Power distribution network photovoltaic power supply power fair distribution method based on game model

Abstract

The invention relates to the technical field of power distribution networks, in particular to a power distribution network photovoltaic power supply power fair distribution method based on a game model. The method comprises the following steps: firstly, deploying a sensor in a power distribution network to collect the position information of photovoltaic power supplies, dividing sub-regions containing a certain number of photovoltaic power supplies through a K-Means algorithm, and counting the quantity of schedulable resources; secondly, participants, a strategy space and a revenue function are defined to construct a game model, and a Nash equilibrium solution method is used for obtaining a plurality of power distribution schemes through multiple iterations; and finally, performing NSGA-II algorithm initial population on the schemes, performing non-dominated sorting, crowding degree calculation, selection crossover mutation operation and population iteration updating, and selecting an optimal resource allocation scheme from Pareto optimal solutions according to the preference of a decision maker, thereby realizing fair distribution of the power of the photovoltaic power supply of the power distribution network, improving the stability of the power grid, the power generation efficiency and the cost effectiveness, and improving the economic benefit of the power distribution network. And reasonable utilization of resources is promoted.

Description

Technical Field

[0001] The invention relates to the technical field of power distribution networks, and in particular to a method for fair distribution of photovoltaic power sources in a power distribution network based on a game model. Background Art

[0002] As the global demand for clean energy continues to grow, photovoltaic power is increasingly used in distribution networks. As a renewable energy power generation method, photovoltaic power has significant advantages such as cleanliness, environmental protection, and sustainability, providing an effective way to alleviate the energy crisis and reduce environmental pollution.

[0003] However, with the gradual increase in the penetration rate of photovoltaic power sources in the distribution network, a series of new problems and challenges have also arisen. The traditional distribution network structure and operation management mode are mainly based on centralized power generation and unidirectional power supply design, and lack sufficient adaptability for the large-scale access of distributed photovoltaic power sources. In terms of power allocation, existing methods often find it difficult to effectively coordinate the power allocation relationship between many photovoltaic power sources. On the one hand, if the power cannot be reasonably allocated, it may lead to excessive utilization of photovoltaic power sources in some areas, while underutilization in other areas, and the potential of photovoltaic power sources in the entire distribution network cannot be fully utilized, reducing resource utilization efficiency. On the other hand, unreasonable power allocation may also cause power quality problems, such as voltage fluctuations, harmonic distortion, etc., affecting the normal operation of users' electrical equipment and reducing power supply reliability. In addition, in terms of fairness, there is a lack of effective mechanisms to ensure the fairness of power allocation among various photovoltaic power sources or sub-regions, which is easy to cause conflicts of interest and system instability.

[0004] Therefore, there is an urgent need for an innovative power allocation method for photovoltaic power sources in distribution networks, which can fully consider various factors, achieve fair and reasonable power allocation, and improve the overall performance and operating efficiency of the distribution network to meet the growing needs of photovoltaic power sources in distribution networks. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a fair distribution method for photovoltaic power sources in distribution networks based on a game model. By collecting photovoltaic power source location information to allocate power grid sub-areas and counting the amount of dispatchable resources, a game model is constructed based on this, and multiple schemes are generated by comprehensively considering fairness and efficiency. Then, the NSGA-II algorithm is used to combine factors such as grid stability, power generation efficiency, and cost-effectiveness to determine the best resource allocation scheme from multiple schemes, so as to achieve fair and reasonable distribution of photovoltaic power sources in distribution networks and improve the overall performance of the power grid.

[0006] The technical solution of the present invention is as follows:

[0007] A fair power allocation method for photovoltaic power sources in distribution networks based on a game model includes:

[0008] The distribution network monitoring system obtains the location information of the photovoltaic power source, divides the distribution network into multiple sub-areas according to the location information of the photovoltaic power source, each sub-area contains a certain number of photovoltaic power sources, and counts the dispatchable resources in each sub-area;

[0009] A power allocation game model for photovoltaic power sources in distribution networks is constructed according to the dispatchable resources in the sub-regions. Using the game model, multiple possible power allocation schemes are generated by considering the fairness and efficiency of photovoltaic power sources.

[0010] Based on the factors of grid stability, photovoltaic power generation efficiency and cost-effectiveness, the NSGA-II algorithm is used to determine the best resource allocation scheme from multiple power allocation schemes.

[0011] In a possible improvement, the location information includes geographic coordinate information of the photovoltaic power source, that is, longitude and latitude information, and electrical connection location information of the photovoltaic power source in the power distribution network.

[0012] In a possible improvement, the location information is divided into sub-areas by using a K-Means algorithm, and the steps include:

[0013] Randomly select K initial cluster centers, where K is the number of sub-regions that are expected to be divided;

[0014] For each photovoltaic power source data point, calculate its distance to the K cluster centers using the following formula:

[0015]

[0016] Among them, x i is the data point of the ith photovoltaic power source, x ik represents its kth attribute value, c j is the jth cluster center, c jk represents its kth attribute value, n is the number of attributes;

[0017] Assign each PV source to the cluster to which the nearest cluster center belongs;

[0018] Recalculate the center of each cluster, and the new cluster center coordinates are the average value of the corresponding attribute values ​​of all photovoltaic power data points in the cluster;

[0019] Repeat the above steps until the cluster center no longer changes significantly or the preset number of iterations is reached;

[0020] In a possible improvement, the dispatchable resource amount P sum , the calculation formula is:

[0021]

[0022] Pi is the actual output power of the i-th photovoltaic power source, and m is the number of photovoltaic power sources in the sub-area.

[0023] In a possible improvement, the game model is:

[0024] G={K,{S k},{U k}};

[0025] Among them, G is the game model function, K is the number of participants, that is, the number of sub-regions, S k is the strategy space of participant k, that is, the strategy space of sub-region k, U k is the payoff function of participant k;

[0026] U k =αη k -βF fair

[0027] Among them, α and β are weight coefficients;

[0028]

[0029] Among them, F fair is the resource allocation difference index, k represents the sub-region number, k = 1, 2, ..., K, K is the number of sub-regions, P sumk Each participant has its own schedulable resources, P i is the actual output power of the i-th photovoltaic power source;

[0030]

[0031] Among them, η k is the total power generation efficiency of the photovoltaic power source in the sub-region, P outk is the actual total output power of the photovoltaic power source in the kth sub-area.

[0032] In a possible improvement, the game model is solved by using Nash equilibrium, and the solution method includes:

[0033] Find a set of strategies So that for each participant k, while the strategies of other participants remain unchanged,

[0034]

[0035] in represents the optimal strategy combination of other participants except participant k;

[0036] In the calculation, the Nash equilibrium solution is approximated by the best response dynamic algorithm. Initially, a power allocation amount is randomly assigned to each sub-area. Then in each iteration, each sub-region selects the power allocation that maximizes its own benefit function as the strategy for the next iteration based on the current power allocation of other sub-regions, that is:

[0037]

[0038] Where t represents the number of iterations.

[0039] The iterative process is repeated until the convergence condition is reached, that is, the change in the power allocation amount of each sub-region between two adjacent iterations is less than the preset threshold or the maximum number of iterations is reached. At this time, the power allocation scheme obtained is a possible Nash equilibrium solution.

[0040] In a possible improvement, the step of determining the best resource allocation scheme from multiple power allocation schemes includes:

[0041] The generated multiple possible power allocation schemes are used as the initial population of the NSGA-II algorithm;

[0042] Perform non-dominated sorting and crowding calculation on the initial population;

[0043] The offspring individuals are generated by the initial population through selection, crossover and mutation operations;

[0044] The offspring individuals generated by selection, crossover and mutation operations are merged with the parent individuals to form a new population and iteratively update it;

[0045] After the iteration, the individuals in the non-dominated layer of the new generation population are the Pareto optimal solutions.

[0046] In a possible improvement, the non-dominated sorting is performed according to a stability index of the power grid, power generation efficiency and cost-effectiveness of the photovoltaic power source;

[0047] The congestion calculation formula is:

[0048]

[0049] Among them, d i For individual y i The congestion degree, M is the number of objective functions;

[0050]

[0051] in, and They are the objective functions f m The maximum and minimum values ​​in the population, f m is the mth objective function, y ik For the corresponding individual, 2≤k≤N-1.

[0052] The beneficial effects brought about by the technical solution provided in the embodiments of the present application include at least the following beneficial effects:

[0053] Through accurate sub-region division and detailed resource statistics, the actual situation of photovoltaic power sources in each sub-region can be fully considered, and the constructed game model effectively balances fairness and efficiency to ensure more reasonable power distribution. It avoids excessive concentration or idleness of resources, improves the utilization rate of photovoltaic power sources in the entire distribution network, achieves relative fairness in power distribution among sub-regions, and reduces potential conflicts caused by uneven distribution.

[0054] Introducing power quality factors and power factors in the process of generating allocation plans makes power allocation more in line with grid operation requirements, which helps reduce power quality issues such as voltage fluctuations and harmonic distortion. The final plan is determined by comprehensively considering multiple factors such as grid stability through the NSGA-II algorithm, which can effectively maintain the stable operation of the grid, enhance the grid's adaptability to photovoltaic power access, improve power supply reliability, and ensure the normal operation of users' electrical equipment.

[0055] The power generation efficiency and cost-effectiveness of photovoltaic power sources are taken into consideration. While improving power generation efficiency, the investment cost, operation and maintenance cost, and economic losses caused by power quality problems are taken into account. This comprehensive consideration helps to reduce overall costs, improve the economic efficiency of distribution network operation, and maximize economic benefits while achieving efficient use of clean energy, thus promoting the sustainable development of distribution networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the overall process of the fair allocation method of photovoltaic power in distribution network based on game model;

[0057] Figure 2 It is a detailed flow chart of step S100 of the method for fair allocation of photovoltaic power in distribution network based on game model;

[0058] Figure 3 It is a detailed flow chart of step S200 of the method for fair power allocation of photovoltaic power sources in distribution network based on game model;

[0059] Figure 4 The figure is a detailed flow chart of step S300 of a method for fair allocation of photovoltaic power in a distribution network based on a game model. DETAILED DESCRIPTION

[0060] 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.

[0061] With the widespread application of photovoltaic power sources in distribution networks, how to reasonably allocate photovoltaic power has become an important issue. Traditional allocation methods often fail to balance fairness, efficiency, and various operating factors of the power grid, leading to problems such as insufficient resource utilization and unstable power quality. Therefore, a more scientific and effective power allocation method is needed to improve the operating performance of the distribution network.

[0062] To solve the above problems, please refer to Figure 1 , which shows a method for fair distribution of photovoltaic power in a distribution network based on a game model provided by an embodiment of the present invention, the method comprising:

[0063] S100: The distribution network monitoring system obtains location information of the photovoltaic power source, divides the distribution network into multiple sub-areas according to the location information of the photovoltaic power source, each sub-area contains a certain number of photovoltaic power sources, and counts the dispatchable resources of each sub-area;

[0064] S200: construct a power allocation game model for photovoltaic power sources in a distribution network according to the dispatchable resource amount of the sub-region; and generate multiple possible power allocation schemes by using the game model and considering the fairness and efficiency of the photovoltaic power sources.

[0065] S300: Based on the stability of the power grid, the power generation efficiency of the photovoltaic power source and the cost-effectiveness factors, the NSGA-II algorithm is used to determine the best resource allocation scheme from multiple power allocation schemes.

[0066] Through the above detailed technical solutions, it is possible to effectively achieve fair distribution of photovoltaic power in the distribution network, fully consider various factors, and improve the overall performance and operating efficiency of the distribution network.

[0067] S100 mainly completes the work of sub-area division and resource quantity statistics of distribution network. By collecting the location information of photovoltaic power sources, the K-Means algorithm is used to divide the distribution network sub-areas and count the accurate dispatchable resources in each sub-area, providing basic data for subsequent power allocation.

[0068] Please refer to Figure 2 , which shows a flowchart of an exemplary method S100 for fair distribution of photovoltaic power in a distribution network based on a game model of the present application, and its contents include:

[0069] S110: Collecting photovoltaic power source location information.

[0070] High-precision location sensors are deployed in the distribution network. The sensors correspond to photovoltaic power sources one by one to obtain the geographic coordinate information of the photovoltaic power source, that is, the longitude and latitude information; as well as the electrical connection location information in the distribution network. The sensors transmit the collected location information to the data acquisition server in the distribution network monitoring system through the wireless communication network.

[0071] S120: Divide the sub-regions using the location information.

[0072] In a possible implementation, sub-region division is performed based on a K-Means algorithm.

[0073] Each photovoltaic power source is regarded as a data point, and its location information is used as the attribute of the data point. The K-Means algorithm is used for sub-area division.

[0074] The calculation steps are as follows:

[0075] (1) Randomly select K initial cluster centers, where K is the number of sub-regions to be divided. The initial cluster centers are randomly selected from the location information dataset of photovoltaic power sources.

[0076] (2) For each photovoltaic power source data point, calculate its distance to the K cluster centers. The calculation formula is:

[0077]

[0078] Among them, x i is the data point of the ith photovoltaic power source, x ik represents its kth attribute value, c j is the jth cluster center, c jk represents its kth attribute value, and n is the number of attributes.

[0079] (3) Allocate each photovoltaic power source to the cluster to which the nearest cluster center belongs.

[0080] (4) Recalculate the center of each cluster. The new cluster center coordinates are the average of the corresponding attribute values ​​of all photovoltaic power data points in the cluster. For geographic coordinate information clustering, the new cluster center longitude is: latitude: Among them, c jnewlon and c jnewlat is the longitude and latitude of the new cluster center, x ilon and x ilat is the attribute value corresponding to the PV power source data point, i.e., longitude and latitude, and m is the number of PV sources in cluster j.

[0081] For the electrical connection location feature clustering, the average values ​​of the corresponding features are similarly calculated.

[0082] (5) Repeat steps (2) to (4) until the cluster center no longer changes significantly or the preset number of iterations is reached.

[0083] Through the above clustering process, the distribution network is divided into K non-overlapping sub-areas, each of which contains several photovoltaic power sources.

[0084] S130: Statistics on the amount of schedulable resources in the sub-region.

[0085] For each sub-region, the sum of the rated powers of the photovoltaic power sources it contains is counted as the initial dispatchable resource quantity estimate P sum0 .

[0086] Consider the real-time operating status of the photovoltaic power source, such as the influence of light intensity, temperature and other factors on its actual output power. According to the power output characteristic model of the photovoltaic power source:

[0087]

[0088] Where, P is the actual output power, P STC is the rated power under standard test conditions, G is the actual light intensity, G STC It is the light intensity under standard test conditions, usually 1000W / m 2 , λ is the power temperature coefficient, T is the actual temperature of the photovoltaic cell, T STC The temperature under standard test conditions is usually 25°C.

[0089] Real-time collection of the light intensity G at each photovoltaic power source location i and temperature T i Data, calculate the actual output power P of each photovoltaic power source i , sum up the actual output power of all photovoltaic power sources in the sub-area to obtain the accurate dispatchable resource amount P sum , the calculation formula is:

[0090]

[0091] S200 focuses on building a game model and solving it to generate a power allocation plan. It defines the game model elements based on the dispatchable resources in the sub-region, builds and solves the power allocation game model of photovoltaic power sources in the distribution network, and obtains multiple power allocation plans that consider fairness and efficiency.

[0092] Please refer to Figure 3 , which shows a flowchart of an exemplary method S200 for fair distribution of photovoltaic power in a distribution network based on a game model of the present application, and its contents include:

[0093] S210: Definition of game model elements.

[0094] (1) Participants: Each sub-region is regarded as a participant, and each participant has its own schedulable resource amount P. sumk , k represents the sub-area number, k = 1, 2, ..., K. And each participant has its own strategy space. The strategy space refers to the range of power allocation that each sub-area can choose when allocating power, which is set to [0, P sumk ].

[0095] (2) Profit function: Define the profit function of each participant, which takes into account fairness and efficiency factors.

[0096] For fairness, the resource allocation difference index is used to measure it, and the calculation formula is:

[0097]

[0098] Among them, F fair It is an indicator of resource allocation diversity.

[0099] For efficiency, the total power generation efficiency η of the PV power source in the sub-region is used. k The power generation efficiency is measured by the actual output power of the photovoltaic power source and the amount of dispatchable resources. Where P outk is the actual total output power of the photovoltaic power source in the kth sub-area.

[0100] The participant's profit function is calculated as follows:

[0101] U k =αη k -βF fair ;

[0102] Where α and β are weight coefficients used to adjust the relative importance of fairness and efficiency in the benefit function.

[0103] S220: Game model construction and solution.

[0104] According to the defined participants, strategy space and benefit function, a game model for photovoltaic power allocation in distribution network is constructed. The model is expressed as:

[0105] G={K,{S k},{U k}};

[0106] Where K is the number of participants, i.e. the number of sub-regions, S k is the strategy space of participant k, that is, the strategy space of sub-region k, U k is the payoff function of participant k.

[0107] A game theory solving method is selected to solve the model and generate multiple possible power allocation schemes.

[0108] In one possible implementation, a Nash equilibrium solution method is used, that is, a set of strategies is found So that for each participant k, while the strategies of other participants remain unchanged,

[0109]

[0110] in Represents the optimal strategy combination of other participants except participant k.

[0111] In the calculation, the Nash equilibrium solution is approximated by the best response dynamic algorithm. Initially, a power allocation amount is randomly assigned to each sub-area. Then in each iteration, each sub-region selects the power allocation that maximizes its own benefit function as the strategy for the next iteration based on the current power allocation of other sub-regions, that is:

[0112]

[0113] Where t represents the number of iterations.

[0114] Repeat the iteration process until the convergence condition is reached, that is, the change in the power allocation amount of each sub-area between two adjacent iterations is less than the preset threshold or the maximum number of iterations is reached. The power allocation scheme obtained at this time is a possible Nash equilibrium solution. By changing the initial strategy multiple times, multiple different Nash equilibrium solutions, that is, multiple possible power allocation schemes, are obtained.

[0115] S300 aims to determine the best resource allocation scheme from multiple power allocation schemes using the NSGA-II algorithm. By using the generated power allocation scheme as the initial population, the NSGA-II algorithm performs non-dominated sorting, congestion calculation, crossover mutation operation and iterative update to determine the best resource allocation scheme from the Pareto optimal solution according to the decision maker's preference.

[0116] Please refer to Figure 3 , which shows a flowchart of an exemplary method S300 for fair distribution of photovoltaic power in a distribution network based on a game model of the present application, and its contents include:

[0117] S310: Using the generated multiple possible power allocation schemes as the initial population of the NSGA-II algorithm.

[0118] The multiple possible power allocation schemes generated in S200 are used as the initial population P of the NSGA-II algorithm. 0 Each power allocation scheme is represented as an individual y i =(Pi1 ,P i2 ,...,P iK ), where P iK is the power allocation of the kth sub-area in the ith scheme. At the same time, determine the population size N, that is, the number of power allocation schemes, the maximum number of iterations T max , crossover probability p c and mutation probability p m .

[0119] S320: Non-dominated sorting and congestion calculation.

[0120] (1) Non-dominated sorting: For individual P in the population 0 , non-dominated sorting is performed according to the pre-set objective function. The objective function includes the stability index of the power grid, the power generation efficiency and cost-effectiveness of the photovoltaic power source, etc.

[0121] For example, the grid stability can be measured by the indicators of voltage stability margin and frequency stability margin; the power generation efficiency is as mentioned above. Cost-effectiveness takes into account the investment cost of photovoltaic power sources, operation and maintenance costs, and economic losses caused by power quality problems due to unreasonable power distribution. The comprehensive cost-effectiveness index is the sum of the above costs.

[0122] For two individuals y i and j , if y i Not inferior to y in all objective functions j , and is better than y in at least one objective function j , then y i Dominate y j By comparing the dominance relationship between all individuals in the population, the population is divided into multiple non-dominated layers F 1 ,F 2 ,..., where F 1 The layer contains all non-dominated individuals, that is, individuals that are not dominated by other individuals. 2 The layer contains F 1 The individuals dominated by individuals in the same layer belong to the non-dominated individuals, and so on.

[0123] (2) Crowding calculation: For each individual in the non-dominated layer, calculate its crowding degree.

[0124] Crowding reflects the density of individuals around an individual. The greater the crowding, the sparser the individuals around the individual, and the better the diversity in the population.

[0125] For example, individual y i For example, the sum of the distances of adjacent individuals on each objective function is calculated as the crowding degree di , for the mth objective function f m , first sort the individuals on the objective function, let f m (y i1 )≤f m (y i2 )≤...≤f m (y iN ), then individual y ik (2≤k≤N-1) in the objective function f m The congestion contribution on is:

[0126]

[0127] in and They are the objective functions f m The maximum and minimum values ​​in a population.

[0128] Individual i The congestion degree d i The calculation formula is:

[0129]

[0130] Where M is the number of objective functions.

[0131] S330: Selection, crossover and mutation operations.

[0132] The selection operation randomly selects a certain number of individuals through a tournament, and selects the best one as the parent according to the non-dominated level and crowding degree to retain the excellent characteristics; the crossover operation uses the simulated binary crossover method to select the genes of the parent individuals with a certain probability, that is, the power allocation. The crossover factor is calculated to exchange part of the information and generate new offspring to increase diversity; the mutation operation performs polynomial mutation on the genes of the offspring individuals with the mutation probability, changes the gene value according to the random number and mutation distribution index, further enhances the diversity of the population, and jointly promotes the evolution of the population in a better direction to find the best resource allocation plan.

[0133] S340: New population generation and iterative update.

[0134] The offspring individuals generated by selection, crossover and mutation operations are merged with the parent individuals to form a new population P. new Then for P new Perform non-dominated sorting and crowding calculation, divide the population into different non-dominated layers in the same way as sub-step S320, and calculate the crowding of each individual.

[0135] According to the non-dominated sorting and crowding information, the new population P new Select a new generation of population Pnewt ,The selection process first prioritizes individuals with low non-dominated ranks. When the number of individuals with the same non-dominated rank exceeds the population size, individuals with high crowding degree are selected until the number of individuals selected reaches the population size N.

[0136] Repeat S330 and S340 until the preset maximum number of iterations T is reached. max .

[0137] S350: Determine the optimal resource allocation solution.

[0138] After the iteration, the new generation population P newt The non-dominated layer F 1 The individual in is the Pareto optimal solution.

[0139] From the Pareto optimal solution, the best resource allocation scheme is selected according to actual needs and decision makers' preferences. For example, if decision makers pay more attention to grid stability, they can select the scheme with the best performance in terms of stability indicators; if they emphasize cost-effectiveness, they can select the scheme with the best cost-effectiveness indicators; if they want to strike a balance between multiple goals, they can use multi-attribute decision-making methods to comprehensively evaluate the Pareto optimal solution and select the scheme with the best comprehensive performance as the final optimal resource allocation scheme for photovoltaic power generation in the distribution network.

[0140] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0141] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0142] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0143] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features of the present invention.

[0144] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A fair power allocation method for photovoltaic power sources in a distribution network based on a game model, characterized in that: include: The distribution network monitoring system obtains the location information of the photovoltaic power source, divides the distribution network into multiple sub-areas according to the location information of the photovoltaic power source, each sub-area contains a certain number of photovoltaic power sources, and counts the dispatchable resources in each sub-area; A power allocation game model for photovoltaic power sources in the distribution network is constructed based on the dispatchable resources in the sub-regions. Multiple possible power allocation schemes are generated by considering the fairness and efficiency of photovoltaic power sources using the game model. Based on the factors of grid stability, photovoltaic power generation efficiency and cost-effectiveness, the NSGA-II algorithm is used to determine the best resource allocation scheme from multiple power allocation schemes.

2. The method for fair distribution of photovoltaic power in a distribution network based on a game model according to claim 1, characterized in that: The location information includes the geographic coordinate information of the photovoltaic power source, that is, the longitude and latitude information, and the electrical connection location information of the photovoltaic power source in the power distribution network.

3. The method for fair distribution of photovoltaic power in a distribution network based on a game model according to claim 2, characterized in that: The location information is divided into sub-areas by using the K-Means algorithm, and the steps include: Randomly select K initial cluster centers, where K is the number of sub-regions that are expected to be divided; For each photovoltaic power source data point, calculate its distance to the K cluster centers using the following formula: Where xi is the data point of the i-th photovoltaic power source, xik represents its k-th attribute value, cj is the j-th cluster center, cjk represents its k-th attribute value, and n is the number of attributes; Assign each PV source to the cluster to which the nearest cluster center belongs; Recalculate the center of each cluster, and the new cluster center coordinates are the average value of the corresponding attribute values ​​of all photovoltaic power data points in the cluster; Repeat the above steps until the cluster center no longer changes significantly or the preset number of iterations is reached.

4. The method for fair distribution of photovoltaic power in a distribution network based on a game model according to claim 1, characterized in that: The schedulable resource amount Psum is calculated as follows: Pi is the actual output power of the i-th photovoltaic power source, and m is the number of photovoltaic power sources in the sub-area.

5. The method for fair distribution of photovoltaic power in a distribution network based on a game model according to claim 1, characterized in that: The game model is: G = {K, {Sk}, {Uk}}; Among them, G is the game model function, K is the number of participants, that is, the number of sub-areas, Sk is the strategy space of participant k, that is, the strategy space of sub-area k, U k is the payoff function of participant k; Uk=αηk-βFfair; Among them, α and β are weight coefficients; Among them, Ffair is the resource allocation difference index, k represents the sub-area number, k = 1, 2, ..., K, K is the number of sub-areas, Psumk is the amount of dispatchable resources for each participant, and Pi is the actual output power of the i-th photovoltaic power source; Among them, ηk is the total power generation efficiency of the photovoltaic power source in the sub-region, and Poutk is the actual total output power of the photovoltaic power source in the kth sub-region.

6. The method for fair distribution of photovoltaic power in a distribution network based on a game model according to claim 5, characterized in that: The game model is solved by using Nash equilibrium, and the solution method includes: Find a set of strategies So that for each participant k, while the strategies of other participants remain unchanged, in represents the optimal strategy combination of other participants except participant k; In the calculation, the Nash equilibrium solution is approximated by the best response dynamic algorithm. Initially, a power allocation amount is randomly assigned to each sub-area. Then in each iteration, each sub-region selects the power allocation that maximizes its own benefit function as the strategy for the next iteration based on the current power allocation of other sub-regions, that is: Where t represents the number of iterations; The iterative process is repeated until the convergence condition is reached, that is, the change in the power allocation amount of each sub-region between two adjacent iterations is less than the preset threshold or the maximum number of iterations is reached. At this time, the power allocation scheme obtained is a possible Nash equilibrium solution.

7. The method for fair distribution of photovoltaic power in a distribution network based on a game model according to claim 1, characterized in that: The step of determining the best resource allocation scheme from a plurality of power allocation schemes comprises: The generated multiple possible power allocation schemes are used as the initial population of the NSGA-II algorithm; Perform non-dominated sorting and crowding calculation on the initial population; The offspring individuals are generated by the initial population through selection, crossover and mutation operations; The offspring individuals generated by selection, crossover and mutation operations are merged with the parent individuals to form a new population and iteratively update it; After the iteration, the individuals in the non-dominated layer of the new generation population are the Pareto optimal solutions.

8. The method for fair distribution of photovoltaic power in a distribution network based on a game model according to claim 7, characterized in that: The non-dominated sorting is performed according to the stability index of the power grid, the power generation efficiency of the photovoltaic power source and the cost-effectiveness; The congestion calculation formula is: Among them, di is the crowding degree of individual yi, and M is the number of objective functions; in, and are the maximum and minimum values ​​of the objective function fm in the population, fm is the mth objective function, yik is the corresponding individual, 2≤k≤N-1.

Images