual recovery efficiency of 91%. Finally, through Bayesian network analysis, the key factors affecting the system recovery efficiency, such as load fluctuation, network topology and equipment failure rate, are determined to provide a basis for subsequent optimization. The whole process is automated through information technology to ensure the accuracy and reliability of the analysis results.

[0053] Specifically, the step S9 includes: S91, acquiring deviation feedback data according to the recovery efficiency evaluation result, analyzing the deviation feedback data, and obtaining distribution characteristics of the deviation feedback data; S92, updating the load rating parameter according to the distribution characteristics, optimizing the resource allocation ratio according to the updated load rating parameter, and obtaining an optimized allocation parameter; S93, iteratively adjusting the resource allocation model according to the optimized allocation parameters to generate a third resource allocation strategy.

[0054] Exemplarily, deviation feedback data is obtained from the system recovery efficiency evaluation result, and the deviation feedback data is processed by statistical tools to obtain the distribution characteristics of the feedback data. The change of load importance is judged by the distribution characteristics, and the load rating parameters are updated by an adaptive algorithm to determine the adjustment value of the load rating parameters. The update demand of resource allocation is obtained according to the adjustment value, and the resource allocation ratio is optimized by a linear programming algorithm to obtain the optimized allocation parameters. The resource allocation model is iteratively adjusted according to the optimized allocation parameters, the reallocation information of power resources is obtained, and the allocation scheme of power resources is determined. The characteristic data of the allocation result is extracted from the allocation scheme, and whether the allocation result meets the balance requirement is judged by comparative analysis to obtain the judgment value of the balance state. If the judgment value is lower than the preset threshold, the resource allocation model is updated by adjusting the allocation parameters to obtain the adjusted allocation result. The configuration information of the power resources is updated according to the adjusted allocation result, and the consistency of the configuration information is judged by data verification to obtain the generation of the third resource allocation strategy.

[0055] For example, based on the results of the system recovery efficiency evaluation, the difference between load distribution and actual demand is identified through deviation feedback data, such as the load deviation of node A is 5MW, node B is 3MW, and node C is 7MW. The adaptive filtering algorithm is used to process the deviation data, and the load importance rating is updated in combination with the historical load characteristics. If the importance rating of node A is adjusted from 8 to 85, node B is adjusted from 7 to 75, and node C is adjusted from 9 to 92. Based on the updated load rating parameters, the genetic algorithm is used to optimize the power resource allocation ratio, and the objective function is to minimize the deviation and maximize the resource utilization. If the resource allocation ratio of node A is adjusted from 30% to 32% after optimization, node B is adjusted from 25% to 27%, and node C is adjusted from 45% to 47%. The optimization results are input into the dynamic power resource allocation model, and the third resource allocation strategy is generated by combining real-time supply and demand data with the network topology. For example, the allocated power of node A is 100MW, node B is 80MW, and node C is 120MW. Through automated processing technology, the real-time and accuracy of the allocation plan are ensured to provide support for the stable operation of the power system.

[0056] Specifically, in step S10, a temporary power supply network is deployed through a third resource allocation strategy, an operation instruction is extracted from the temporary equipment access parameters, and the equipment output power and access status are adjusted by a remote control system to obtain the adjusted equipment operation data. The real-time status of the critical load is obtained from the adjusted equipment operation data, and the change trend of the real-time status is analyzed by a data processing tool to determine the guarantee rate of the critical load. For the guarantee rate of the critical load, if the guarantee rate reaches a preset threshold, the stable operation state of the distribution network is judged by a logical check to obtain the stability data of the distribution network. According to the stability data of the distribution network, the allocation efficiency of the power resources is analyzed by a support vector machine algorithm to determine the optimization requirements of the temporary power supply network. The instruction parameters of the remote control system are adjusted by optimizing the requirements, the updated output power and access status are obtained, and the dynamic adjustment information of the distribution network is obtained. The characteristic data of the distribution network state is extracted from the dynamic adjustment information, and the distribution law of the characteristic data is analyzed by a clustering algorithm to determine the collaborative recovery capability of the distribution network. For the collaborative recovery capability, the characteristic data is processed by a statistical tool to determine the final configuration state of the power resources and obtain a stable operation plan for the temporary power supply network.

[0057] Exemplarily, according to the third resource allocation strategy, when deploying a temporary power supply network, operating instructions such as voltage, frequency, and power factor are extracted from the access parameters of the temporary equipment. If the voltage instruction of node D is 15kV, the frequency is 50Hz, and the power factor is 95. Through the remote control system, combined with real-time monitoring data, the fuzzy control algorithm is used to adjust the output power and access status of the equipment. If the output power of node D is adjusted from 50MW to 55MW, node E is adjusted from 40MW to 45MW, and node F is adjusted from 60MW to 65MW. During the adjustment process, based on the real-time collected load data, the Kalman filter algorithm is used to analyze the guarantee rate of key loads. The guarantee rate of node D is increased from 85% to 92%, node E is increased from 80% to 88%, and node F is increased from 90% to 95%. When the critical load protection rate reaches the expected threshold (such as 90%), the system automatically determines that it has entered a stable operating state and generates stable operating state parameters for the distribution network. When the voltage of node D is stable at 15kV±2kV, the frequency is stable at 50Hz±1Hz, and the power factor is stable at 95±02, the stable operating state parameters are compared and analyzed with historical data through automated processing technology to ensure that the recovery process of the distribution network is consistent with the expected goals.

[0058] The embodiment of the present invention discloses a method for coordinated power restoration of a distribution network after a disaster. By acquiring real-time supply and demand status data and network topology constraint information, analyzing and extracting a set of key load nodes and boundary conditions of power flow distribution, a resource allocation model is constructed, and an optimized allocation strategy is obtained by integrating power flow analysis and multi-objective optimization methods. Combined with temporary equipment access parameters, the temporary power supply network configuration and capacity allocation plan are determined, and then the deployment path of the temporary power supply configuration is planned to deploy a temporary power supply network to coordinate power restoration of the distribution network. It is able to quickly evaluate the supply and demand status in a post-disaster environment, optimize resource allocation, ensure power supply to key loads, and improve the efficiency and stability of distribution network restoration, providing an effective solution for the rapid recovery of the power system after a disaster.

[0059] See also Figure 2 , Figure 2 1 is a schematic diagram of a coordinated power restoration system 10 for a distribution network after a disaster provided by an embodiment of the present invention. The coordinated power restoration system 10 for a distribution network after a disaster includes: The load rating parameter determination module 11 is used to obtain the real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, extract and analyze the real-time supply and demand data, and determine the current load rating parameters of the distribution network; wherein the real-time supply and demand data includes load demand data and power supply data; The distribution boundary condition determination module 12 is used to calculate the priority weight of each load node according to the load rating parameters, load historical data and user type, so as to determine the key load node set of the distribution network; calculate the feasible path of the power flow of the distribution network according to the network topology information, and determine the power flow distribution boundary conditions of the distribution network; A first allocation strategy determination module 13 is used to construct a resource allocation model according to the power flow distribution boundary conditions and the key load node set, and adopt the resource allocation model to determine a first resource allocation strategy through the real-time supply and demand data; The second allocation strategy determination module 14 is used to construct a multi-objective optimization model according to the resource allocation efficiency and social benefits, as well as the flow distribution boundary conditions; the multi-objective optimization model is used to optimize the first resource allocation strategy to obtain a second resource allocation strategy; wherein the multi-objective optimization model includes a comprehensive objective function and objective constraints; A temporary power supply configuration determination module 15, configured to determine a temporary power supply configuration of the distribution network according to the power flow distribution data in the second resource allocation strategy, the priority weight and the access parameters of the temporary equipment; A capacity allocation strategy determination module 16 is used to determine the power supply redundancy of each key load node according to the resource allocation ratio in the second resource allocation strategy and the output power of the temporary power supply configuration, so as to obtain the capacity allocation strategy of the distribution network; The first deployment path planning module 17 is used to plan the deployment path of the temporary power supply configuration according to the power flow distribution data and voltage stability constraint parameters in the second resource allocation strategy and the information of the temporary power supply configuration, so as to coordinate the restoration of power to the distribution network.

[0060] Furthermore, the distribution network coordinated power restoration system 10 after a disaster further includes: An efficiency evaluation result acquisition module is used to analyze the current load demand data according to the capacity allocation plan and the temporary power supply configuration to obtain key load guarantee rate and network stability index, and analyze the key load guarantee rate and network stability index through a simulation algorithm to obtain a recovery efficiency evaluation result; a third allocation strategy generating module, configured to update the load rating parameter and the resource allocation ratio according to the recovery efficiency evaluation result, so as to adjust the resource allocation model and generate a third resource allocation strategy; The second deployment path planning module is used to plan the deployment path of the adjusted temporary power supply configuration according to the third resource allocation strategy to coordinate power restoration of the distribution network.

[0061] A distribution network collaborative power restoration system 10 based on post-disaster power supply provided by an embodiment of the present invention can implement all processes of the distribution network collaborative power restoration method based on post-disaster power supply of the above-mentioned embodiment. The functions of each module in the system and the technical effects achieved are respectively the same as the functions of the distribution network collaborative power restoration method based on post-disaster power supply of the above-mentioned embodiment, and will not be repeated here.

[0062] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for coordinated power restoration of a distribution network after a disaster, characterized in that: include: Acquire real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, extract and analyze the real-time supply and demand data, and determine the current load rating parameters of the distribution network; wherein the real-time supply and demand data includes load demand data and power supply data; According to the load rating parameters, load historical data and user types, the priority weight of each load node is calculated to determine the key load node set of the distribution network; according to the network topology information, the feasible path of the power flow of the distribution network is calculated, and the power flow distribution boundary conditions of the distribution network are determined; Constructing a resource allocation model according to the flow distribution boundary conditions and the key load node set, and using the resource allocation model to determine a first resource allocation strategy through the real-time supply and demand data; A multi-objective optimization model is constructed according to resource allocation efficiency and social benefits, as well as the boundary conditions of the power flow distribution; the first resource allocation strategy is optimized using the multi-objective optimization model to obtain a second resource allocation strategy; wherein the multi-objective optimization model includes a comprehensive objective function and objective constraints; Determining a temporary power supply configuration of the distribution network according to the power flow distribution data in the second resource allocation strategy, the priority weight and the access parameters of the temporary equipment; Determine the power supply redundancy of each critical load node according to the resource allocation ratio in the second resource allocation strategy and the output power of the temporary power supply configuration, and obtain the capacity allocation strategy of the distribution network; According to the power flow distribution data and voltage stability constraint parameters in the second resource allocation strategy, as well as the information of the temporary power supply configuration, a deployment path of the temporary power supply configuration is planned to coordinate power restoration of the distribution network.

2. The method for coordinated power restoration of a distribution network after a disaster according to claim 1, characterized in that: After planning the deployment path of the temporary power supply configuration, the method further includes: According to the capacity allocation plan and the temporary power supply configuration, the current load demand data is analyzed to obtain key load guarantee rate and network stability index, and the key load guarantee rate and network stability index are analyzed by simulation algorithm to obtain restoration efficiency evaluation result; According to the recovery efficiency evaluation result, updating the load rating parameter and the resource allocation ratio to adjust the resource allocation model and generate a third resource allocation strategy; According to the third resource allocation strategy, a deployment path of the adjusted temporary power supply configuration is planned to coordinate power restoration of the distribution network.

3. The method for coordinated power restoration of distribution networks after disasters according to claim 1, characterized in that: The acquiring of real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, extracting and analyzing the real-time supply and demand data, and determining the current load rating parameters of the distribution network include: Obtain the real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, use the time series analysis method to analyze the change trend of the real-time supply and demand data, and obtain the load fluctuation characteristics and the corresponding power supply status; Calculating the matching deviation between the load fluctuation characteristics and the corresponding power supply state, determining the current supply and demand imbalance area and degree of the distribution network, so as to obtain the current supply and demand matching deviation distribution of the distribution network; According to the supply-demand matching deviation distribution, a current load rating parameter of the distribution network is determined.

4. The method for coordinated power restoration of a distribution network after a disaster according to claim 1, characterized in that: The method of calculating the priority weight of each load node according to the load rating parameters, load historical data and user type to determine the key load node set of the distribution network; calculating the feasible path of the power flow of the distribution network according to the network topology information, and determining the power flow distribution boundary conditions of the distribution network, includes: Calculate the priority weight of each load node according to the load rating parameters, load historical data and user type; All load nodes corresponding to the priority weight being greater than a preset load weight threshold are taken as a set of key load nodes of the distribution network; Calculating a feasible path of the power flow of the distribution network through a topology analysis algorithm according to the line connection status and damaged node location information in the network topology information; According to the feasible path, the boundary constraint conditions of the power flow are determined, and according to the boundary constraint conditions and the line connection status, the boundary conditions of the power flow distribution of the distribution network are obtained.

5. The method for coordinated power restoration of distribution networks after disasters according to claim 1, characterized in that: The step of constructing a resource allocation model according to the flow distribution boundary conditions and the key load node set, and using the resource allocation model to determine a first resource allocation strategy through the real-time supply and demand data includes: Constructing a resource allocation model according to the flow distribution boundary conditions and the key load node set; wherein the resource allocation model includes an allocation objective function and allocation constraints; The real-time supply and demand data is input into the resource allocation model, and according to the real-time supply and demand data, the allocation objective function and the allocation constraints, an optimization algorithm is used to iteratively adjust the resource allocation ratio between each load node until the resource allocation model converges, thereby obtaining a first resource allocation strategy.

6. The method for coordinated power restoration of a distribution network after a disaster according to claim 1, characterized in that: The multi-objective optimization model is constructed according to the resource allocation efficiency and social benefits, as well as the flow distribution boundary conditions; The multi-objective optimization model is used to optimize the first resource allocation strategy to obtain a second resource allocation strategy, including: Constructing a comprehensive objective function of a multi-objective optimization model according to resource allocation efficiency and social benefits, and establishing objective constraints of the multi-objective optimization model according to the flow distribution boundary conditions and actual operation constraints; According to the real-time supply and demand data, the comprehensive objective function and the objective constraints, an optimization algorithm is used to iteratively adjust the first resource allocation strategy until the multi-objective optimization model converges to obtain a second resource allocation strategy.

7. The method for coordinated power restoration of a distribution network after a disaster according to claim 2, characterized in that: According to the capacity allocation scheme and the temporary power supply configuration, the current load demand data is analyzed to obtain key load guarantee rate and network stability index, and the key load guarantee rate and network stability index are analyzed by simulation algorithm to obtain the recovery efficiency evaluation result, including: Calculate each critical load guarantee rate according to the load demand data by using the allocated capacity of the capacity allocation strategy and the power supply capacity of the temporary power supply configuration; According to the load demand data, the network stability index is evaluated through the network topology and power transmission conditions of the temporary power supply configuration; The critical load guarantee rate and the network stability index are analyzed by a simulation algorithm to obtain a restoration efficiency evaluation result.

8. The method for coordinated power restoration of a distribution network after a disaster according to claim 2, characterized in that: The updating of the load rating parameter and the resource allocation ratio according to the recovery efficiency evaluation result to adjust the resource allocation model and generate a third resource allocation strategy includes: Obtaining deviation feedback data according to the recovery efficiency evaluation result, analyzing the deviation feedback data, and obtaining distribution characteristics of the deviation feedback data; Update the load rating parameter according to the distribution characteristics, optimize the resource allocation ratio according to the updated load rating parameter, and obtain an optimized allocation parameter; The resource allocation model is iteratively adjusted according to the optimized allocation parameters to generate a third resource allocation strategy.

9. A coordinated power restoration system for distribution networks after disasters, characterized in that: include: A load rating parameter determination module is used to obtain real-time supply and demand data and network topology information of the post-disaster distribution network in the target area, extract and analyze the real-time supply and demand data, and determine the current load rating parameters of the distribution network; wherein the real-time supply and demand data includes load demand data and power supply data; A distribution boundary condition determination module is used to calculate the priority weight of each load node according to the load rating parameters, load historical data and user type, so as to determine the key load node set of the distribution network; calculate the feasible path of the power flow of the distribution network according to the network topology information, and determine the power flow distribution boundary conditions of the distribution network; A first allocation strategy determination module is used to construct a resource allocation model according to the power flow distribution boundary conditions and the key load node set, and adopt the resource allocation model to determine a first resource allocation strategy through the real-time supply and demand data; A second allocation strategy determination module is used to construct a multi-objective optimization model according to resource allocation efficiency and social benefits, as well as the boundary conditions of the power flow distribution; the multi-objective optimization model is used to optimize the first resource allocation strategy to obtain a second resource allocation strategy; wherein the multi-objective optimization model includes a comprehensive objective function and objective constraints; A temporary power supply configuration determination module, configured to determine the temporary power supply configuration of the distribution network according to the power flow distribution data in the second resource allocation strategy, the priority weight and the access parameters of the temporary equipment; A capacity allocation strategy determination module, configured to determine the power supply redundancy of each key load node according to the resource allocation ratio in the second resource allocation strategy and the output power of the temporary power supply configuration, and obtain the capacity allocation strategy of the distribution network; The first deployment path planning module is used to plan the deployment path of the temporary power supply configuration according to the power flow distribution data and voltage stability constraint parameters in the second resource allocation strategy and the information of the temporary power supply configuration, so as to coordinate the restoration of power to the distribution network.

10. The coordinated power restoration system for distribution networks after disasters according to claim 9, characterized in that: Also includes: An efficiency evaluation result acquisition module is used to analyze the current load demand data according to the capacity allocation plan and the temporary power supply configuration to obtain key load guarantee rate and network stability index, and analyze the key load guarantee rate and network stability index through a simulation algorithm to obtain a recovery efficiency evaluation result; A third allocation strategy generating module, configured to update the load rating parameter and the resource allocation ratio according to the recovery efficiency evaluation result, so as to adjust the resource allocation model and generate a third resource allocation strategy; The second deployment path planning module is used to plan the deployment path of the adjusted temporary power supply configuration according to the third resource allocation strategy to coordinate power restoration of the distribution network.

Images