Distributed power supply optimal configuration method and system for self-healing of power distribution network

Abstract

The invention discloses a power distribution network self-healing-oriented distributed power supply optimal configuration method and system, and the method comprises the steps: constructing an operation layer model with the lowest total operation cost of a power distribution network as a target through optimizing the power flow of the power distribution network and dividing distributed power supplies according to a topological structure of the power distribution network; based on the operation layer, constructing a planning layer model with the lowest annual total cost of the power distribution network as a target; combining the operation layer model, the planning layer model and the operation-planning double-layer model, and solving the operation-planning double-layer model through an improved particle swarm algorithm to obtain an optimal configuration scheme of the distributed power supply; and planning the installation position and the installation capacity of the distributed power supply according to the optimal configuration scheme of the distributed power supply. The operation stability of the power distribution network is improved by reconstructing the power distribution network structure and additionally arranging the distributed power supply configuration, and the power utilization quality of the user is further improved.

Description

Technical Field

[0001] The present application belongs to the technical field of distribution network planning, and specifically relates to a distributed power supply optimization configuration method and system for distribution network self-healing. Background Art

[0002] The network structure of the distribution network is flexible and variable. After a fault or a sudden load change occurs, the self-healing process of the distribution network can change the switch state on the line to adjust the network structure and then change the power flow distribution, so as to achieve rapid recovery of power supply to some areas and ensure the stability of the power grid operation. With the development of smart grids, more and more distributed power sources are connected to the distribution network, which can independently supply power to the surrounding areas and form an island operation state. Network reconstruction and island operation based on distributed resources are both effective means of self-healing of distribution networks. In addition, distribution networks are prone to faults and sudden load changes, which directly affect social production and residents' daily power supply. Therefore, it is necessary to focus on the self-healing ability of the distribution network to improve the stability of the power grid operation and study the capacity and site selection planning of distributed power sources.

[0003] At present, the existing method to improve the self-healing ability of the distribution network is to construct a two-layer model based on network reconstruction and island division for optimization. However, this method generally considers network reconstruction first and then island division. The obtained distributed power configuration scheme is difficult to cover the variable fault conditions during the actual operation of the distribution network, and cannot significantly improve the self-healing ability of the distribution network and quickly resume operation after a fault. The improvement of the operating stability of the distribution network is not significant. Summary of the invention

[0004] This application proposes a distributed power supply optimization configuration method and system for distribution network self-healing, which improves the operating stability of the distribution network and further improves the power quality of users by reconstructing the distribution network structure and adding distributed power supply configurations.

[0005] Configuration optimization of distributed power sources improves the operational stability of the distribution network and reduces the cost required for self-healing.

[0006] A first aspect of the present application provides a distributed power supply optimization configuration method for distribution network self-healing, the method comprising:

[0007] According to the topology of the distribution network, by optimizing the power flow of the distribution network and dividing the distributed power sources, an operation layer model is constructed with the goal of minimizing the total operation cost of the distribution network;

[0008] Based on the operation layer model, a planning layer model is constructed with the goal of minimizing the total annual cost of the distribution network;

[0009] Combining the operation layer model and the planning layer model, an operation-planning double-layer model is formed, and the operation-planning double-layer model is solved by an improved particle swarm algorithm to obtain an optimal configuration scheme for distributed power sources;

[0010] According to the optimal configuration plan of distributed power sources, the installation location and installation capacity of distributed power sources are planned.

[0011] The above scheme is based on the topological structure of the power grid. By considering the reconstruction of the distribution network flow and the division of distributed power sources at the same time, the operation layer model is constructed to reduce the total operation cost of the distribution network as much as possible while improving the self-healing ability of the distribution network; then, based on the cost target of the operation layer model, a planning layer model with the lowest annual total cost of the distribution network is constructed. The two models are then combined to form a two-layer model with inter-layer influence. By improving the particle swarm algorithm, the optimal configuration scheme of distributed power sources that can minimize the total cost of the distribution network is found on the basis of improving the self-healing ability of the distribution network, so as to achieve the optimization of the distributed power configuration while improving the economy of the distribution network, greatly improve the self-healing ability of the distribution network after faults and reduce the cost required to optimize the self-healing ability, and improve the operation stability of the distribution network and the quality of electricity consumption of users.

[0012] In a possible implementation method of the first aspect, according to the topology of the distribution network, by optimizing the distribution network flow and dividing the distributed power sources, an operation layer model with the goal of minimizing the total operation cost of the distribution network is constructed, specifically:

[0013] According to the preset power flow constraints and the preset distributed power output constraints, the power flow distribution and the division of distributed power sources in the distribution network topology are optimized to obtain a unified self-healing strategy;

[0014] According to the historical operation data of the distribution network, the operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network;

[0015] The operation layer model is constructed by using the unified self-healing strategy and the operation layer objective function.

[0016] The above scheme obtains a unified self-healing strategy by simultaneously considering the improvement of the self-healing ability of the distribution network by distribution network reconstruction and distributed power generation division; then, based on the unified self-healing strategy, the total operating cost of the distribution network and the operating layer model are constructed to link the flexible transformation of the distribution network topology structure with the economic cost generated, thereby significantly improving the economic benefits of the distribution network while significantly improving the self-healing ability of the distribution network.

[0017] In a possible implementation method of the first aspect, based on the historical operation data of the distribution network, an operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network, specifically:

[0018] According to the historical operation data of the distribution network and the preset distribution network self-healing process set, the load reduction cost, self-healing cost and network loss cost generated by the distribution network in the self-healing process are constructed respectively;

[0019] Combine the load reduction cost, self-healing cost and network loss cost to construct the total operating cost of the distribution network;

[0020] The operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network.

[0021] In a possible implementation method of the first aspect, the load reduction cost, the self-healing cost, and the network loss cost are specifically:

[0022] The load reduction cost is specifically expressed as follows:

[0023]

[0024] In the formula, C load For load reduction costs, N T is the total number of time intervals within the fault period, N load is the total number of load nodes in the distribution network, β1, β2, and β3 are all load weight coefficients, and S load1 , S load2 , S load3 are the load reduction amounts corresponding to each level, and Δt is the length of each time interval during the fault period;

[0025] The self-healing cost is specifically expressed as:

[0026]

[0027] In the formula, C re is the self-healing cost, c swtich is the switch operation cost coefficient, N line is the total number of branches in the distribution network, is the switch state of the branch with node i as the starting point and node j as the end point during the fault recovery period t, c DG is the output cost of distributed generation, N DG is the total number of nodes of distributed generation, is the output of distributed generation in period t;

[0028] The network loss cost is specifically expressed as:

[0029]

[0030] In the formula, C net is the network loss cost, c line is the network loss cost coefficient, is the active power loss of line ij during period t.

[0031] In a possible implementation method of the first aspect, based on the operation layer, a planning layer model with the lowest annual total cost of the distribution network as the goal is constructed, specifically:

[0032] According to the preset distributed power capacity constraints, considering the installed capacity and number of distributed power, the distributed power investment cost is constructed;

[0033] Combine the distributed generation investment cost and the total operating cost of the distribution network at the operating layer to construct the total annual cost of the distribution network;

[0034] With the goal of minimizing the annual total cost of the distribution network, a target planning layer model is constructed.

[0035] The above scheme reflects the reconstruction of the distribution network and the division of distributed power sources through the installed capacity and number of distributed power sources, and constructs the distributed power source investment cost brought about by these topological changes, and then combines the total operating cost of the distribution network at the operating layer to achieve comprehensive planning of the location and capacity of distributed power sources.

[0036] In a possible implementation method of the first aspect, the distributed power source investment cost is specifically:

[0037] The specific expression of the distributed power source investment cost is:

[0038]

[0039] In the formula, C inv is the investment cost of distributed power generation, r is the discount rate, n is the useful life of distributed power generation, N DG is the number of distributed generation to be installed, c inv,DG is the unit capacity investment cost of distributed power generation, P DGi max Install the capacity of the distributed generation at node i.

[0040] In a possible implementation method of the first aspect, the operation-planning two-layer model is solved by an improved particle swarm algorithm to obtain an optimal configuration scheme for distributed power sources, specifically:

[0041] Based on a preset typical scenario data set, the installation location and installation capacity of the distributed generation in the operation-planning two-layer model are iterated by using an improved particle swarm algorithm to optimize the distribution network self-healing process under the typical scenario data set;

[0042] When the particle swarm algorithm is terminated, the optimal installation location and optimal installation capacity of the distributed power source are obtained;

[0043] According to the distribution network topology and the optimal installation location and optimal installation capacity, an optimal configuration plan for distributed power sources is constructed.

[0044] The above scheme first uses a typical scenario data set and an improved particle swarm algorithm to iterate the installation location and installation capacity of distributed power sources, ensuring that the self-healing capability of the distribution network under various fault scenarios is improved as much as possible, and improving the comprehensiveness of the optimal configuration scheme of distributed power sources; in the iterative process, the installation location and installation capacity of distributed power sources are used as variables to obtain the optimal installation location and optimal installation capacity that can both improve the self-healing capability of the distribution network and reduce the self-healing cost, thereby optimizing the power quality of users.

[0045] In a possible implementation method of the first aspect, the particle swarm algorithm is improved, specifically:

[0046] A preset power exponential operator is introduced into the inertia weight of the particle swarm optimization algorithm to obtain an improved particle swarm algorithm; wherein the inertia weight acts on the iterative convergence speed of the installation position and the installation capacity.

[0047] The above solution optimizes the iteration speed of the algorithm by introducing a power exponential operator, reduces the complexity of solving the model, and improves the efficiency of solving the model.

[0048] The second aspect of the present application provides a distributed power optimization configuration system for distribution network self-healing, the system comprising: an operation layer model construction module, a planning layer model construction module, an optimal configuration solution generation module and a distributed power planning module;

[0049] The operation layer model building module is used to build an operation layer model with the goal of minimizing the total operation cost of the distribution network by optimizing the distribution network flow and dividing the distributed power sources according to the distribution network topology;

[0050] The planning layer model building module is used to build a planning layer model with the lowest annual total cost of the distribution network as the goal based on the operation layer model;

[0051] The optimal configuration scheme generation module is used to combine the operation layer model and the planning layer model to form an operation-planning double-layer model, and solve the operation-planning double-layer model through an improved particle swarm algorithm to obtain an optimal configuration scheme for distributed power sources;

[0052] The distributed power planning module is used to plan the installation location and installation capacity of the distributed power according to the distributed power optimal configuration plan.

[0053] A third aspect of the present application provides a terminal device, which includes: a terminal device including a processor and a memory, the memory storing a computer program, and the processor implementing the steps of a distributed power supply optimization configuration method for distribution network self-healing as described in any one of the embodiments of the present application when executing the computer program. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the implementation manner will be briefly introduced below. Obviously, the drawings described below are only some implementation manners of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0055] Figure 1 It is a specific flow chart of a distributed power supply optimization configuration method for distribution network self-healing provided by a certain embodiment of the present application;

[0056] Figure 2 It is a topological structure diagram of a unified self-healing strategy of a distributed power supply optimization configuration method for distribution network self-healing provided by a certain embodiment of the present application;

[0057] Figure 3 It is a schematic diagram of a distribution network load loss rate of a distributed power supply optimization configuration method for distribution network self-healing provided by a certain embodiment of the present application;

[0058] Figure 4 It is a schematic diagram of distribution network voltage safety of a distributed power supply optimization configuration method for distribution network self-healing provided by a certain embodiment of the present application;

[0059] Figure 5 It is a schematic diagram of the load recovery of a distribution network provided in a distributed power supply optimization configuration method for distribution network self-healing in a certain embodiment of the present application;

[0060] Figure 6 It is a specific structural diagram of a distributed power supply optimization configuration system for distribution network self-healing provided by a certain embodiment of the present application;

[0061] Figure 7 A structural diagram of a terminal device is provided for a certain embodiment of the present application. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0063] It should be understood that the step numbers used in this article are only for the convenience of description and are not intended to limit the order in which the steps are executed.

[0064] First embodiment

[0065] Failures in the distribution network and sudden changes in load will affect residents' daily electricity consumption and social production. Therefore, the current main research direction is to improve the self-healing ability of the distribution network after a failure by studying the capacity and site planning of distributed power sources. Distribution network reconstruction and island division are both methods to improve the self-healing ability of the distribution network. How to unify and combine these two methods to improve the stability of power grid operation is the research direction of the embodiments of this application.

[0066] like Figure 1 As shown, Figure 1 A specific flow chart of a distributed power supply optimization configuration method for distribution network self-healing is provided for a certain embodiment of the present application. The distributed power supply optimization configuration method for distribution network self-healing of this embodiment includes steps S1 to S4, which are described in detail as follows:

[0067] Step S1, based on the topological structure of the distribution network, by optimizing the distribution network flow and dividing the distributed power sources, an operation layer model is constructed with the goal of minimizing the total operation cost of the distribution network.

[0068] In the embodiment of the present application, an operation layer model is first constructed to improve the self-healing capability of the active distribution network under faults and load mutations. Then, in the operation layer model, the output of the distributed power source and the distribution network reconstruction plan are optimized with the lowest total operation cost of the distribution network as the objective function.

[0069] In the process of constructing the operation layer model, the distribution network self-healing strategy is constructed through the distribution network reconstruction and the division of distributed power sources. The network reconstruction is used to optimize the network flow to achieve the effect of reducing network loss and balancing load.

[0070] Considering the self-healing process of the distribution network under line faults and the self-healing process under normal operating conditions, the main purpose of the distribution network self-healing is to restore as much load as possible, and consider the self-healing cost of the distribution network under the premise of ensuring that the load is fully restored to power supply.

[0071] Therefore, in the embodiment of the present application, on the basis of the unified self-healing strategy, the cost of implementing the unified self-healing strategy is reduced as much as possible, so the objective function of the operation layer model is divided into three parts: the load reduction cost, self-healing cost and network loss cost generated by the distribution network during the self-healing process. Among them, considering the self-healing process of the distribution network under line faults in any period of time, the objective function is divided into several discrete time periods, while the objective function of the self-healing process of the distribution network under normal operating conditions is the full operation period. Each period of the two types of self-healing processes corresponds to the self-healing scheme of their respective time periods, and the objective function of the operation layer model is the cumulative value of the total cost of each time period, so that the self-healing strategy of the distribution network can reach the optimal value of the entire fault period.

[0072] Therefore, according to the historical operation data of the distribution network and the preset distribution network self-healing process set, the load reduction cost, self-healing cost and network loss cost generated by the distribution network in the self-healing process are constructed respectively. Then, the load reduction cost, self-healing cost and network loss cost are combined to construct the total operation cost of the distribution network. Finally, the operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network.

[0073] Among them, the objective function of the operating layer model is specifically expressed as:

[0074] minC ope =∑(C load +C re +C net );

[0075] In the formula, C ope is the total operating cost of the distribution network, C load For load reduction costs, C re is the self-healing cost, C net The network loss cost.

[0076] The load reduction cost is specifically expressed as:

[0077]

[0078] In the formula, C load For load reduction costs, N T is the total number of time intervals within the fault period, N load is the total number of load nodes in the distribution network, β1, β2, and β3 are all load weight coefficients, and S load1 , S load2 , S load3 are the load reduction amounts corresponding to each level, and Δt is the length of each time interval during the fault period;

[0079] The self-healing cost is specifically expressed as:

[0080]

[0081] In the formula, C re is the self-healing cost, c swtich is the switch operation cost coefficient, N line is the total number of branches in the distribution network, is the switch state of the branch with node i as the starting point and node j as the end point during the fault recovery period t, c DG is the output cost of distributed generation, N DG is the total number of nodes of distributed generation, is the output of distributed generation in period t;

[0082] The network loss cost is specifically expressed as:

[0083]

[0084] In the formula, C net is the network loss cost, c line is the network loss cost coefficient, is the active power loss of line ij during period t.

[0085] Specifically, the load reduction cost accounts for the highest proportion in the objective function. When calculating the load reduction cost, the weight level of the load in the distribution network should be taken into account to ensure the continuous power supply of important loads. The self-healing cost is also called the switch operation cost and the output cost of the distributed power source, which means that the economic operation of the distribution network switch is guaranteed under the premise of ensuring power supply, and the output is reasonably allocated.

[0086] Furthermore, according to the preset power flow constraints and the preset distributed power output constraints, the distribution network self-healing strategy is optimized to obtain a unified self-healing strategy. In the embodiment of the present application, the DistFlow power flow constraint of the radial network is selected, and a 0-1 variable representing line disconnection is introduced, so that the DistFlow power flow constraint is more applicable to the self-healing strategy based on the flexible transformation of the distribution network topology.

[0087] Then, according to the DistFlow flow constraint, the flow constraint of the distribution network is constructed. The specific expression is:

[0088]

[0089] In the formula, the flow from the starting node to the end node in the branch information is defined as the positive direction. They represent the sum of active power injected into node j in period t, the sum of active power outflow, the predicted active load, the sum of active outputs of all distributed generation sources, and the amount of active load reduction respectively; Represents the sum of reactive power injected into node j during period t, the sum of reactive power outflow, predicted reactive load, the sum of reactive output of all distributed generation sources, and reactive load reduction; They represent the sum of active and reactive line losses injected into the line of node j at time t; M represents a constant with a large value; represents the square term of the voltage at node i during period t; It represents the square term of the line current of the line with node i as the starting node and node j as the ending node in time period t.

[0090] In order to ensure that the power flow constraint of the constructed distribution network has the same topological structure as the original distribution network, the same connectivity distribution network radiation is required. The power flow constraint has the same topological structure as the original distribution network if and only if the following two conditions are met: 1. The graph model of the power flow constraint is radial, and each subgraph is a connected graph; 2. The number of branches of the graph model is equal to the number of nodes minus the given number of subgraphs. When the above two conditions are met, the power flow constraint also includes:

[0091]

[0092] In the formula, For the virtual flow of line js, is the virtual power flow of line ij, v is the graph node, The net virtual power injected into the distributed generation node.

[0093] In order to achieve the first condition, according to graph theory, a single commodity flow theory method can be used to represent the connectivity of the subgraph, because the virtual network corresponding to the flow constraint has the same topological structure as the original distribution network, so they have the same connectivity. In the embodiment of the present application, the distributed power source is used as the source node.

[0094] In addition, the distribution network self-healing strategy also needs to meet the output constraints of distributed power sources. The specific expression is:

[0095]

[0096] Where P DGjmin , P DGjmax are the lower and upper limits of the active output of distributed generation, respectively. is the power factor angle, and its upper limit is generally taken as a fixed value.

[0097] Based on the above-mentioned power flow constraints and distributed power output constraints, a unified self-healing strategy is obtained. Then, based on the unified self-healing strategy, an operation layer model for improving the self-healing ability of the active distribution network under faults and load mutations is constructed through the operation layer objective function. Among them, the unified self-healing strategy fully considers the coordination of distribution network islands and network reconstruction, and solves the problem that the single self-healing strategy has poor load recovery and power supply effect due to faults and load mutations.

[0098] Step S2: Based on the operation layer model, a planning layer model is constructed with the goal of minimizing the annual total cost of the distribution network.

[0099] In the embodiment of the present application, the installation location and installation capacity of the distributed power source are planned with the goal of minimizing the annual total cost of the distribution network, and a planning layer model is constructed.

[0100] Furthermore, the total annual cost of the distribution network includes the power investment cost of the installation and use of distributed generation, as well as the total operating cost of the distribution network at the operating level.

[0101] Among them, the objective function of the planning layer model is specifically expressed as:

[0102] minC plan =C inv +C ope ;

[0103] In the formula, C plan is the total annual cost of the distribution network, C inv is the distributed generation investment cost, C ope is the total operating cost of the distribution network.

[0104] The investment cost of the distributed power source is specifically expressed as:

[0105]

[0106] In the formula, C inv is the investment cost of distributed power generation, r is the discount rate, n is the useful life of distributed power generation, N DG is the number of distributed generation to be installed, c inv,DG is the unit capacity investment cost of distributed power source, in RMB / kW, P DGi max is the installed capacity of distributed generation at node i, in kW.

[0107] When considering the installed capacity and number of distributed power sources, it is also necessary to meet the distributed power capacity constraints to construct the distributed power investment cost. The specific expression is:

[0108]

[0109] In the formula, is the maximum installed capacity of the distributed generation at the i-th node.

[0110] Step S3, combining the operation layer model and the planning layer model to construct an operation-planning double-layer model, and solving the operation-planning double-layer model by using an improved particle swarm algorithm to obtain an optimal configuration solution for distributed power sources.

[0111] In the embodiment of the present application, the operation layer model and the planning layer model are first combined to form an operation-planning double-layer model, and then the traditional particle swarm algorithm is improved to obtain an improved particle swarm algorithm.

[0112] The particle swarm algorithm is an optimization algorithm that uses particles to simulate the predation behavior of birds, continuously updates the particle position and speed, and searches for the optimal solution of the target. The algorithm is widely used due to its fast convergence speed and strong search ability. The embodiment of the present application introduces inertia weight and learning factor to improve the particle swarm algorithm, further optimize the particle search ability, and improve the convergence of the operation.

[0113] In the traditional particle swarm algorithm, during the iteration, each particle completes the update by tracking the individual extreme value and the global extreme value, specifically:

[0114] v i (t+1)=wv i (t)+c1r1(pbest(i,t)-x i (t))+c2r2(p best,i -x i (t));

[0115] x i (t+1)=x i (t)+v i (t+1);

[0116] In the formula, v i is, w is the inertia weight, c1 and c2 are learning factors, r1 and r2 are, pbest(i,t) is the optimal position of the particle found during the tth iteration of the i-th particle, p best is the best position found by the group so far, x i is the current particle position.

[0117] The embodiment of the present application introduces inertia weight and learning factor to improve the particle swarm algorithm to enhance the algorithm's optimization ability and convergence. There are three important parameters in the speed iteration formula of the particle swarm algorithm: inertia weight w, learning factors c1, c2. The inertia weight w affects the degree of influence of the particle's previous flight speed on the current flight speed. The selection of inertia weight is of great significance for balancing the global search ability and local search ability of the algorithm. In the iterative process, the globality and locality of the algorithm must be considered, and appropriate inertia weights must be adopted for searching. Therefore, the embodiment of the present application utilizes a power exponential operator, adds it to the inertia weight, and based on the total number of iterations, allows each particle to expand the search space during the search, thereby increasing population diversity and improving the iterative convergence speed.

[0118] The dynamic change expression of the inertia weight after the introduction of the power exponential operator is:

[0119]

[0120] In the formula, g is the current iteration number, a, b, d are fixed parameters, g max is the maximum number of iterations.

[0121] In addition, it is also necessary to set an appropriate learning factor to improve the particle swarm algorithm so that particles can search. In the early stage, attention should be paid to the ability of individuals to recognize themselves, and in the later stage, attention should be paid to the ability of individuals to obtain social information. The learning factor set in the embodiment of the present application is specifically expressed as follows:

[0122] c1=(e1-e2)g / g max +e2;

[0123] c2=(f1-f2)g / g max +f2;

[0124] Where e1, e2, f1, and f2 are fixed parameters and are related to the solution scale of the algorithm.

[0125] The improved particle swarm algorithm and the iterative formulas for particle speed and position are the same as the original algorithm. Compared with the basic algorithm, when the speed formula of the particle swarm algorithm based on dynamic adjustment of inertia weight and learning factor is adjusted, the algorithm changes dynamically with the number of iterations, its global search capability is effectively improved, the convergence of particles is also enhanced, the complexity of solving the operation-planning two-layer model is reduced, and the rapid convergence of the algorithm is improved.

[0126] Then, based on the preset typical scenario data set, the installation location and installation capacity of the distributed generation in the operation-planning two-layer model are iterated by using the improved particle swarm algorithm to optimize the distribution network self-healing process under the typical scenario data set. The specific process is as follows:

[0127] 1) Input the distribution network parameters, use the K-means multi-scenario analysis method to process the annual historical data of wind, light and load, and use multiple typical days in different seasons and climates to determine the random characteristics of wind, light and load, obtain the typical day scenario data and construct a typical scenario data set;

[0128] 2) Initialize the particle position and particle velocity, that is, the installation position and installation capacity of the distributed generation in the planning layer, as the input of the operation layer;

[0129] 3) Taking the optimization objectives and constraints of the operation layer as input, optimize the self-healing process of the distribution network at the operation layer and feed it back to the planning layer;

[0130] 4) The planning layer receives feedback data from the operation layer, calculates the constraints and objective functions of the planning layer, and optimizes and updates the installation location and capacity of the distributed generation;

[0131] 5) Update the particle population and repeat the above steps 2) to 4);

[0132] 6) If the preset convergence conditions are met, the improved particle swarm algorithm is terminated and the optimal installation location and optimal installation capacity of the distributed generation are obtained; otherwise, step 5) is repeated.

[0133] According to the distribution network topology and the optimal installation location and optimal installation capacity, the optimal configuration scheme of distributed power generation is constructed. Through the operation-planning two-layer model, the planning layer and the operation layer interact with each other to jointly decide on the optimal configuration scheme of distributed power generation of the distribution network that can improve the self-healing ability and operation stability.

[0134] Step S4: planning the installation location and installation capacity of the distributed power source according to the optimal configuration plan of the distributed power source.

[0135] In an embodiment of the present application, through the distributed power supply optimal configuration plan, the installation location and installation capacity of the distributed power supply are planned in the distribution network, and additional distributed power supply configuration is added so that it can achieve island operation for the fault area when connected to the distribution network, thereby ensuring the stability and safety of the voltage in the area, and quickly restoring the power supply to key loads to improve the quality of electricity use for users.

[0136] The reasonable distributed power planning provided by the distributed power optimal configuration solution can reduce the power supply recovery time and improve the reliability and stability of the overall system, greatly improve the self-healing ability of the distribution network system, and optimize the power quality of users.

[0137] Implementing the embodiments of the present application has the following beneficial effects:

[0138] The embodiment of the present application is based on the topological structure of the power grid. By simultaneously considering the reconstruction of the distribution network flow and the division of distributed power sources, the operation layer model is constructed to reduce the total operation cost of the distribution network as much as possible while improving the self-healing ability of the distribution network; then, based on the cost target of the operation layer model, a planning layer model with the lowest annual total cost of the distribution network is constructed. The two models are then combined to form a two-layer model with influence between layers. By improving the particle swarm algorithm, the optimal configuration scheme of distributed power sources that can minimize the total cost of the distribution network is found on the basis of improving the self-healing ability of the distribution network, so as to achieve the optimization of the distributed power configuration while improving the economy of the distribution network, greatly improve the self-healing ability of the distribution network after a fault and reduce the cost required to optimize the self-healing ability, and improve the operation stability of the distribution network and the quality of electricity used by users.

[0139] Second embodiment

[0140] Furthermore, in order to verify the distributed power supply optimization configuration effect of the above method embodiment, the embodiment of the present application selects the IEEE33-node distribution network for verification.

[0141] The distribution network provided in the embodiment of the present application includes 33 nodes, 32 segment switches and 5 contact switches, the reference voltage is 12.66kV, the reference power is 1000MV, and the photovoltaic units are connected at nodes {7, 13, 27}, with capacities of 500, 300 and 400kV respectively; the wind turbine units are connected at nodes {10, 16, 17, 30, 33}, with capacities of 500, 300, 200, 200 and 300kW. The computing platform of the embodiment of the present application adopts the MATLAB-YALMIP platform, and the CPLEX solver is called for solution. The micro gas turbine is a planned distributed power source, and the nodes to be selected are {2, 3, 4, 6, 8, 10, 12, 15, 18, 20, 26, 27, 29, 30, 31, 32}. The rated capacity of a single micro gas turbine is 50kW, and the upper limit of installation per node is 15 units, and 2 places are to be connected. The operation layer model considers the single line fault operation in any period of time and the typical daily routine operation scenario without fault. In the embodiment of the present application, it is assumed that the probability of line failure is the same and the line maintenance recovery time is 4 hours.

[0142] Figure 2 This is a topological diagram of the unified self-healing strategy of the distribution network after the line fault. It records that the unified self-healing strategy is adopted after the line fault, which reduces the load reduction of the distribution network within 4 hours and keeps the voltage level in a safe range. As shown in the figure, the red line indicates the line disconnection state, the upper left corner is the distribution network recovery in the first hour, the upper right corner is the distribution network recovery in the second hour, the lower left corner is the distribution network recovery in the third hour, and the lower right corner is the distribution network recovery in the fourth hour.

[0143] Figure 3The node voltage values ​​of the distribution network are provided. In the figure, after the fault, the line maintenance recovery time is divided into 4 intervals per hour. The figure reflects the voltage of nodes 1-37 of the distribution network. The blue line, orange line, yellow line and purple line in the figure represent the node voltage in different time periods. After the fault occurs in the distribution network branch, the proposed method keeps the voltage level of each node in a safe range.

[0144] Figure 4 The load loss rate of the distribution network is provided, which is also called the probability of power shortage, and is used to reflect the available power capacity and reliability of the power system. The blue solid line in the figure represents the predicted load that can maintain the stable operation of the distribution network, the red dotted line represents the actual supply load of the distribution network, the horizontal axis is the time interval after the failure of the distribution network branch, and the vertical axis is the load value of the distribution network. In the early stage after the failure of the distribution network branch, the load that the distribution network can provide is obviously lower than expected, but under the implementation of the unified self-healing strategy, it can be restored to the predicted load within 4 hours, and quickly recover to stable operation after the failure, which greatly improves the operation stability and self-healing ability of the distribution network.

[0145] Furthermore, the following Table 1 shows the optimal configuration scheme of some distributed power sources of the embodiment of the present application obtained based on the unified self-healing strategy and the operation-planning two-layer model.

[0146] Table 1 Optimal installation location and optimal installation capacity of distributed power generation

[0147]

[0148]

[0149] Based on the above table, in order to verify the effectiveness of the optimal configuration scheme of distributed power sources in the embodiment of the present application in improving the self-healing capability of the distribution network, the following two self-healing schemes are considered:

[0150] Solution 1: Considering the self-healing operation of the distribution network with distributed generation;

[0151] Solution 2: Self-healing operation of the distribution network without considering distributed generation.

[0152] That is, the difference between the two schemes is that Scheme 1 adopts the optimal configuration scheme of distributed power sources, while Scheme 2 does not adopt the optimal configuration scheme of distributed power sources.

[0153] Figure 5The load recovery of the distribution network in fault scenarios under two schemes is provided. The figure above shows the load recovery of the distribution network after adding distributed power sources based on scheme 1. It can be seen that although the distribution network failed to reach the expected load at the beginning of the fault, the gap with the expected load was not large. During the period of line maintenance and restoration, the load can quickly return to stability. The full load supply can be restored in the second hour, and the self-healing ability of the distribution network is significantly improved; the figure below shows the load recovery of the distribution network based on scheme 2 without adding distributed power sources. It can be seen that there is a significant load reduction during the entire fault recovery time, which greatly reduces the user's power consumption experience.

[0154] Therefore, by adding distributed power sources to the distribution network, it is possible to independently supply power to the surrounding areas under operating conditions facing faults and load mutations, forming an island operation state. While ensuring voltage stability and safety, the load can be quickly and effectively restored, greatly improving the self-healing ability of the distribution network system, and verifying the effectiveness of the optimized configuration of distributed power sources in supplying power to distribution network users and the actual social welfare effect.

[0155] Third embodiment

[0156] Furthermore, in order to execute the distributed power supply optimization configuration system for distribution network self-healing corresponding to the above method embodiment to achieve corresponding functions and technical effects, Figure 6 A structural diagram of a distributed power supply optimization configuration system for distribution network self-healing is provided. For ease of description, only the parts related to this embodiment are shown. The distributed power supply optimization configuration system for distribution network self-healing provided by the embodiment of the present application includes:

[0157] The operation layer model building module 201 is used to build an operation layer model with the goal of minimizing the total operation cost of the distribution network by optimizing the distribution network flow and dividing the distributed power sources according to the distribution network topology.

[0158] In an embodiment of the present application, according to preset power flow constraints and preset distributed power source output constraints, the power flow distribution of the distribution network topology and the division of distributed power sources are optimized to obtain a unified self-healing strategy; according to the historical operation data of the distribution network, an operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network; and the operation layer model is constructed through the unified self-healing strategy and the operation layer objective function.

[0159] The planning layer model building module 202 is used to build a planning layer model with the goal of minimizing the annual total cost of the distribution network based on the operation layer model.

[0160] In an embodiment of the present application, based on the preset distributed power capacity constraints, the distributed power investment cost is constructed taking into account the installed capacity and installation quantity of the distributed power. The distributed power investment cost and the total operating cost of the distribution network of the operating layer are combined to construct the annual total cost of the distribution network. With the goal of minimizing the annual total cost of the distribution network, a target planning layer model is constructed.

[0161] The optimal configuration solution generation module 203 is used to combine the operation layer model and the planning layer model to construct an operation-planning double-layer model, and solve the operation-planning double-layer model through an improved particle swarm algorithm to obtain an optimal configuration solution for distributed power sources.

[0162] In an embodiment of the present application, based on a preset typical scenario data set, the installation location and installation capacity of the distributed power source in the operation-planning two-layer model are iterated by an improved particle swarm algorithm to optimize the distribution network self-healing process under the typical scenario data set; when the particle swarm algorithm terminates, the optimal installation location and optimal installation capacity of the distributed power source are obtained; and based on the distribution network topology and the optimal installation location and optimal installation capacity, an optimal configuration plan for the distributed power source is constructed.

[0163] The distributed power planning module 204 is used to plan the installation location and installation capacity of the distributed power according to the distributed power optimal configuration solution.

[0164] In an embodiment of the present application, through the distributed power supply optimal configuration plan, the installation location and installation capacity of the distributed power supply are planned in the distribution network, and additional distributed power supply configuration is added so that it can achieve island operation for the fault area when connected to the distribution network, thereby ensuring the stability and safety of the voltage in the area, and quickly restoring the power supply to key loads to improve the quality of electricity use for users.

[0165] In some embodiments, the operation layer model building module 201 further includes:

[0166] First, an operation layer model is constructed to improve the self-healing ability of the active distribution network under faults and load mutations. Then, in the operation layer model, the output of distributed generation and the distribution network reconstruction plan are optimized with the lowest total operation cost of the distribution network as the objective function.

[0167] In the process of constructing the operation layer model, the distribution network self-healing strategy is constructed through the distribution network reconstruction and the division of distributed power sources. The network reconstruction is used to optimize the network flow to achieve the effect of reducing network loss and balancing load.

[0168] Considering the self-healing process of the distribution network under line faults and the self-healing process under normal operating conditions, the main purpose of the distribution network self-healing is to restore as much load as possible, and consider the self-healing cost of the distribution network under the premise of ensuring that the load is fully restored to power supply.

[0169] Therefore, in the embodiment of the present application, on the basis of the unified self-healing strategy, the cost of implementing the unified self-healing strategy is reduced as much as possible, so the objective function of the operation layer model is divided into three parts: the load reduction cost, self-healing cost and network loss cost generated by the distribution network during the self-healing process. Among them, considering the self-healing process of the distribution network under line faults in any period of time, the objective function is divided into several discrete time periods, while the objective function of the self-healing process of the distribution network under normal operating conditions is the full operation period. Each period of the two types of self-healing processes corresponds to the self-healing scheme of their respective time periods, and the objective function of the operation layer model is the cumulative value of the total cost of each time period, so that the self-healing strategy of the distribution network can reach the optimal value of the entire fault period.

[0170] Therefore, according to the historical operation data of the distribution network and the preset distribution network self-healing process set, the load reduction cost, self-healing cost and network loss cost generated by the distribution network in the self-healing process are constructed respectively. Then, the load reduction cost, self-healing cost and network loss cost are combined to construct the total operation cost of the distribution network. Finally, the operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network.

[0171] Among them, the objective function of the operating layer model is specifically expressed as:

[0172] minC ope =∑(C load +C re +C net );

[0173] In the formula, C ope is the total operating cost of the distribution network, C load For load reduction costs, C re is the self-healing cost, C net The network loss cost.

[0174] The load reduction cost is specifically expressed as:

[0175]

[0176] In the formula, C load For load reduction costs, N T is the total number of time intervals within the fault period, N load is the total number of load nodes in the distribution network, β1, β2, and β3 are all load weight coefficients, and S load1 , S load2 , S load3are the load reduction amounts corresponding to each level, and Δt is the length of each time interval during the fault period;

[0177] The self-healing cost is specifically expressed as:

[0178]

[0179] In the formula, C re is the self-healing cost, c swtich is the switch operation cost coefficient, N line is the total number of branches in the distribution network, is the switch state of the branch with node i as the starting point and node j as the end point during the fault recovery period t, c DG is the output cost of distributed generation, N DG is the total number of nodes of distributed generation, is the output of distributed generation in period t;

[0180] The network loss cost is specifically expressed as:

[0181]

[0182] In the formula, C net is the network loss cost, c line is the network loss cost coefficient, is the active power loss of line ij during period t.

[0183] Specifically, the load reduction cost accounts for the highest proportion in the objective function. When calculating the load reduction cost, the weight level of the load in the distribution network should be taken into account to ensure the continuous power supply of important loads. The self-healing cost is also called the switch operation cost and the output cost of the distributed power source, which means that the economic operation of the distribution network switch is guaranteed under the premise of ensuring power supply, and the output is reasonably allocated.

[0184] Furthermore, according to the preset power flow constraints and the preset distributed power output constraints, the distribution network self-healing strategy is optimized to obtain a unified self-healing strategy. In the embodiment of the present application, the DistFlow power flow constraint of the radial network is selected, and a 0-1 variable representing line disconnection is introduced, so that the DistFlow power flow constraint is more applicable to the self-healing strategy based on the flexible transformation of the distribution network topology.

[0185] Then, according to the DistFlow flow constraint, the flow constraint of the distribution network is constructed. The specific expression is:

[0186]

[0187]

[0188] In the formula, the flow from the starting node to the end node in the branch information is defined as the positive direction. They represent the sum of active power injected into node j in period t, the sum of active power outflow, the predicted active load, the sum of active outputs of all distributed generation sources, and the amount of active load reduction respectively; Represents the sum of reactive power injected into node j during period t, the sum of reactive power outflow, predicted reactive load, the sum of reactive output of all distributed generation sources, and reactive load reduction; They represent the sum of active and reactive line losses injected into the line of node j at time t; M represents a constant with a large value; represents the square term of the voltage at node i during period t; It represents the square term of the line current of the line with node i as the starting node and node j as the ending node in time period t.

[0189] In order to ensure that the power flow constraint of the constructed distribution network has the same topological structure as the original distribution network, the same connectivity distribution network radiation is required. The power flow constraint has the same topological structure as the original distribution network if and only if the following two conditions are met: 1. The graph model of the power flow constraint is radial, and each subgraph is a connected graph; 2. The number of branches of the graph model is equal to the number of nodes minus the given number of subgraphs. When the above two conditions are met, the power flow constraint also includes:

[0190]

[0191]

[0192] In the formula, For the virtual flow of line js, is the virtual power flow of line ij, v is the graph node, The net virtual power injected into the distributed generation node.

[0193] In order to achieve the first condition, according to graph theory, a single commodity flow theory method can be used to represent the connectivity of the subgraph, because the virtual network corresponding to the flow constraint has the same topological structure as the original distribution network, so they have the same connectivity. In the embodiment of the present application, the distributed power source is used as the source node.

[0194] In addition, the distribution network self-healing strategy also needs to meet the output constraints of distributed power sources. The specific expression is:

[0195]

[0196] Where P DGjmin , P DGjmax are the lower and upper limits of the active output of distributed generation, respectively. is the power factor angle, and its upper limit is generally taken as a fixed value.

[0197] Based on the above-mentioned power flow constraints and distributed power output constraints, a unified self-healing strategy is obtained. Then, based on the unified self-healing strategy, an operation layer model for improving the self-healing ability of the active distribution network under faults and load mutations is constructed through the operation layer objective function. Among them, the unified self-healing strategy fully considers the coordination of distribution network islands and network reconstruction, and solves the problem that the single self-healing strategy has poor load recovery and power supply effect due to faults and load mutations.

[0198] In some embodiments, the planning layer model building module 202 is specifically:

[0199] In the embodiment of the present application, the installation location and installation capacity of the distributed power source are planned with the goal of minimizing the annual total cost of the distribution network, and a planning layer model is constructed.

[0200] Furthermore, the total annual cost of the distribution network includes the power investment cost of the installation and use of distributed power sources, as well as the total operating cost of the distribution network at the operating level.

[0201] Among them, the objective function of the planning layer model is specifically expressed as:

[0202] minC plan =C inv +C ope ;

[0203] In the formula, C plan is the total annual cost of the distribution network, C inv is the distributed generation investment cost, C ope is the total operating cost of the distribution network.

[0204] The investment cost of the distributed power source is specifically expressed as:

[0205]

[0206] In the formula, C inv is the investment cost of distributed power generation, r is the discount rate, n is the useful life of distributed power generation, N DG is the number of distributed generation to be installed, c inv,DG is the unit capacity investment cost of distributed power source, in RMB / kW, P DGi max is the installed capacity of distributed generation at node i, in kW.

[0207] When considering the installed capacity and number of distributed power sources, it is also necessary to meet the distributed power capacity constraints to construct the distributed power investment cost. The specific expression is:

[0208]

[0209] In the formula, is the maximum installed capacity of the distributed generation at the i-th node.

[0210] In some embodiments, the optimal configuration solution generation module 203 is specifically:

[0211] In the embodiment of the present application, the operation layer model and the planning layer model are first combined to form an operation-planning double-layer model, and then the traditional particle swarm algorithm is improved to obtain an improved particle swarm algorithm.

[0212] The particle swarm algorithm is an optimization algorithm that uses particles to simulate the predation behavior of birds, continuously updates the particle position and speed, and searches for the optimal solution of the target. The algorithm is widely used due to its fast convergence speed and strong search ability. The embodiment of the present application introduces inertia weight and learning factor to improve the particle swarm algorithm, further optimize the particle search ability, and improve the convergence of the operation.

[0213] In the traditional particle swarm algorithm, during the iteration, each particle completes the update by tracking the individual extreme value and the global extreme value, specifically:

[0214] v i (t+1)=wv i (t)+c1r1(pbest(i,t)-x i (t))+c2r2(p best,i -x i (t));

[0215] x i (t+1)=x i (t)+v i (t+1);

[0216] In the formula, v i is, w is the inertia weight, c1 and c2 are learning factors, r1 and r2 are, pbest(i,t) is the optimal position of the particle found during the tth iteration of the i-th particle, p best is the best position found by the group so far, x i is the current particle position.

[0217] The embodiment of the present application introduces inertia weight and learning factor to improve the particle swarm algorithm to enhance the algorithm's optimization ability and convergence. There are three important parameters in the speed iteration formula of the particle swarm algorithm: inertia weight w, learning factors c1, c2. The inertia weight w affects the degree of influence of the particle's previous flight speed on the current flight speed. The selection of inertia weight is of great significance for balancing the global search ability and local search ability of the algorithm. In the iterative process, the globality and locality of the algorithm must be considered, and appropriate inertia weights must be adopted for searching. Therefore, the embodiment of the present application utilizes a power exponential operator, adds it to the inertia weight, and based on the total number of iterations, allows each particle to expand the search space during the search, thereby increasing population diversity and improving the iterative convergence speed.

[0218] The dynamic change expression of the inertia weight after the introduction of the power exponential operator is:

[0219]

[0220] In the formula, g is the current iteration number, a, b, d are fixed parameters, g max is the maximum number of iterations.

[0221] In addition, it is also necessary to set an appropriate learning factor to improve the particle swarm algorithm so that particles can search. In the early stage, attention should be paid to the ability of individuals to recognize themselves, and in the later stage, attention should be paid to the ability of individuals to obtain social information. The learning factor set in the embodiment of the present application is specifically expressed as follows:

[0222] c1=(e1-e2)g / g max +e2;

[0223] c2=(f1-f2)g / g max +f2;

[0224] Where e1, e2, f1, and f2 are fixed parameters and are related to the solution scale of the algorithm.

[0225] The improved particle swarm algorithm and the iterative formulas for particle speed and position are the same as the original algorithm. Compared with the basic algorithm, when the speed formula of the particle swarm algorithm based on dynamic adjustment of inertia weight and learning factor is adjusted, the algorithm changes dynamically with the number of iterations, its global search capability is effectively improved, the convergence of particles is also enhanced, the complexity of solving the operation-planning two-layer model is reduced, and the rapid convergence of the algorithm is improved.

[0226] Then, based on the preset typical scenario data set, the installation location and installation capacity of the distributed generation in the operation-planning two-layer model are iterated by using the improved particle swarm algorithm to optimize the distribution network self-healing process under the typical scenario data set. The specific process is as follows:

[0227] 1) Input the distribution network parameters, use the K-means multi-scenario analysis method to process the annual historical data of wind, light and load, and use multiple typical days in different seasons and climates to determine the random characteristics of wind, light and load, obtain the typical day scenario data and construct a typical scenario data set;

[0228] 2) Initialize the particle position and particle velocity, that is, the installation position and installation capacity of the distributed generation in the planning layer, as the input of the operation layer;

[0229] 3) Taking the optimization objectives and constraints of the operation layer as input, optimize the self-healing process of the distribution network at the operation layer and feed it back to the planning layer;

[0230] 4) The planning layer receives feedback data from the operation layer, calculates the constraints and objective functions of the planning layer, and optimizes and updates the installation location and capacity of the distributed generation;

[0231] 5) Update the particle population and repeat the above steps 2) to 4);

[0232] 6) If the preset convergence conditions are met, the improved particle swarm algorithm is terminated and the optimal installation location and optimal installation capacity of the distributed generation are obtained; otherwise, step 5) is repeated.

[0233] According to the distribution network topology and the optimal installation location and optimal installation capacity, the optimal configuration scheme of distributed power generation is constructed. Through the operation-planning two-layer model, the planning layer and the operation layer interact with each other to jointly decide on the optimal configuration scheme of distributed power generation of the distribution network that can improve the self-healing ability and operation stability.

[0234] Implementing the embodiments of the present application has the following beneficial effects:

[0235] The embodiment of the present application is based on the topological structure of the power grid. By simultaneously considering the reconstruction of the distribution network flow and the division of distributed power sources, the operation layer model is constructed to reduce the total operation cost of the distribution network as much as possible while improving the self-healing ability of the distribution network; then, based on the cost target of the operation layer model, a planning layer model with the lowest annual total cost of the distribution network is constructed. The two models are then combined to form a two-layer model with influence between layers. By improving the particle swarm algorithm, the optimal configuration scheme of distributed power sources that can minimize the total cost of the distribution network is found on the basis of improving the self-healing ability of the distribution network, so as to achieve the optimization of the distributed power configuration while improving the economy of the distribution network, greatly improve the self-healing ability of the distribution network after a fault and reduce the cost required to optimize the self-healing ability, and improve the operation stability of the distribution network and the quality of electricity used by users.

[0236] Furthermore, Figure 7 This is a structural diagram of a terminal device provided in one embodiment of the present application. Figure 7As shown, the terminal device 3 of this embodiment includes: at least one processor 30 (in Figure 7 Only one is shown) and a memory 31 and a computer program 32 stored in the memory 31 and executable on the at least one processor. When the processor 30 executes the computer program 32, the steps of a distributed power source optimization configuration method for distribution network self-healing as described in any one of the embodiments of the present application can be implemented.

[0237] The terminal device 3 may be a computing device such as a desktop computer, a cloud server, or a laptop computer. The computing device may include but is not limited to a processor 30 and a memory 31 . Figure 7 This is merely an example of the terminal device 3 and does not constitute a limitation on the terminal device 3 , which may include more or fewer components than those shown in the figure.

[0238] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A distributed power supply optimization configuration method for distribution network self-healing, characterized in that: include: According to the topological structure of the distribution network, by optimizing the power flow of the distribution network and dividing the distributed power sources, an operation layer model is constructed with the goal of minimizing the total operation cost of the distribution network; Based on the operation layer model, a planning layer model is constructed with the goal of minimizing the total annual cost of the distribution network; Combining the operation layer model and the planning layer model, constructing an operation-planning double-layer model, and solving the operation-planning double-layer model by using an improved particle swarm algorithm to obtain an optimal configuration solution for distributed power sources; According to the optimal configuration plan of distributed power sources, the installation location and installation capacity of distributed power sources are planned.

2. The distributed power supply optimization configuration method for distribution network self-healing according to claim 1 is characterized in that: According to the distribution network topology, by optimizing the distribution network flow and dividing the distributed power sources, an operation layer model with the goal of minimizing the total operation cost of the distribution network is constructed, specifically: According to the preset power flow constraints and the preset distributed power output constraints, the power flow distribution of the distribution network topology and the division of distributed power sources are optimized to obtain a unified self-healing strategy; According to the historical operation data of the distribution network, the operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network; The operation layer model is constructed by using the unified self-healing strategy and the operation layer objective function.

3. The distributed power supply optimization configuration method for distribution network self-healing according to claim 2 is characterized in that: According to the historical operation data of the distribution network, the operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network, which is specifically: According to the historical operation data of the distribution network and the preset distribution network self-healing process set, the load reduction cost, self-healing cost and network loss cost generated by the distribution network in the self-healing process are constructed respectively; Combine the load reduction cost, self-healing cost and network loss cost to construct the total operating cost of the distribution network; The operation layer objective function is constructed with the goal of minimizing the total operation cost of the distribution network.

4. The distributed power supply optimization configuration method for distribution network self-healing according to claim 3 is characterized in that: The load reduction cost, self-healing cost and network loss cost are specifically: The load reduction cost is specifically expressed as follows: In the formula, C load For load reduction costs, N T is the total number of time intervals within the fault period, N load is the total number of load nodes in the distribution network, β1, β2, and β3 are all load weight coefficients, and S load1 , S load2 , S load3 are the load reduction amounts corresponding to each level, and Δt is the length of each time interval during the fault period; The self-healing cost is specifically expressed as: In the formula, C re is the self-healing cost, c swtich is the switch operation cost coefficient, N line is the total number of branches in the distribution network, is the switch state of the branch with node i as the starting point and node j as the end point during the fault recovery period t, c DG is the output cost of distributed generation, N DG is the total number of nodes of distributed generation, is the output of distributed generation in period t; The network loss cost is specifically expressed as: In the formula, C net is the network loss cost, c line is the network loss cost coefficient, is the active power loss of line ij during period t.

5. The distributed power supply optimization configuration method for distribution network self-healing according to claim 1 is characterized in that: Based on the operation layer, a planning layer model is constructed with the goal of minimizing the total annual cost of the distribution network, specifically: According to the preset distributed power capacity constraints, considering the installed capacity and number of distributed power, the distributed power investment cost is constructed; Combine the distributed generation investment cost and the total operating cost of the distribution network at the operating layer to construct the total annual cost of the distribution network; With the goal of minimizing the annual total cost of the distribution network, a target planning layer model is constructed.

6. The distributed power supply optimization configuration method for distribution network self-healing according to claim 5 is characterized in that: The distributed power generation investment cost is specifically: The investment cost of the distributed power source is specifically expressed as follows: In the formula, C inv is the investment cost of distributed power generation, r is the discount rate, n is the useful life of distributed power generation, N DG is the number of distributed generation to be installed, c inv,DG is the unit capacity investment cost of distributed power generation, P DGimax Install the capacity of the distributed generation at node i.

7. The distributed power supply optimization configuration method for distribution network self-healing according to claim 1 is characterized in that: The improved particle swarm algorithm is used to solve the operation-planning double-layer model to obtain the optimal configuration scheme of distributed power sources, which is specifically: Based on a preset typical scenario data set, the installation location and installation capacity of the distributed generation in the operation-planning two-layer model are iterated by using an improved particle swarm algorithm to optimize the distribution network self-healing process under the typical scenario data set; When the particle swarm algorithm is terminated, the optimal installation location and optimal installation capacity of the distributed power source are obtained; According to the distribution network topology and the optimal installation location and optimal installation capacity, an optimal configuration plan for distributed power sources is constructed.

8. The distributed power supply optimization configuration method for distribution network self-healing according to claim 7 is characterized in that: The improved particle swarm algorithm is specifically: A preset power exponential operator is introduced into the inertia weight of the particle swarm optimization algorithm to obtain an improved particle swarm algorithm; wherein the inertia weight acts on the iterative convergence speed of the installation position and the installation capacity.

9. A distributed power optimization configuration system for distribution network self-healing, characterized in that: include: Operation layer model building module, planning layer model building module, optimal configuration solution generation module and distributed power generation planning module; The operation layer model building module is used to build an operation layer model with the goal of minimizing the total operation cost of the distribution network by optimizing the distribution network flow and dividing the distributed power sources according to the distribution network topology; The planning layer model building module is used to build a planning layer model with the goal of minimizing the annual total cost of the distribution network based on the operation layer model; The optimal configuration scheme generation module is used to combine the operation layer model and the planning layer model to form an operation-planning double-layer model, and solve the operation-planning double-layer model through an improved particle swarm algorithm to obtain an optimal configuration scheme for distributed power sources; The distributed power planning module is used to plan the installation location and installation capacity of the distributed power according to the distributed power optimal configuration plan.

10. A terminal device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a distributed power supply optimization configuration method for distribution network self-healing according to any one of claims 1 to 8 are implemented.

Images