Substation supply area division method considering distributed photovoltaic consumption

Abstract

The invention discloses a transformer substation supply area division method considering distributed photovoltaic consumption, and the method comprises the following steps: S1, building a bilevel planning model with the division distance and minimum abandoned light as targets; s2, performing connectivity test on the land parcels in the supply area by adopting a union set checking algorithm; s3, establishing an improved genetic algorithm based on a graph theory, and generating and optimizing a donor region division scheme meeting connectivity constraints; and S4, proposing a bilayer programming model solving algorithm based on an improved genetic algorithm, and outputting an optimal supply zone division scheme through iterative optimization. Through the bilevel planning model and the improved genetic algorithm, photovoltaic consumption optimization and supply area connectivity guarantee are realized, and the method has the advantages of high efficiency, reliability and economical efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply area division, and in particular to a power supply area division method for a substation considering distributed photovoltaic consumption. Background Art

[0002] In the current power system planning and operation, distributed photovoltaic power generation, as an important form of renewable energy, has received widespread attention and application. With the country's increasing support for clean energy, especially under the promotion of policies such as the "Urban Rooftop Photovoltaic Action", the penetration rate of photovoltaic power generation in the power system has increased significantly. This trend has led to fundamental changes in the load characteristics of the power grid, and traditional power system planning methods are facing unprecedented challenges.

[0003] In the prior art, the methods for dividing the supply area of ​​substations are mainly based on the load density of the planning area. These methods usually assume that the load distribution is relatively uniform, and do not fully consider the volatility of photovoltaic power generation and the impact of reverse flow on the operation of the power grid. In addition, the existing supply area division methods are mostly based on a single objective function, usually only focusing on the optimization of power supply reliability and power supply cost, while ignoring the problem of photovoltaic power generation. Since the power generation capacity of distributed photovoltaics is greatly affected by natural conditions, its output has significant intermittent and volatile characteristics. When the photovoltaic power generation power exceeds the local load, it is easy to cause reverse flow, which in turn affects the voltage stability and power quality of the power grid. In this case, if the supply area division of the substation fails to fully consider the characteristics of photovoltaic power generation, it may lead to serious abandonment of light in the power supply area, reduce the utilization rate of photovoltaic power generation, and violate the original intention of energy conservation, emission reduction and efficient use of clean energy.

[0004] In addition, the existing supply area division methods often ignore the connectivity issues between the plots in the supply area during the optimization process. In power grid planning, the connectivity of the power supply area is a basic requirement to ensure safe and reliable power supply. If the plots in the power supply area fail to form a connected power network, it may cause insufficient power supply or power outages in local areas, thus affecting the power supply quality and stability of the entire supply area. However, the traditional genetic algorithm fails to consider the connectivity constraints when performing crossover and mutation operations, and the generated offspring individuals often do not meet the connectivity requirements, thus affecting the final optimization results.

[0005] In response to the above problems, some relevant research has been carried out in the existing technology. For example, a study proposed a supply area division method considering distributed photovoltaics, attempting to optimize the photovoltaic power generation capacity by adjusting the supply area boundaries. However, most of these methods are still based on static load forecasting models, lack in-depth analysis of the dynamic characteristics of photovoltaic power generation, and fail to achieve a quantitative assessment of the photovoltaic power generation capacity. In addition, the research on supply area connectivity in the existing technology is also relatively limited. Usually, connectivity tests are only performed in the later stages of supply area division, and connectivity constraints are not fully taken into account in the algorithm design stage. This post-test method often leads to low algorithm efficiency, and the optimization results are difficult to meet the actual application needs.

[0006] Therefore, how to provide a method for substation supply area division that takes distributed photovoltaic consumption into consideration is an urgent problem that technicians in this field need to solve. Summary of the invention

[0007] One purpose of the present invention is to propose a method for dividing the supply area of ​​a substation taking into account distributed photovoltaic absorption. The present invention uses a two-layer planning model combined with an improved genetic algorithm to innovatively solve the difficult problems of distributed photovoltaic absorption and substation supply area division. By dynamically evaluating the absorption capacity of photovoltaic power generation and detecting the connectivity of the supply area in real time, the present invention effectively reduces the phenomenon of abandoned light and improves the efficiency of photovoltaic utilization. At the same time, the improved algorithm ensures that the supply area division scheme has higher power supply security and reliability while optimizing economic efficiency. This method shows significant advantages in the context of large-scale photovoltaic access, and has the advantages of high efficiency, reliability, and economy, providing advanced technical means for power system planning.

[0008] A method for dividing a substation supply area considering distributed photovoltaic consumption according to an embodiment of the present invention includes the following steps:

[0009] S1. Construct a two-level planning model with the goal of minimizing the partition distance and abandoned light;

[0010] S2, using the check-union algorithm to check the connectivity of the plots in the supply area;

[0011] S3. Establish an improved genetic algorithm based on graph theory to generate and optimize the supply area division scheme that meets the connectivity constraints;

[0012] S4. A two-level programming model solving algorithm based on an improved genetic algorithm is proposed, and the optimal supply area division scheme is output through iterative optimization.

[0013] Optionally, the two-layer planning model includes an upper layer and a lower layer, the upper layer is a supply division model, and the lower layer is a simulation operation model. The upper layer transmits the supply area division plan to the lower layer, and after the lower layer receives the supply area division plan and calculates the operating cost, the calculated operating cost data is provided to the upper layer for optimizing the supply area division.

[0014] Optionally, the objective function of the supply division model is to minimize the combined supply area division distance index and the operating cost:

[0015] minF=λ div D div +λ ope C ope ;

[0016]

[0017]

[0018] Where F represents the objective function of the supply area division model, λ div Represents the weight coefficient of the distance indicator, λ ope Represents the weight coefficient of operating cost, D div It represents the distance index from the land to the substation after the supply area is divided, C ope represents the operating cost under the supply area division, NS represents the number of substations in the planning area, NP represents the number of plots in the planning area, and d ij represents the straight-line distance from plot i to substation j, represents the unit cost of abandoned light, c loss represents the unit cost of network loss, Indicates the total amount of abandoned photovoltaic power in the simulation operation, Indicates the total network loss power in the simulation operation, Indicates the total amount of abandoned photovoltaic power in the simulation operation, Indicates the total network loss in the simulation operation; a ij Indicates the supply area of ​​the plot, which is a 0 / 1 variable. If plot i belongs to the supply area of ​​substation j, then a ij =1, otherwise a ij =0;

[0019] The supply area division model constraints include single power supply constraints, connectivity constraints and substation capacity constraints;

[0020] The single power supply constraint stipulates that any plot can only be powered by one substation:

[0021]

[0022] The connectivity constraint stipulates that the power supply area of ​​the same substation should meet the connectivity:

[0023]

[0024] in, It represents the judgment function of the connected component of the supply area of ​​the strain station j. When the connected component = 1, the supply area meets the connectivity;

[0025] The substation capacity constraint states that the load supplied by the substation cannot exceed the rated capacity:

[0026]

[0027] in, represents the capacity of substation j, represents the active power of the load of plot i at time t, R s Indicates the specified load capacity ratio.

[0028] Optionally, the objective function of the simulation operation model is to minimize the operation cost, and the objective function f is expressed as:

[0029] minf=C ope ;

[0030] The constraints of the simulation operation model include power flow constraints, node voltage constraints, branch current constraints and distributed photovoltaic output constraints;

[0031] The power flow constraints stipulate that each node and line must meet the following requirements:

[0032]

[0033] Among them, P j,t represents the active power of node j at time t, Q j,t represents the reactive power of node j at time t, o(j) and v(j) represent the set of nodes connected to node j with node j as the starting point and the end point, P ij,t and Q ij,t represents the active power and reactive power flowing through branch ij at time t, r ij and x ij represents the resistance and reactance of branch ij, They represent the active power and reactive power of the photovoltaic power of node j at time t, Represents the reactive power of the load of plot i at time t represents the reactive power of the load of plot i at time t, u j,t represents the square of the voltage of node j at time t, i ij,t represents the square of the current in branch ij at time t, ||·|| 2 represents the two-norm of ;

[0034] The node voltage constraints are:

[0035] U j,min ≤U j,t ≤U j,max ;

[0036] Among them, U j,t represents the voltage of node j at time t; U j,min and U j,max They represent the lower and upper limits of the voltage at node j respectively;

[0037] The branch current constraint is:

[0038] -I ij,max ≤I ij,t ≤I ij,max ;

[0039] Among them, I ij,t I represents the current magnitude of branch ij at time t; ij,max Indicates the upper limit of the current allowed to pass through branch ij;

[0040] The distributed photovoltaic output constraints are:

[0041]

[0042] in, represents the maximum power that node j can emit at time t, represents the distributed photovoltaic capacity connected to plot i, represents the per-unit value of photovoltaic output without considering the abandonment of solar power at time t, φ pv Indicates the power factor angle of photovoltaic.

[0043] Optionally, the S2 specifically includes:

[0044] S21, mapping the plots in the same supply area to an undirected graph, where the plots are represented as nodes in the graph, and the adjacent relationships between the plots are represented as edges in the graph. If there is an edge connecting two nodes, it means that the plots represented by the two nodes are adjacent;

[0045] S22, setting the root node of all nodes to the node itself;

[0046] S23, select the edge l that has not been traversed ij , where i and j represent the node numbers at both ends of the edge, query the root nodes fa(i) and fa(j) to which nodes i and j belong;

[0047] S24. If the root nodes of nodes i and j are different, set the root node of node fa(i) to fa(j);

[0048] S25, repeat steps S23 and S24 until all edges are traversed;

[0049] S26. Arrange the affiliation between nodes, count and output the number of root nodes. If all nodes belong to the same root node, that is, there is only one connected component, then the connectivity requirement is met.

[0050] Optionally, the S3 specifically includes:

[0051] S31, setting the initial two adjacent supply areas in the parent generation division scheme;

[0052] S32, searching for adjacent land blocks between adjacent supply areas;

[0053] S33, randomly select 70% of the adjacent plots as the set to be mutated;

[0054] S34, changing the supply area of ​​the land parcels one by one, and using the union-find algorithm to check the connectivity of the supply areas. If the supply area after the change does not meet the connectivity constraint, the ownership of the land parcel is changed back to the original supply area;

[0055] S35, traverse all adjacent supply areas in the parent generation division scheme one by one, and repeat steps S32-S34;

[0056] S36. A supply area division scheme that satisfies connectivity constraints.

[0057] Optionally, the S4 specifically includes:

[0058] S41. Each individual in the genetic algorithm represents a supply area division scheme, and the individual gene encoding is in

[0059] S42, generating a child population according to the improved child population generation process proposed in step S3, and merging the parent population with the child population;

[0060] S43, evaluate the fitness function of each individual in the population, including the distance index and operation cost of the zoning scheme;

[0061] S44, performing non-dominated sorting and crowding calculation on the individuals in the population according to the individual fitness function obtained in step S43, and updating the optimal solution as the new parent population;

[0062] S45. If the maximum number of iterations has not been reached, repeat steps S42-S44;

[0063] S46. Output the optimal supply area division plan.

[0064] The beneficial effects of the present invention are:

[0065] First, the present invention constructs a two-layer planning model that can dynamically evaluate and optimize the absorption capacity of distributed photovoltaics while dividing the supply area. By combining the supply area division with the simulation operation, the upper model and the lower model feed back each other, realizing the comprehensive optimization of the supply area division and the operation cost, thereby effectively reducing the phenomenon of abandoned light and improving the utilization rate of photovoltaic power generation. This feature is particularly important in the context of the gradual expansion of the scale of photovoltaic power generation, which can significantly improve the utilization efficiency of clean energy and reduce the impact on the environment.

[0066] In addition, the present invention uses a check-union algorithm to perform real-time inspection of the connectivity of the supply area, ensuring the connectivity between the blocks within the power supply area and avoiding local power supply interruptions or instability caused by improper supply area division. This method effectively ensures the integrity and power supply security of the power network through real-time connectivity inspection during the division process, so that the supply area division scheme is not only optimized in terms of economy, but also reaches high standards in terms of physical connectivity.

[0067] Furthermore, the present invention improves the genetic algorithm by incorporating the supply area connectivity constraint into the design of the genetic algorithm, so that the generated supply area division scheme can meet the connectivity requirements in practical applications while optimizing the target. This improvement improves the efficiency and practicality of the algorithm and ensures that the final optimization result has higher feasibility and reliability in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0069] Figure 1 A flow chart of a method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption proposed by the present invention;

[0070] Figure 2 A schematic diagram of a double-layer supply area division model of a substation supply area division method considering distributed photovoltaic consumption proposed by the present invention;

[0071] Figure 3 This is a flow chart for solving a double-layer supply area division model for a substation supply area division method considering distributed photovoltaic consumption proposed in the present invention. DETAILED DESCRIPTION

[0072] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0073] refer to Figure 1-3A method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption includes the following steps:

[0074] S1. Construct a two-level planning model with the goal of minimizing the partition distance and abandoned light;

[0075] S2, using the check-union algorithm to check the connectivity of the plots in the supply area;

[0076] S3. Establish an improved genetic algorithm based on graph theory to generate and optimize the supply area division scheme that meets the connectivity constraints;

[0077] S4. A two-level programming model solving algorithm based on an improved genetic algorithm is proposed, and the optimal supply area division scheme is output through iterative optimization.

[0078] In this embodiment, the two-layer planning model includes an upper layer and a lower layer. The upper layer is a supply division model, and the lower layer is a simulation operation model. The upper layer transmits the supply area division plan to the lower layer. After the lower layer receives the supply area division plan and calculates the operating cost, the calculated operating cost data is provided to the upper layer for optimizing the supply area division.

[0079] In this implementation, the objective function of the supply division model is to minimize the combined supply area division distance index and the operating cost:

[0080] minF=λ div D div +λ ope C ope ;

[0081]

[0082] Where F represents the objective function of the supply area division model, λ div Represents the weight coefficient of the distance indicator, λ ope Represents the weight coefficient of operating cost, D div It represents the distance index from the land to the substation after the supply area is divided, C ope represents the operating cost under the supply area division, NS represents the number of substations in the planning area, NP represents the number of plots in the planning area, and d ij represents the straight-line distance from plot i to substation j, represents the unit cost of abandoned light, c loss represents the unit cost of network loss, Indicates the total amount of abandoned photovoltaic power in the simulation operation, Indicates the total network loss power in the simulation operation, Indicates the total amount of abandoned photovoltaic power in the simulation operation, Indicates the total network loss in the simulation operation; a ijIndicates the supply area of ​​the plot, which is a 0 / 1 variable. If plot i belongs to the supply area of ​​substation j, then a ij =1, otherwise a ij =0;

[0083] The supply area division model constraints include single power supply constraints, connectivity constraints and substation capacity constraints;

[0084] The single power supply constraint stipulates that any plot can only be powered by one substation:

[0085]

[0086] The connectivity constraint stipulates that the power supply area of ​​the same substation should meet the connectivity:

[0087]

[0088] in, It represents the judgment function of the connected component of the supply area of ​​the strain station j. When the connected component = 1, the supply area meets the connectivity;

[0089] The substation capacity constraint states that the load supplied by the substation cannot exceed the rated capacity:

[0090]

[0091] in, represents the capacity of substation j, represents the active power of the load of plot i at time t, R s Indicates the specified load capacity ratio.

[0092] In this implementation, the objective function of the simulation operation model is to minimize the operation cost, and the objective function f is expressed as:

[0093] minf=C ope ;

[0094] The constraints of the simulation operation model include power flow constraints, node voltage constraints, branch current constraints and distributed photovoltaic output constraints;

[0095] The power flow constraints stipulate that each node and line must meet the following requirements:

[0096]

[0097] Among them, P j,t represents the active power of node j at time t, Q j,t represents the reactive power of node j at time t, o(j) and v(j) represent the set of nodes connected to node j with node j as the starting point and the end point, P ij,t and Qij,t represents the active power and reactive power flowing through branch ij at time t, r ij and x ij represents the resistance and reactance of branch ij, They represent the active power and reactive power of the photovoltaic power of node j at time t, Represents the reactive power of the load of plot i at time t represents the reactive power of the load of plot i at time t, u j,t represents the square of the voltage of node j at time t, i ij,t represents the square of the current in branch ij at time t, ||·|| 2 represents the two-norm of ;

[0098] The node voltage constraints are:

[0099] U j,min ≤U j,t ≤U j,max ;

[0100] Among them, U j,t represents the voltage of node j at time t; U j,min and U j,max They represent the lower and upper limits of the voltage at node j respectively;

[0101] The branch current constraint is:

[0102] -I ij,max ≤I ij,t ≤I ij,max ;

[0103] Among them, I ij,t I represents the current magnitude of branch ij at time t; ij,max Indicates the upper limit of the current allowed to pass through branch ij;

[0104] The distributed photovoltaic output constraints are:

[0105]

[0106] in, represents the maximum power that node j can emit at time t, represents the distributed photovoltaic capacity connected to plot i, represents the per-unit value of photovoltaic output without considering the abandonment of solar power at time t, φ pv Indicates the power factor angle of photovoltaic.

[0107] In this implementation, S2 specifically includes:

[0108] S21, mapping the plots in the same supply area to an undirected graph, where the plots are represented as nodes in the graph, and the adjacent relationships between the plots are represented as edges in the graph. If there is an edge connecting two nodes, it means that the plots represented by the two nodes are adjacent;

[0109] S22, setting the root node of all nodes to the node itself;

[0110] S23, select the edge l that has not been traversed ij , where i and j represent the node numbers at both ends of the edge, query the root nodes fa(i) and fa(j) to which nodes i and j belong;

[0111] S24. If the root nodes of nodes i and j are different, set the root node of node fa(i) to fa(j);

[0112] S25, repeat steps S23 and S24 until all edges are traversed;

[0113] S26. Arrange the affiliation between nodes, count and output the number of root nodes. If all nodes belong to the same root node, that is, there is only one connected component, then the connectivity requirement is met.

[0114] In this implementation, S3 specifically includes:

[0115] S31, setting the initial two adjacent supply areas in the parent generation division scheme;

[0116] S32, searching for adjacent land blocks between adjacent supply areas;

[0117] S33, randomly select 70% of the adjacent plots as the set to be mutated;

[0118] S34, changing the supply area of ​​the land parcels one by one, and using the union-find algorithm to check the connectivity of the supply areas. If the supply area after the change does not meet the connectivity constraint, the ownership of the land parcel is changed back to the original supply area;

[0119] S35, traverse all adjacent supply areas in the parent generation division scheme one by one, and repeat steps S32-S34;

[0120] S36. A supply area division scheme that satisfies connectivity constraints.

[0121] In this implementation, S4 specifically includes:

[0122] S41. Each individual in the genetic algorithm represents a supply area division scheme, and the individual gene encoding is in

[0123] S42, generating a child population according to the improved child population generation process proposed in step S3, and merging the parent population with the child population;

[0124] S43, evaluate the fitness function of each individual in the population, including the distance index and operation cost of the zoning scheme;

[0125] S44, performing non-dominated sorting and crowding calculation on the individuals in the population according to the individual fitness function obtained in step S43, and updating the optimal solution as the new parent population;

[0126] S45. If the maximum number of iterations has not been reached, repeat steps S42-S44;

[0127] S46. Output the optimal supply area division plan.

[0128] Embodiment 1:

[0129] In order to verify the feasibility of the present invention in implementation, the present invention is applied to a typical power system planning scenario, and we selected a medium-sized city in East China as the research object. The city is located in a plain with abundant light resources. In recent years, with the advancement of the national "Urban Rooftop Photovoltaic Action", distributed photovoltaic power generation has gradually become popular in various residential areas and industrial parks. However, with the continuous improvement of the penetration rate of distributed photovoltaics, the city's distribution network faces some challenges in daily operation. Specifically, due to the intermittent and uncertain nature of photovoltaic power generation, the power grid frequently experiences voltage over-limit and light abandonment problems during peak hours, especially during the noon period when the photovoltaic output is higher. The reverse flow phenomenon becomes more and more obvious, seriously affecting the stability of the power grid.

[0130] In this context, the substation supply area division method considering distributed photovoltaic consumption proposed in the present invention has been applied in the power system planning of the city. First, we conducted comprehensive data collection and modeling of the city's distribution network, focusing on the existing substation layout, load characteristics of each area, and the installation of distributed photovoltaics. Through the analysis of historical data, we found that photovoltaic power generation has the highest output during the noon period in spring and autumn, while in winter and summer, due to the existence of air-conditioning load, the grid load shows obvious peak and valley characteristics. These characteristics provide a realistic background for the application of the method of the present invention.

[0131] When applying the method of the present invention, we first constructed a two-layer planning model. The upper supply zone division model aims to minimize the supply zone division distance and operating cost, and the lower simulation operation model is used to calculate the operating cost under a given supply zone division scheme. Through the design of this two-layer model, we can dynamically evaluate the photovoltaic power generation capacity while considering the physical connectivity of the plots in the supply zone. During the model construction process, we used the check-union algorithm to check the connectivity of the plots in the supply zone to ensure that all plots can form a connected power network during the supply zone division process, avoiding the risk of power outages in the power supply area.

[0132] In the specific application, we first re-divided the existing supply area. The previous supply area division was mainly based on load density, but failed to fully consider the characteristics of photovoltaic power generation. Therefore, the re-divided supply area took more into account the distribution of photovoltaic power generation and the impact of reverse flow. In this process, we combined the improved genetic algorithm to generate a series of supply area division schemes that meet the connectivity constraints, and finally determined an optimal scheme through iterative optimization.

[0133] Under this optimal solution, the supply area division of all substations has been optimized, and the phenomenon of abandoned light has been significantly reduced. For example, in the re-divided supply area, an industrial park with a large photovoltaic installed capacity and a nearby residential area with relatively high load are merged into one supply area. In this way, during the peak photovoltaic output period at noon, the excess power generated by the industrial park can be directly supplied to the residential area through the power grid, avoiding the occurrence of reverse flow and abandoned light.

[0134] Table 1 Comparison of partitioning scheme indicators before and after optimization

[0135] Division scheme Distance indicator / km Annual abandoned photovoltaic power / MW·h Before optimization 285.92 9172.93 After optimization 303.83 5712.78

[0136] By comparing the data in Table 1, it can be seen that the supply area division schemes before and after optimization have undergone significant changes in distance indicators and annual abandoned photovoltaic power. First, from the perspective of distance indicators, the distance indicator of the optimized supply area division scheme has increased slightly, from 285.92 kilometers before optimization to 303.83 kilometers. This change shows that the optimized supply area division scheme has increased the average distance from each plot to the substation to a certain extent. This may be because when the supply area was redivided, in order to better balance the load and improve the absorption capacity of photovoltaic power generation, some plots were reassigned to more distant substations, resulting in an increase in distance indicators.

[0137] However, despite the increase in the distance index, the optimized scheme showed significant improvement in the annual abandoned photovoltaic power. Before the optimization, the annual abandoned photovoltaic power in the supply area was as high as 9172.93MWh, while after the optimization, this figure dropped significantly to 5712.78MWh, a decrease of about 3450MWh, a decrease of nearly 38%. This result shows that although the optimized supply area division scheme has increased in geographical distance, the photovoltaic absorption capacity has been significantly improved. This improvement directly illustrates the effectiveness of the optimization strategy, that is, through reasonable supply area division, it can not only balance the electricity demand and photovoltaic power supply of each region, but also effectively reduce the phenomenon of abandoned light, thereby improving the utilization rate of distributed photovoltaics.

[0138] This data comparison further proves the core advantage of the present invention, that is, when planning the supply area, it not only focuses on the traditional power supply distance index, but also pays more attention to the comprehensive optimization of photovoltaic power generation capacity and operating costs. This optimization scheme can effectively improve the overall efficiency and sustainability of the power supply system in practical applications, and provides a more scientific and reasonable planning method for the widespread application of distributed photovoltaics.

[0139] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption, characterized in that: The steps include: S1. Construct a two-level planning model with the goal of minimizing the partition distance and abandoned light; S2. Use the check-union algorithm to check the connectivity of the plots in the supply area; S3. Establish an improved genetic algorithm based on graph theory to generate and optimize the supply area division scheme that meets the connectivity constraints; S4. A two-level programming model solving algorithm based on an improved genetic algorithm is proposed, and the optimal supply area division scheme is output through iterative optimization.

2. A method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption according to claim 1, characterized in that: The two-layer planning model includes an upper layer and a lower layer, the upper layer is a supply division model, and the lower layer is a simulation operation model. The upper layer transmits the supply area division plan to the lower layer, and after the lower layer receives the supply area division plan and calculates the operation cost, it provides the calculated operation cost data to the upper layer for optimizing the supply area division.

3. A method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption according to claim 2, characterized in that: The objective function of the supply division model is to minimize the combined distance index of the supply division area and the operating cost: min F=λ div D div +λ ope C ope ; Where F represents the objective function of the supply area division model, λ div Represents the weight coefficient of the distance indicator, λ ope Represents the weight coefficient of operating cost, D div It represents the distance index from the land to the substation after the supply area is divided, C ope represents the operating cost under the supply area division, NS represents the number of substations in the planning area, NP represents the number of plots in the planning area, and d ij represents the straight-line distance from plot i to substation j, represents the unit cost of abandoned light, c loss represents the unit cost of network loss, Indicates the total amount of abandoned photovoltaic power in the simulation operation, Indicates the total network loss power in the simulation operation, Indicates the total amount of abandoned photovoltaic power in the simulation operation, Indicates the total network loss in the simulation operation; a ij Indicates the supply area of ​​the plot, which is a 0 / 1 variable. If plot i belongs to the supply area of ​​substation j, then a ij =1, otherwise a ij =0; The supply area division model constraints include single power supply constraints, connectivity constraints and substation capacity constraints; The single power supply constraint stipulates that any plot can only be powered by one substation: The connectivity constraint stipulates that the power supply area of ​​the same substation should meet the connectivity: in, It represents the judgment function of the connected component of the supply area of ​​the strain station j. When the connected component = 1, the supply area meets the connectivity; The substation capacity constraint states that the load supplied by the substation cannot exceed the rated capacity: in, represents the capacity of substation j, represents the active power of the load of plot i at time t, R s Indicates the specified load capacity ratio.

4. A method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption according to claim 2, characterized in that: The objective function of the simulation operation model is to minimize the operation cost. The objective function f is expressed as: my f=C ope ; The constraints of the simulation operation model include power flow constraints, node voltage constraints, branch current constraints and distributed photovoltaic output constraints; The power flow constraints stipulate that each node and line must meet the following requirements: Among them, P j,t represents the active power of node j at time t, Q j,t represents the reactive power of node j at time t, o(j) and v(j) represent the set of nodes connected to node j with node j as the starting point and the end point, P ij,t and Q ij,t represents the active power and reactive power flowing through branch ij at time t, r ij and x ij represents the resistance and reactance of branch ij, They represent the active power and reactive power of the photovoltaic power plant at node j at time t, Represents the reactive power of the load of plot i at time t represents the reactive power of the load of plot i at time t, u j,t represents the square of the voltage of node j at time t, i ij,t represents the square of the current of branch ij at time t, ||·||2 represents the second norm of ·; The node voltage constraints are: IN j,min ≤U j,t ≤U j,max ; Among them, U j,t represents the voltage of node j at time t; U j,min and U j,max They represent the lower and upper limits of the voltage at node j respectively; The branch current constraint is: -I ij,max ≤I ij,t ≤I ij,max ; Among them, I ij,t I represents the current magnitude of branch ij at time t; ij,max Indicates the upper limit of the current allowed to pass through branch ij; The distributed photovoltaic output constraints are: in, represents the maximum power that node j can emit at time t, represents the distributed photovoltaic capacity connected to plot i, represents the per-unit value of photovoltaic output without considering the abandonment of solar power at time t, φ pv Indicates the power factor angle of photovoltaic.

5. A method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption according to claim 1, characterized in that: The S2 specifically includes: S21, mapping the plots in the same supply area to an undirected graph, where the plots are represented as nodes in the graph, and the adjacent relationships between the plots are represented as edges in the graph. If there is an edge connecting two nodes, it means that the plots represented by the two nodes are adjacent; S22, setting the root node of all nodes to the node itself; S23, select the edge l that has not been traversed ij , where i and j represent the node numbers at both ends of the edge, query the root nodes fa(i) and fa(j) to which nodes i and j belong; S24. If the root nodes of nodes i and j are different, set the root node of node fa(i) to fa(j); S25, repeat steps S23 and S24 until all edges are traversed; S26. Arrange the affiliation between nodes, count and output the number of root nodes. If all nodes belong to the same root node, that is, there is only one connected component, then the connectivity requirement is met.

6. A method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption according to claim 1, characterized in that: The S3 specifically includes: S31, setting the initial two adjacent supply areas in the parent generation division scheme; S32, searching for adjacent land blocks between adjacent supply areas; S33, randomly select 70% of the adjacent plots as the set to be mutated; S34, changing the supply area ownership of the land parcels in turn, and using a union-find algorithm to check the connectivity of the supply areas. If the supply area after the change does not meet the connectivity constraint, the ownership of the land parcel is changed back to the original supply area; S35, traverse all adjacent supply areas in the parent generation division scheme one by one, and repeat steps S32-S34; S36. A supply area division scheme that satisfies connectivity constraints.

7. A method for dividing the supply area of ​​a substation considering distributed photovoltaic consumption according to claim 1, characterized in that: The S4 specifically includes: S41. Each individual in the genetic algorithm represents a donor area division scheme, and the individual gene encoding is in S42, generating a child population according to the improved child population generation process proposed in step S3, and merging the parent population with the child population; S43, evaluate the fitness function of each individual in the population, including the distance index and operation cost of the zoning scheme; S44, performing non-dominated sorting and crowding calculation on the individuals in the population according to the individual fitness function obtained in step S43, and updating the optimal solution as the new parent population; S45. If the maximum number of iterations has not been reached, repeat steps S42-S44; S46. Output the optimal supply area division plan.

Images