Multi-zone-area ordered charging scheduling method and system based on simulated annealing genetic algorithm

Abstract

The invention provides a multi-zone-area ordered charging scheduling method and system based on a simulated annealing genetic algorithm, and the method comprises the steps: generating an initial population through employing a real number coding mode, and enabling a population individual to represent a charging scheduling scheme; constructing a fitness function in the genetic algorithm by taking load balancing and network loss minimization as optimization targets; selecting individuals by adopting a roulette selection and elitism strategy, and reserving a plurality of individuals with the highest fitness to enter the next generation by adopting a self-adaptive crossover mutation operation; taking the optimal individual of the current population as an initial solution, and executing simulated annealing operation; after optimization of a preset number of iterations, an optimal charging scheduling scheme is output, and the load rate, the voltage and the charging pile state of each transformer area are monitored in real time; and when abnormality is detected, charging power distribution is dynamically adjusted, and the load of the overload area is transferred to the low-load area. Through fusion of simulated annealing and a genetic algorithm, multi-zone-area charging scheduling is optimized, load balancing and network loss minimization are realized, and the stability and dynamic adaptability of a power grid are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of charging scheduling, and in particular to a multi-station area orderly charging scheduling method and system based on a simulated annealing genetic algorithm. Background Art

[0002] With the rapid adoption of electric vehicles, the charging load in power substations has exploded. These charging loads exhibit significant randomness and volatility, significantly impacting the power grid within these substations. On the one hand, these random load fluctuations can easily lead to load imbalances within the substation, causing overloads in some areas, increasing line voltage drops, and degrading user-side voltage quality, potentially even falling below normal levels, impacting the user's electricity experience. On the other hand, traditional single-substation optimization strategies face significant deficiencies in power supply reliability, resource utilization, and adaptability to load fluctuations when addressing the growth in charging loads.

[0003] Orderly charging scheduling across multiple power stations optimizes power distribution through cross-regional load transfer, but the optimization algorithms it relies on face key challenges: Traditional single-station strategies struggle to cope with complex load distributions, and in multi-station optimization, traditional algorithms (such as single genetic algorithms) lack multi-objective coordination and are prone to falling into local optimal solutions, resulting in limited load balancing and network loss control. While simulated annealing algorithms can escape local optimal solutions with a certain probability, their search efficiency is relatively low. Furthermore, existing methods are insufficiently adaptable to dynamic load changes and lack real-time monitoring and dynamic adjustment mechanisms, making it difficult to cope with real-time fluctuations in charging loads, resulting in reduced reliability in actual operation. Therefore, how to integrate algorithmic advantages and improve multi-objective optimization efficiency and dynamic adaptability has become a technical bottleneck in the current field of multi-station charging scheduling. Summary of the Invention

[0004] In order to solve the above problems, the present invention proposes a multi-station orderly charging scheduling method and system based on simulated annealing genetic algorithm, which effectively solves the multi-objective coordination problem in multi-station optimization by integrating the improved simulated annealing algorithm and genetic algorithm.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a multi-zone orderly charging scheduling method based on a simulated annealing genetic algorithm, comprising: The initial population is generated using real number coding, and each individual in the population represents a charging scheduling scheme; With load balancing and minimizing network loss as optimization goals, the fitness function in the genetic algorithm is constructed; Roulette wheel selection and elite retention strategies are used to select individuals, and adaptive crossover and mutation operations are used to retain the individuals with the highest fitness to enter the next generation; Take the optimal individual of the current population as the initial solution and perform simulated annealing operation; After optimization for a preset number of iterations, the optimal charging scheduling plan is output, and the load rate, voltage and charging pile status of each area are monitored in real time; when an abnormality is detected, the charging power distribution is dynamically adjusted to transfer part of the load of the overloaded area to the low-load area.

[0006] In a second aspect, the present invention provides a multi-zone orderly charging scheduling system based on a simulated annealing genetic algorithm, comprising: The initial population generation module is used to generate the initial population using real number coding. The population individuals represent a charging scheduling plan. The fitness function construction module is used to construct the fitness function in the genetic algorithm with load balancing and minimizing network loss as optimization goals; Genetic operation module, which is used to select individuals using roulette wheel selection and elite retention strategy, and adopt adaptive crossover and mutation operation to retain the individuals with the highest fitness to enter the next generation; The simulated annealing optimization module is used to perform simulated annealing operations using the optimal individual of the current population as the initial solution; The scheduling plan output and monitoring adjustment module is used to output the optimal charging scheduling plan after optimization for a preset number of iterations, and monitor the load rate, voltage and charging pile status of each area in real time; when an abnormality is detected, the charging power distribution is dynamically adjusted to transfer part of the load of the overloaded area to the low-load area.

[0007] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the multi-station orderly charging scheduling method based on a simulated annealing genetic algorithm described in the first aspect.

[0008] In a fourth aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the multi-station orderly charging scheduling method based on a simulated annealing genetic algorithm described in the first aspect are implemented.

[0009] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention combines an improved simulated annealing algorithm with a genetic algorithm, leveraging the strengths of both algorithms. The genetic algorithm's global search capabilities and the simulated annealing algorithm's local search capabilities and ability to escape local optima enable the algorithm to more effectively search for the global optimal solution, improving the optimization of orderly charging scheduling across multiple stations.

[0010] 2. By designing a reasonable encoding method, fitness function and improved genetic operation, the present invention can better handle the multi-objective coordination problem in multi-station optimization, achieve load balancing, minimize network losses, and improve the operating economy and power quality of the power grid.

[0011] 3. The present invention introduces adaptive crossover probability, mutation probability and dynamic variable step length, as well as a dynamic temperature adjustment strategy of the simulated annealing algorithm, which enables the algorithm to automatically adjust the search strategy according to the characteristics of the problem and the evolutionary state, thereby improving the convergence speed and search efficiency of the algorithm and reducing the calculation time.

[0012] 4. The real-time monitoring and dynamic adjustment mechanism of the present invention ensures that the dispatching plan can adapt to the load changes in the actual power grid operation, improves the stability and reliability of the power grid operation, and also improves the user's charging experience and satisfaction.

[0013] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their description are used to explain the present invention but do not constitute a limitation of the present invention.

[0015] Figure 1 A main flow chart of a multi-zone orderly charging scheduling method based on a simulated annealing genetic algorithm provided by an embodiment of the present invention; Figure 2 A flowchart of a multi-zone orderly charging scheduling method based on a simulated annealing genetic algorithm is provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0017] Example 1 like Figure 1 As shown, this embodiment discloses a multi-zone orderly charging scheduling method based on a simulated annealing genetic algorithm, comprising the following steps: S1: Generate the initial population using real number coding, where each individual in the population represents a charging scheduling scheme; S2: Taking load balancing and minimizing network loss as optimization goals, construct the fitness function in the genetic algorithm; S3: Roulette wheel selection and elite retention strategies are used to select individuals, and adaptive crossover and mutation operations are used to retain the individuals with the highest fitness to enter the next generation; S4: Take the optimal individual of the current population as the initial solution and perform simulated annealing operation; S5: After optimizing for a preset number of iterations, the optimal charging scheduling plan is output, and the load rate, voltage, and charging pile status of each area are monitored in real time. When an anomaly is detected, the charging power distribution is dynamically adjusted to transfer part of the load in the overloaded area to the underloaded area.

[0018] Next, combine Figure 2 , a multi-station area orderly charging scheduling method based on simulated annealing genetic algorithm disclosed in this embodiment is described in detail.

[0019] To address the existing problems of insufficient single-algorithm optimization accuracy, easy trapping in local optimality, and poor adaptability to complex nonlinear problems, this embodiment deeply integrates the genetic algorithm (GA) and the simulated annealing algorithm (SA) to form a collaborative optimization mechanism. First, the genetic algorithm is used as the main framework to generate an initial solution set and perform a global search through selection, crossover, and mutation operations. Then, after each generation of genetic operations, the simulated annealing algorithm is introduced to perform local fine optimization of the current optimal individual, leveraging its characteristic of probabilistically accepting inferior solutions to break through local extreme values. Finally, a cooling strategy is used to gradually converge to the global optimal solution.

[0020] This dynamic fusion method achieves complementary advantages of the two algorithms at different search stages, leveraging both the swarm intelligence and parallel search capabilities of the genetic algorithm and the local deep mining characteristics of the simulated annealing algorithm, significantly improving the optimization efficiency and solution quality of orderly charging scheduling in multiple areas.

[0021] (1) Charging load data collection and preprocessing Using smart meters, historical charging load data of each substation is collected based on preset step sizes to obtain more detailed load change information.

[0022] The collected data is cleaned to remove outliers and missing values. Missing values are filled using a moving average method. Outliers are corrected or removed based on the upper and lower limits of the data. The daily load data is then divided into multiple time windows for subsequent data analysis.

[0023] As an implementation method, in actual applications, smart meters installed in each substation continuously collect historical charging load data for at least one month, with a one-hour collection step. The collected data is cleaned. If the load data at a certain point in time significantly deviates from the normal range without a reasonable reason, it is identified as an outlier and removed. For a small number of missing values, the average of the adjacent data is used to fill in the gaps. The 24-hour daily load data is divided into 48 30-minute time windows to provide basic data for subsequent data analysis and optimized scheduling.

[0024] (2) Encoding and initializing the population The real number coding method is used to encode the charging scheduling plan of multiple areas.

[0025] Specifically, the charging power distribution of each substation in different time windows is used as a gene position to form a chromosome.

[0026] For example, if there are n stations and m time windows, the chromosome length is n × m. Initialize a population of size N, where each individual is a randomly generated feasible charging scheduling plan, ensuring that each plan meets constraints such as power balance, line capacity, and charging pile capacity.

[0027] As an implementation method, when there are three stations in a certain area, the upper limit of charging power of each station is . Using real number coding, the chromosome length is .

[0028] Initialize a population of 50 and ensure that each gene position (i.e. the charging power of each station in each time window) satisfies , while satisfying the power balance constraints of the entire power grid.

[0029] In this embodiment, real-number encoding uses the charging power allocation of each substation in different time windows as genetic bits to form chromosomes, intuitively reflecting the charging scheduling plan. Compared with other encoding methods, this method more accurately represents the solution space, avoids information loss during the encoding and decoding processes, facilitates the subsequent operation of the genetic algorithm, helps the algorithm converge to the global optimal solution more quickly, and improves the efficiency and quality of generating orderly charging scheduling plans for multiple substations.

[0030] (3) Fitness function design With load balancing and minimizing network loss as optimization goals, a comprehensive fitness function in the genetic algorithm is constructed.

[0031] First, the load balance degree is measured by calculating the standard deviation of the load rate of each area. The smaller the standard deviation, the more balanced the load. The load rate calculation formula is: Area load rate = Area actual load power / Area rated capacity. Assume that the mean of the area load rate is , then the fitness component of the load balancing objective is for: (1) in, For Taiwan The load factor, The total number of areas.

[0032] Secondly, network losses include line losses and transformer losses. Based on the topology of the power grid, line parameters, and the load conditions of each substation, the network losses are calculated using the power flow calculation method.

[0033] Assume the total network loss is , then the fitness component of minimizing network loss is: (2) The comprehensive fitness function is: (3) in, 、 is the weight coefficient, and ,The values of the weight coefficients are determined through multiple experiments and ,experience to balance the importance of different optimization objectives.

[0034] As an implementation method, after multiple tests and analyses, the weight coefficient is determined 、 According to the calculation formula of the fitness function in formula (3), the fitness value of each individual in the population is calculated.

[0035] In this embodiment, the load balancing degree is measured by the standard deviation of the load rate of the substation, the network loss is calculated using the power flow calculation method, and the relationship between the two is balanced by the weight coefficient. In this way, when searching for the optimal solution, the algorithm can take into account both the load balance of the power grid and the reduction of network loss, effectively solve the multi-objective coordination problem, and improve the economic efficiency and power quality of the power grid operation.

[0036] (IV) Select operation A combination of roulette wheel selection and elite retention strategy is adopted: roulette wheel selection is used to achieve probabilistic optimization of the population, so that individuals with high fitness are inherited to the next generation with a higher probability. At the same time, the elite retention strategy is used to directly retain the current optimal individuals, avoiding the loss of excellent genes due to random selection, balancing population diversity and convergence speed, and ensuring the stability and efficiency of the optimization direction.

[0037] First, the probability of each individual being selected is calculated based on their fitness. The higher the fitness, the greater the probability of selection. A roulette wheel is used to select some individuals for the next generation. At the same time, several individuals with the best fitness in the current population are retained and directly replicated to the next generation to avoid the loss of excellent genes.

[0038] As an implementation method, the selection probability of each individual in the population is calculated, such as individual The fitness of , the total fitness of the population is , then the individual The probability of selection . 30 individuals are selected to enter the next generation through a roulette wheel method, while the top 10 individuals in the current population are retained and directly copied to the next generation.

[0039] In this embodiment, roulette wheel selection determines selection probability based on individual fitness, giving individuals with high fitness a greater chance of advancing to the next generation, achieving a degree of meritocracy. The elite retention strategy directly retains the individuals with the best fitness in the current population, preventing the loss of excellent genes. The combination of these two strategies not only ensures population diversity, enabling the algorithm to search a larger solution space, but also ensures the inheritance of excellent solutions, accelerating algorithm convergence, increasing the probability of finding the global optimal solution, and optimizing multi-area charging scheduling.

[0040] (5) Cross Operation The traditional single-point crossover method is improved by combining multi-point crossover with adaptive crossover probability.

[0041] Multiple crossover points are randomly determined based on the length of the chromosome, and the genes on both sides of the crossover point are exchanged. Adaptive crossover probability Adjust according to the fitness of the individual. The higher the fitness, the lower the crossover probability, so as to retain the structure of excellent individuals. The specific formula is: (4) in, and are the maximum and minimum crossover probabilities, respectively, 、 and are the maximum fitness, average fitness and individual fitness in the population respectively. 's adaptability.

[0042] As an embodiment, the two parent individuals selected are , the chromosome length is 144. Three crossover points are randomly determined. The genes on both sides of the crossover point are exchanged to generate two offspring individuals. During the crossover process, the crossover probability is calculated according to the adaptive crossover probability formula , if the randomly generated number is less than , then perform a crossover operation.

[0043] In this embodiment, multi-point crossover increases the diversity of gene combinations by randomly determining multiple crossover points to exchange genes, expanding the algorithm's search range and helping to escape local optimal solutions. Adaptive crossover probability is adjusted based on individual fitness, with individuals with high fitness having a low crossover probability, thus preserving excellent individual structures. This combination strikes a balance between exploring new solutions and retaining high-quality genes, improving algorithm search efficiency and enabling genetic algorithms to better function in multi-station charging scheduling optimization.

[0044] (6) Mutation Operation Introduce adaptive mutation probability and dynamic variable step length.

[0045] Specifically, the mutation probability Adjustments are made based on the individual's fitness and evolutionary generations. The lower the fitness and the larger the evolutionary generations, the higher the probability of mutation, thus increasing the diversity of the population. The formula is: (5) in, and are the maximum and minimum mutation probabilities, respectively. is the minimum fitness in the population, is the current evolutionary generation, is the maximum evolutionary generation.

[0046] The mutation probability is dynamically adjusted according to individual fitness and evolutionary generations. When the fitness is low or the evolutionary generations are large, the mutation probability is increased, which prompts the algorithm to enhance its exploration ability and escape from local solutions when the population diversity is insufficient or stagnant.

[0047] Furthermore, the dynamic variable step length is adjusted based on the problem scale and the current evolutionary state. A larger variable step length is used in the early stages of evolution to accelerate search speed; a smaller variable step length is used in the later stages of evolution to conduct localized, refined searches. This effectively combines global search with local optimization, improving the algorithm's adaptability to complex solution spaces. It should be understood that the larger and smaller values are relative settings, and those skilled in the art can adjust them based on practical needs. This is not a limitation here.

[0048] In this embodiment, the adaptive mutation probability is adjusted based on individual fitness and the number of evolutionary generations. Low fitness or a large number of evolutionary generations results in a higher mutation probability, which increases population diversity and prevents premature convergence. The dynamically variable step length uses a larger value in the early stages of evolution to speed up the search, and a smaller value later for a more refined local search. This allows the algorithm to flexibly adjust mutation operations based on the evolutionary state, improving search efficiency, more quickly finding the global optimal solution to the problem of orderly charging scheduling across multiple stations, and enhancing the algorithm's adaptability to complex problems.

[0049] (VII) Simulated Annealing Algorithm Fusion After each generation of evolution in the genetic algorithm, a simulated annealing algorithm is introduced to further optimize the population. "Evolution" here refers to the complete iterative cycle of the genetic algorithm (the entire process from selection to mutation). Simulated annealing is introduced after each generation of genetic operations generates a new population, performing local optimization on the optimal individuals. By leveraging its probabilistic acceptance of inferior solutions, it compensates for the genetic algorithm's tendency to become trapped in a local optimum after mutation, thus enabling the two algorithms to complement each other during the iterative process.

[0050] Specifically, the optimal individual in the current population is used as the initial solution. According to the principle of simulated annealing algorithm, the solution is disturbed at a certain temperature to generate a new solution. The fitness difference between the new solution and the current solution is calculated. ,if , then accept the new solution; if , then with probability Accept new interpretations, including is the current temperature. As the iteration progresses, the temperature is lowered according to a certain cooling strategy. , so that the algorithm gradually converges to the global optimal solution. The cooling strategy adopts the exponential cooling method, that is, (6) in is the initial temperature, is the cooling coefficient , is the number of iterations.

[0051] In this embodiment, the simulated annealing algorithm uses the optimal individual in the current population as the initial solution and perturbs it at a certain temperature to generate a new solution. Acceptance is determined based on the fitness difference between the new solution and the current solution, as well as the temperature. The temperature is then lowered using an exponential cooling strategy. This algorithm exploits its characteristic of accepting inferior solutions with a certain probability, helping the algorithm escape local optima, compensating for the genetic algorithm's vulnerability to local optima. This complements the genetic algorithm's strengths, enabling a more efficient search for the global optimal solution and improving the optimization of orderly charging scheduling across multiple stations.

[0052] (8) Optimized dispatching of multiple stations After iterative optimization using an improved simulated annealing genetic algorithm, an optimal charging scheduling scheme was obtained. This scheme coordinates charging load distribution across multiple substations in a multi-station system, achieving load balancing, reducing network losses, and improving charging pile utilization. During the scheduling process, load changes, voltage conditions, and charging pile usage status at each substation are monitored in real time, and the scheduling scheme is dynamically adjusted based on actual conditions to ensure stable grid operation. As an implementation method, after 50 generations of iterative optimization using an improved SA genetic algorithm, an optimal charging scheduling scheme was obtained. In actual grid operation, the charging load of each substation is distributed according to this scheme. Simultaneously, a real-time monitoring system is used to collect data on the load, voltage, and charging pile usage status of each substation every 15 minutes. If a substation's load rate is found to be too high or its voltage is outside the normal range, the scheduling scheme is fine-tuned based on the actual situation, appropriately reducing the charging power of that substation and transferring part of the charging load to other substations with lighter loads, ensuring stable grid operation and normal charging needs of users.

[0053] This specific embodiment combines an improved simulated annealing algorithm with a genetic algorithm. The powerful global search capability of the genetic algorithm and the local search and local optimal jump-out characteristics of the simulated annealing algorithm enable the algorithm to find the global optimal solution more efficiently, thereby improving the optimization effect of orderly charging scheduling in multiple areas. At the same time, the reasonably designed encoding method, fitness function, and improved genetic operation can effectively handle multi-objective collaborative problems, achieve load balancing and minimize network losses, and improve the economic efficiency and power quality of power grid operation. The adaptive crossover probability, mutation probability, dynamic variable asynchronous length, and dynamic temperature adjustment strategy of simulated annealing enable the algorithm to automatically optimize the search strategy according to the problem and evolutionary state, accelerate convergence speed, improve search efficiency, and reduce calculation time. This real-time monitoring and dynamic adjustment mechanism can ensure that the scheduling plan can adapt to changes in grid load, ensure stable and reliable operation of the grid, and enhance the user charging experience.

[0054] Example 2 This embodiment provides a multi-zone orderly charging scheduling system based on a simulated annealing genetic algorithm, including: The initial population generation module is used to generate the initial population using real number coding. The population individuals represent a charging scheduling plan. The fitness function construction module is used to construct the fitness function in the genetic algorithm with load balancing and minimizing network loss as optimization goals; Genetic operation module, which is used to select individuals using roulette wheel selection and elite retention strategy, and adopt adaptive crossover and mutation operation to retain the individuals with the highest fitness to enter the next generation; The simulated annealing optimization module is used to perform simulated annealing operations using the optimal individual of the current population as the initial solution; The scheduling plan output and monitoring adjustment module is used to output the optimal charging scheduling plan after optimization for a preset number of iterations, and monitor the load rate, voltage and charging pile status of each area in real time; when an abnormality is detected, the charging power distribution is dynamically adjusted to transfer part of the load of the overloaded area to the low-load area.

[0055] Example 3 This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the multi-station orderly charging scheduling method based on a simulated annealing genetic algorithm as described in the first embodiment above are implemented.

[0056] Example 4 This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the multi-station orderly charging scheduling method based on a simulated annealing genetic algorithm as described in the first embodiment above are implemented.

[0057] The steps or modules involved in Examples 2 to 4 above correspond to those in Example 1. For detailed implementations, please refer to the relevant description of Example 1. The term "computer-readable storage medium" should be understood to mean a single medium or multiple media that includes one or more instruction sets; it should also be understood to include any medium that can store, encode, or carry an instruction set for execution by a processor and cause the processor to perform any method of the present invention.

[0058] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A multi-zone orderly charging scheduling method based on simulated annealing genetic algorithm, characterized in that: include: The initial population is generated using real number coding, and each individual in the population represents a charging scheduling scheme; With load balancing and minimizing network loss as optimization goals, the fitness function in the genetic algorithm is constructed; Roulette wheel selection and elite retention strategies are used to select individuals, and adaptive crossover and mutation operations are used to retain the individuals with the highest fitness to enter the next generation; Take the optimal individual of the current population as the initial solution and perform simulated annealing operation; After optimization for a preset number of iterations, the optimal charging scheduling plan is output, and the load rate, voltage and charging pile status of each area are monitored in real time; when an abnormality is detected, the charging power distribution is dynamically adjusted to transfer part of the load of the overloaded area to the low-load area.

2. The method for orderly charging scheduling of multiple zones based on simulated annealing genetic algorithm according to claim 1, characterized in that: The real number encoding is to determine the chromosomes of the individuals in the population according to the number of stations and preset time windows.

3. The method for orderly charging scheduling of multiple zones based on simulated annealing genetic algorithm according to claim 1, characterized in that: The fitness component of the load balancing objective is: in, For Taiwan The load factor, is the total number of districts; The fitness component of the network loss minimization objective is: 。 4. The method for orderly charging scheduling of multiple zones based on simulated annealing genetic algorithm according to claim 1, characterized in that: The adaptive crossover and mutation operation includes multi-point crossover and adaptive crossover probability, as well as adaptive mutation probability and dynamically variable step length.

5. The method for orderly charging scheduling of multiple zones based on simulated annealing genetic algorithm according to claim 1, characterized in that: The adaptive crossover operation is specifically as follows: multiple crossover points are randomly selected on the chromosome, and the genes on both sides of the crossover points of the parent individuals are exchanged; the crossover probability is dynamically adjusted with the individual fitness, and individuals with high fitness use a lower crossover probability to retain good genes.

6. The method for orderly charging scheduling of multiple zones based on simulated annealing genetic algorithm according to claim 1, characterized in that: The adaptive mutation operation is specifically as follows: the mutation probability is dynamically adjusted according to the individual fitness and evolutionary generations, and individuals with low fitness or large evolutionary generations adopt a higher mutation probability; the mutation step length is larger in the early stage of evolution for rapid search, and gradually decreases in the later stage for fine optimization.

7. The method for orderly charging scheduling of multiple zones based on simulated annealing genetic algorithm according to claim 1, characterized in that: The simulated annealing operation is specifically as follows: taking the current optimal individual as the initial solution, gradually reducing the temperature according to the exponential cooling strategy; after each perturbation generates a new solution, if the new solution is better, it is accepted; otherwise, the inferior solution is accepted according to the criterion of decreasing probability with temperature to avoid falling into the local optimum.

8. A multi-zone orderly charging scheduling system based on simulated annealing genetic algorithm, characterized by: include: The initial population generation module is used to generate the initial population using real number coding. The population individuals represent a charging scheduling plan. The fitness function construction module is used to construct the fitness function in the genetic algorithm with load balancing and minimizing network loss as optimization goals; Genetic operation module, which is used to select individuals using roulette wheel selection and elite retention strategy, and adopt adaptive crossover and mutation operation to retain the individuals with the highest fitness to enter the next generation; The simulated annealing optimization module is used to perform simulated annealing operations using the optimal individual of the current population as the initial solution; The scheduling plan output and monitoring adjustment module is used to output the optimal charging scheduling plan after optimization for a preset number of iterations, and monitor the load rate, voltage and charging pile status of each area in real time; when an abnormality is detected, the charging power distribution is dynamically adjusted to transfer part of the load of the overloaded area to the low-load area.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of a multi-station orderly charging scheduling method based on a simulated annealing genetic algorithm are implemented as described in any one of claims 1 to 7.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the multi-station area orderly charging scheduling method based on simulated annealing genetic algorithm are implemented as described in any one of claims 1 to 7.

Images