Power distribution network three-phase load dynamic balance adjusting system and method

Abstract

The invention relates to the technical field of power systems, and discloses a power distribution network three-phase load dynamic balance adjusting system and method, and the system comprises a region division module which collects the data of a power distribution network topological structure and load distribution through a power grid system, divides a power distribution network into a plurality of independent regions according to the collected data, and transmits the regions to the power grid system; distributing a unique identifier for each region and setting an independent intelligent agent, wherein the intelligent agent configures an adaptive region load scale and a region complexity degree; according to the method, the power distribution network is divided into the independent areas, and the intelligent agent adaptive to the load scale and complexity is configured for each area, so that accurate acquisition of regional load characteristics is facilitated, and the intelligent agent can accurately acquire the load characteristics based on the three-phase voltage, current and active power data acquired in real time. An initial strategy population containing multiple adjustment operation combinations is generated by applying a coevolution algorithm, and an optimal strategy is screened out through simulation evaluation, so that strategy generation is more targeted.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a three-phase load dynamic balancing regulation system and method for a distribution network. Background Art

[0002] In modern power systems, the distribution network is a key link connecting the transmission network and end users. Its stability and efficiency directly affect the operating quality of the entire power system. With the rapid development of industrialization and urbanization, electricity demand is increasing, and load types are becoming increasingly diverse, including a large number of single-phase loads and distributed power sources. This has led to an increasingly prominent problem of three-phase load imbalance in the distribution network. Three-phase load imbalance not only increases line losses and reduces energy efficiency, but also causes voltage fluctuations and frequency offsets, affecting power supply quality and system stability.

[0003] However, traditional technologies have many shortcomings in the dynamic balance regulation of three-phase loads in distribution networks. When regulating the entire distribution network, traditional methods ignore the load differences and dynamic change characteristics between different regions, making the regulation strategy lack of pertinence and flexibility, resulting in load imbalance problems between local areas. Traditional methods also have limitations in information acquisition and processing, making it difficult to grasp the operating status and load changes of the distribution network in real time and accurately, affecting the timeliness and effectiveness of the regulation strategy. In addition, traditional methods lack a global optimization mechanism and cannot achieve strategy coordination and complementarity among multiple regions, limiting the further improvement of the overall regulation effect.

[0004] Therefore, a three-phase load dynamic balancing regulation system and method for a distribution network are developed. Summary of the Invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a three-phase load dynamic balancing regulation system and method for a distribution network. The present invention divides the distribution network into independent areas and configures an intelligent agent that adapts to the load scale and complexity for each area, so as to facilitate the accurate collection of regional load characteristics. The intelligent agent uses a collaborative evolutionary algorithm based on the real-time collected three-phase voltage, current and active power data to generate an initial strategy population containing a variety of regulation operation combinations, and selects the optimal strategy through simulation evaluation, making the strategy generation more targeted. Moreover, through the regional and hierarchical optimization mechanism, the accuracy and adaptability of the regulation strategy are significantly improved, avoiding the local load imbalance problem caused by traditional methods.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a three-phase load dynamic balancing and regulating system for a distribution network, the system comprising:

[0007] Regional division module: The power grid system collects data on the distribution network topology and load distribution, divides the distribution network into multiple independent regions based on the collected data, assigns a unique identifier to each region and sets up an independent intelligent agent. The intelligent agent configuration adapts to the regional load scale and regional complexity;

[0008] Local optimization module: This module uses an intelligent agent to obtain real-time data on three-phase voltage, current, and active power, and uses a co-evolutionary algorithm to generate and store a preliminary load regulation strategy for the region.

[0009] Information sharing module: Each regional intelligent agent exchanges information through the communication network, sends the generated local load regulation strategy to the intelligent agents in adjacent regions, receives the strategies of adjacent regions at the same time, calculates the adaptability of load characteristics between regions, and stores the received strategies;

[0010] Collaborative Optimization Module: This module optimizes the strategy with the highest adaptability based on the adaptability of the adjacent regional strategies and its own real-time load information. It then calculates the adaptability of the optimized strategy and compares it with the adaptability of the original strategy. Based on the comparison results, it decides whether to update the local strategy.

[0011] Global optimization module: Regularly collects new strategies and the adaptability of new strategies after optimization in each region, sets weights based on regional load scale and regional complexity, conducts global optimization evaluation, and issues promotion instructions to relevant regions for strategies with good global adaptability.

[0012] Furthermore, in the local optimization module, using the co-evolutionary algorithm to generate a preliminary load regulation strategy for the region includes the following steps:

[0013] The intelligent agent cleans the acquired real-time data of three-phase voltage, current, and active power, removes outliers and noise data, extracts key characteristic parameters, and generates a standardized load data set;

[0014] Based on the preprocessed load data set, combined with historical regulation experience and regional load characteristics, an initial strategy population containing multiple regulation operation combinations is randomly generated. The regulation operation combinations include adjusting user wiring methods and distributed generation output;

[0015] Apply each strategy in the initial strategy population to the current load scenario for simulation, evaluate its effectiveness in improving three-phase imbalance and reducing line losses, and generate strategy fitness and strategy evaluation reports;

[0016] Based on the strategy fitness and strategy evaluation report, the strategies with better performance are selected as the parent generation, and the strategies with poor performance are eliminated to form an elite strategy subset;

[0017] Performing crossover and mutation operations on the elite strategy subset to generate child strategies that contain the advantageous features of the parent strategy, wherein the crossover and mutation operations include wiring adjustment and power control schemes that integrate different strategies;

[0018] Merge the offspring strategy with the parent strategy, evaluate and screen again, and repeat the evolution process to gradually improve the adaptability and effectiveness of the strategy;

[0019] When the number of iterations reaches the preset upper limit or the strategy effect converges, the strategy with the best performance is selected as the preliminary load regulation strategy for this area and formatted and stored.

[0020] Furthermore, in the information sharing module, the inter-region load characteristic adaptability is calculated using the following formula: ,in, For the The strategy for each region is The load characteristic adaptability of each region, For the The strategic fitness of each region, is the similarity of load characteristics between the two regions obtained by calculating the correlation coefficient of the historical load curves of the i-th region and the j-th region.

[0021] Furthermore, in the collaborative optimization module, the screened strategies are optimized based on the adaptability of the adjacent region strategies and the real-time load information of the current three-phase voltage, current and active power data in the region. The optimization process includes adjusting the adjustment parameters in the strategy and modifying the adjustment operation steps based on the actual situation in the region.

[0022] Furthermore, in the collaborative optimization module, the optimized strategy fitness is calculated using the following formula: ,in, For the The new strategy fitness after region optimization, is the retention coefficient, and its value range is , For the The adaptability of the original strategy in each region, For the The strategy for each region is The load characteristic adaptability of each area.

[0023] Furthermore, in the collaborative optimization module, the optimized strategy fitness is simulated and compared with the original strategy fitness, and when deciding whether to update the local strategy based on the comparison result, the calculation formula is used: ,in, is the strategy fitness difference evaluation value, For the The new strategy fitness after region optimization, For the The adaptability of the original strategy in each region, is the difference coefficient, with a value range of 0.05-0.1, which is used to set the minimum difference ratio for determining whether the new strategy is better. The value of determines whether to update the local policy;

[0024] when When the fitness of the optimized new strategy exceeds the fitness of the original strategy, the updated local strategy is determined to be the strategy after collaborative optimization. When , the fitness of the optimized new strategy does not exceed the fitness of the original strategy, and the original strategy is retained.

[0025] Furthermore, in the global optimization module, the new strategies and the fitness of the new strategies after optimization in each region are collected regularly. , set the weight of each region according to the regional load scale and regional complexity , calculate the global optimization evaluation value , and its calculation formula is: ,in, To optimize the evaluation value globally; The total number of regions allocated for the distribution network; For the The weight of the region, For the The fitness of the new strategy after region optimization.

[0026] Furthermore, in the global optimization module, the strategy with good global adaptability is: Ranked in the top 20% of all regional strategies.

[0027] On the other hand, a method for dynamic balancing and regulating three-phase loads in a distribution network is provided, wherein the specific steps of the method are as follows:

[0028] Regional division: Based on the topological structure and load distribution data of the distribution network, the distribution network is divided into multiple independent regions, and an independent intelligent agent is set up for each region;

[0029] Local optimization: Each regional intelligent agent collects the three-phase load information of the region in real time, generates and stores the preliminary load regulation strategy for the region through the collaborative evolution algorithm;

[0030] Information sharing: Each regional intelligent agent sends its own load regulation strategy to the intelligent agents in adjacent regions through the communication network, and simultaneously receives the load regulation strategies sent by adjacent regions, calculates the inter-region load characteristic adaptability, and stores it;

[0031] Collaborative optimization: By combining the adaptability of adjacent regional strategies with its own load information, the received load regulation strategy is optimized to generate a more suitable load regulation strategy for the region. The optimized strategy fitness is calculated. If the fitness of the optimized new strategy exceeds the fitness of the original strategy, the strategy for the region is updated; otherwise, the original strategy is retained.

[0032] Global optimization: Collect the optimized new strategies and the adaptability of the new strategies in each region, conduct a global evaluation, and issue promotion instructions to relevant regions for strategies with good global adaptability.

[0033] Compared with the existing technology, the three-phase load dynamic balancing regulation system and method of the distribution network have the following beneficial effects:

[0034] 1. The present invention divides the distribution network into independent areas and configures an intelligent agent adapted to the load scale and complexity for each area, facilitating the precise collection of regional load characteristics. Based on the real-time collected three-phase voltage, current and active power data, the intelligent agent uses a co-evolutionary algorithm to generate an initial strategy population containing multiple adjustment operation combinations, and selects the optimal strategy through simulation evaluation, making strategy generation more targeted. Moreover, through the regional and hierarchical optimization mechanism, the accuracy and adaptability of the adjustment strategy are significantly improved, avoiding the local load imbalance problem caused by traditional methods.

[0035] 2. The present invention constructs a cross-regional strategy collaboration framework through the information sharing module and the collaborative optimization module. The intelligent agent in each region sends the local strategy to the adjacent region, calculates the load characteristic adaptability, and selects strategies with high adaptability for parameter adjustment and operation step optimization. When the strategy of one region has a high adaptability in another region due to similar load characteristics, the strategy parameters will be adjusted in combination with the real-time data of the other region to generate an optimization strategy that better meets the needs of the other region. In addition, the global optimization module regularly collects the optimized strategies and adaptability of each region, sets weights according to the regional load scale and complexity, calculates the global optimization evaluation value, and promotes the top 20% of the global adaptability strategies, ensuring the cross-regional reuse of high-quality strategies and avoiding the limitations of local optimization.

[0036] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0038] Figure 1 This is a framework diagram of a three-phase load dynamic balancing regulation system for a distribution network.

[0039] Figure 2 The figure is a flow chart of a method for dynamic balancing adjustment of three-phase loads in a distribution network. DETAILED DESCRIPTION

[0040] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0041] Example 1:

[0042] Regional division module: The power grid system collects the topology of the distribution network in urban residential areas (including transformer locations and line connection relationships) and load distribution data (peak and off-peak periods of electricity load in each building), divides the residential area into three independent areas (Area A is a high-rise residential complex, Area B is a multi-story residential complex, and Area C is a commercial supporting area). Each area is assigned a unique identifier (such as A-01, B-02, and C-03) and configured with an independent intelligent agent. Area A is equipped with a high-performance intelligent agent due to its large load scale and complex electrical equipment, while Areas B and C are equipped with conventional intelligent agents.

[0043] Local optimization module: The intelligent agent in Area A collects three-phase voltage, current, and active power data in real time. After eliminating abnormal fluctuations caused by meter failures, it extracts key features such as voltage deviation rate and current imbalance to generate a standardized load data set. Combined with historical regulation experience, the collaborative evolution algorithm randomly generates an initial strategy population containing operation combinations such as "adjusting the wiring phase sequence of three residential buildings" and "reducing the output of the community photovoltaic panels." After simulation evaluation (such as improving three-phase imbalance and reducing line losses), the parent strategy is selected, and the child strategy is generated through crossover and mutation operations. After repeated evolution, the preliminary regulation strategy is determined and stored.

[0044] Information Sharing Module: The intelligent agent in Area A sends preliminary strategies to the intelligent agents in Areas B and C via the community LAN. It also receives strategies from Area B (e.g., "adjusting the three-phase power distribution for shops") and Area C (e.g., "increasing the phase switching frequency of charging pile power supply") and calculates the load characteristic compatibility between Areas A and B. The calculation formula is: ,in, For the The strategy for each region is The load characteristic adaptability of each region, For the The strategic fitness of each region, The similarity of the load characteristics of the two regions is obtained by calculating the correlation coefficient of the historical load curves of the i-th region and the j-th region, and the received strategy and adaptation results are stored.

[0045] Collaborative Optimization Module: The intelligent agent in Area A combines real-time load (such as the current three-phase current imbalance) to optimize the strategy with higher adaptability in Area B (for example, changing "shops" to "residential buildings") and calculates the fitness of the optimized strategy. The calculation formula is: ,in, For the The new strategy fitness after region optimization, is the retention coefficient, and its value range is , For the The adaptability of the original strategy in each region, For the The strategy for each region is The load characteristic adaptability of each area is calculated by the formula: ,in, is the strategy fitness difference evaluation value, For the The new strategy fitness after region optimization, For the The adaptability of the original strategy in each region, is the difference threshold coefficient, which ranges from 0.05 to 0.1 and is used to set the minimum difference ratio for determining whether the new strategy is better. (After optimization, the three-phase imbalance is reduced and the line loss is reduced), then update the local strategy, such as Figure 1 shown.

[0046] Global optimization module: collects optimized strategies and fitness of areas A, B, and C every day , set weights based on regional load size and complexity , calculate the global optimization evaluation value , and its calculation formula is: ,in, To optimize the evaluation value globally; The total number of regions allocated for the distribution network; For the The weight of each region is set according to the regional load scale and regional complexity. For the The new strategy fitness after regional optimization is evaluated by the global optimization value. Ranking, if the strategy of area A ranks in the top 20% in the global evaluation, then a promotion instruction will be issued to areas B and C.

[0047] In summary, for the distribution network of urban residential communities, the community is divided into three independent areas and corresponding intelligent agents are configured. The collaborative evolutionary algorithm of the local optimization module is used to generate a preliminary regulation strategy. The information sharing module realizes inter-regional strategy interaction and fitness calculation. The collaborative optimization module optimizes and evaluates and updates the strategies with high fitness. The global optimization module regularly collects strategies for global evaluation and promotes the top 20% strategies to related areas, effectively realizing the dynamic balance regulation of the three-phase load of the community distribution network.

[0048] Example 2:

[0049] Regional division module: The power grid system is used to obtain the distribution network topology (including line distribution and transformer connection relationships) and load data (power load of production equipment in each factory building and power load of office areas) of an industrial park. The park is divided into three independent areas: Area A (heavy machinery plant), Area B (electronics processing plant), and Area C (office buildings and dormitories). These areas are assigned identifiers (A-001, B-002, and C-003). A high-performance intelligent agent is configured for Area A (to adapt to the characteristics of large-scale equipment and large load fluctuations), and conventional intelligent agents are configured for Areas B and C.

[0050] Local optimization module: The intelligent agent in Area A collects three-phase voltage, current, and active power data, removes the pulse noise of the welding equipment, and extracts features such as load impact frequency and three-phase current difference. It then generates an initial strategy population using a collaborative evolutionary algorithm, including the operation combinations of "adjusting the phase sequence of the punching machine connection" and "increasing the output of the factory's self-provided generator." After simulation evaluation and cross-mutation iteration, it determines the preliminary adjustment strategy and stores it, such as Figure 2 shown.

[0051] Information sharing module: The intelligent agent in Area A sends policies to Areas B and C via industrial Ethernet. It also receives policies from Area B (such as "adjusting the three-phase power supply distribution of the patch production line") and Area C (such as "controlling the split-phase power supply of the air conditioner in the dormitory area") and calculates the load characteristic compatibility between Areas A and B. The calculation formula is: and store the received strategies and fitness results.

[0052] Collaborative Optimization Module: The intelligent agent in Area A combines real-time load (such as the current three-phase voltage imbalance) to optimize the strategy with higher adaptability in Area B (such as changing the "patch production line" to the "stamping production line") and calculates the fitness of the optimized strategy. The calculation formula is: , and then calculate the formula: , calculated , the local policy is not updated.

[0053] Global optimization module: collects optimized strategies and fitness of area A, area B and area C every day , set weights based on the load scale and complexity of the three areas A, B and C , calculate the global optimization evaluation value , and its calculation formula is: , by evaluating the global optimization value Ranking, if the strategy of area B ranks in the top 20% in the global evaluation, then a promotion instruction will be issued to areas A and C.

[0054] In summary, for the industrial park distribution network, the regional division module first divides the regions and configures intelligent agents. The local optimization module generates preliminary adjustment strategies with the help of the collaborative evolutionary algorithm. The information sharing module promotes the interaction of strategies between regions and calculates the fitness. The collaborative optimization module optimizes the strategies with high fitness and decides whether to update them based on the evaluation results. The global optimization module collects strategies, conducts global evaluation, and then promotes the top 20% strategies, thereby achieving dynamic balanced adjustment of the three-phase load of the industrial park distribution network.

[0055] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A three-phase load dynamic balancing adjustment system and method for a distribution network, characterized in that: The system includes: Regional division module: The power grid system collects data on the distribution network topology and load distribution, divides the distribution network into multiple independent regions based on the collected data, assigns a unique identifier to each region and sets up an independent intelligent agent. The intelligent agent configuration adapts to the regional load scale and regional complexity; Local optimization module: This module uses an intelligent agent to obtain real-time data on three-phase voltage, current, and active power, and uses a co-evolutionary algorithm to generate and store a preliminary load regulation strategy for the region. Information sharing module: Each regional intelligent agent exchanges information through the communication network, sends the generated local load regulation strategy to the intelligent agents in adjacent regions, receives the strategies of adjacent regions at the same time, calculates the adaptability of load characteristics between regions, and stores the received strategies; Collaborative Optimization Module: This module optimizes the strategy with the highest adaptability based on the adaptability of the adjacent regional strategies and its own real-time load information. It then calculates the adaptability of the optimized strategy and compares it with the adaptability of the original strategy. Based on the comparison results, it decides whether to update the local strategy. Global optimization module: Regularly collects new strategies and the adaptability of new strategies after optimization in each region, sets weights based on regional load scale and regional complexity, conducts global optimization evaluation, and issues promotion instructions to relevant regions for strategies with good global adaptability.

2. A distribution network three-phase load dynamic balancing adjustment system and method according to claim 1, characterized in that: In the local optimization module, using the co-evolutionary algorithm to generate a preliminary load regulation strategy for the region includes the following steps: The intelligent agent cleans the acquired real-time data of three-phase voltage, current, and active power, removes outliers and noise data, extracts key characteristic parameters, and generates a standardized load data set; Based on the preprocessed load data set, combined with historical regulation experience and regional load characteristics, an initial strategy population containing multiple regulation operation combinations is randomly generated. The regulation operation combinations include adjusting user wiring methods and distributed generation output; Apply each strategy in the initial strategy population to the current load scenario for simulation, evaluate its effectiveness in improving three-phase imbalance and reducing line losses, and generate strategy fitness and strategy evaluation reports; Based on the strategy fitness and strategy evaluation report, the strategies with better performance are selected as the parent generation, and the strategies with poor performance are eliminated to form an elite strategy subset; Performing crossover and mutation operations on the elite strategy subset to generate child strategies that contain the advantageous features of the parent strategy, wherein the crossover and mutation operations include wiring adjustment and power control schemes that integrate different strategies; Merge the offspring strategy with the parent strategy, evaluate and screen again, and repeat the evolution process to gradually improve the adaptability and effectiveness of the strategy; When the number of iterations reaches the preset upper limit or the strategy effect converges, the strategy with the best performance is selected as the preliminary load regulation strategy for this area and formatted and stored.

3. A distribution network three-phase load dynamic balancing adjustment system and method according to claim 1, characterized in that: In the information sharing module, the inter-region load characteristic adaptability is calculated using the following formula: ,in, For the The strategy for each region is The load characteristic adaptability of each region, For the The strategic fitness of each region, is the similarity of load characteristics between the two regions obtained by calculating the correlation coefficient of the historical load curves of the i-th region and the j-th region.

4. A distribution network three-phase load dynamic balancing adjustment system and method according to claim 1, characterized in that: In the collaborative optimization module, the selected strategies are optimized based on the adaptability of the strategies in the adjacent regions and the real-time load information of the current three-phase voltage, current and active power data in the region. The optimization process includes adjusting the adjustment parameters in the strategy and modifying the adjustment operation steps based on the actual situation in the region.

5. A distribution network three-phase load dynamic balancing adjustment system and method according to claim 4, characterized in that: In the collaborative optimization module, the optimized strategy fitness is calculated using the following formula: ,in, For the The new strategy fitness after region optimization, is the retention coefficient, and its value range is , For the The fitness of the original strategy in each region, For the The strategy for each region is The load characteristic adaptability of each area.

6. A distribution network three-phase load dynamic balancing adjustment system and method according to claim 5, characterized in that: In the collaborative optimization module, the optimized strategy fitness is simulated and compared with the original strategy fitness. When deciding whether to update the local strategy based on the comparison results, the calculation formula is: ,in, is the strategy fitness difference evaluation value, For the The new strategy fitness after region optimization, For the The fitness of the original strategy in each region, is the difference coefficient, with a value range of 0.05-0.1, which is used to set the minimum difference ratio for determining whether the new strategy is better. The value of determines whether to update the local policy; when When the fitness of the optimized new strategy exceeds the fitness of the original strategy, the updated local strategy is determined to be the strategy after collaborative optimization. When , the fitness of the optimized new strategy does not exceed the fitness of the original strategy, and the original strategy is retained.

7. A distribution network three-phase load dynamic balancing adjustment system and method according to claim 1, characterized in that: In the global optimization module, the new strategies and the fitness of the new strategies after optimization in each region are collected regularly. , set the weight of each region according to the regional load scale and regional complexity , calculate the global optimization evaluation value , and its calculation formula is: ,in, To optimize the evaluation value globally; The total number of regions allocated for the distribution network; For the The weight of the region, For the The fitness of the new strategy after region optimization.

8. A distribution network three-phase load dynamic balancing adjustment system and method according to claim 7, characterized in that: In the global optimization module, the strategy with good global adaptability is: Ranked in the top 20% of all regional strategies.

9. A method for dynamically balancing and regulating three-phase loads in a distribution network, the method being applicable to a dynamic balancing and regulating system for three-phase loads in a distribution network according to any one of claims 1 to 9, characterized in that: The method includes: Regional division: Based on the topological structure and load distribution data of the distribution network, the distribution network is divided into multiple independent regions, and an independent intelligent agent is set up for each region; Local optimization: Each regional intelligent agent collects the three-phase load information of the region in real time, generates and stores the preliminary load regulation strategy for the region through the collaborative evolution algorithm; Information sharing: Each regional intelligent agent sends its own load regulation strategy to the intelligent agents in adjacent regions through the communication network, and simultaneously receives the load regulation strategies sent by adjacent regions, calculates the inter-region load characteristic adaptability, and stores it; Collaborative optimization: By combining the adaptability of adjacent regional strategies with its own load information, the received load regulation strategy is optimized to generate a more suitable load regulation strategy for the region. The optimized strategy fitness is calculated. If the fitness of the optimized new strategy exceeds the fitness of the original strategy, the strategy for the region is updated; otherwise, the original strategy is retained. Global optimization: Collect the optimized new strategies and the adaptability of the new strategies in each region, conduct a global evaluation, and issue promotion instructions to relevant regions for strategies with good global adaptability.

Images