Renewable energy distribution optimization method based on genetic algorithm

Abstract

The invention discloses a renewable energy distribution optimization method based on a genetic algorithm. The method comprises the following steps: S1, generating an initial population; s2, defining a fitness function, and evaluating the fitness of each individual in the population; s3, an improved roulette selection method is adopted, and individuals are selected according to the fitness; s4, performing crossover operation on the selected individuals, and generating new individuals by adopting a multi-point crossover and uniform crossover combined method; s5, performing mutation operation on the new individuals, and adopting a self-adaptive mutation mechanism; s6, making a dynamic adjustment strategy; s7, new individuals generated through selection, intersection, variation and dynamic adjustment are used for replacing old individuals, and a new generation of population is formed; s8, repeating the steps S2-S7 until a preset termination condition is met; and S9, after a termination condition is met, outputting an individual with the highest fitness in the population as a final renewable energy source distribution scheme. Renewable energy distribution is optimized through the improved genetic algorithm, and efficient and stable distribution is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grids, and in particular to a renewable energy distribution optimization method based on a genetic algorithm. Background Art

[0002] With the continuous growth of global energy demand and the enhancement of environmental protection awareness, renewable energy has gradually become an important part of the modern power system. Renewable energy such as wind power, solar energy and hydropower has the advantages of being clean, renewable and widely distributed, and plays an important role in reducing greenhouse gas emissions and addressing climate change. However, due to the intermittent and volatile power generation characteristics of these renewable energy sources, their stable distribution and efficient utilization in the power system face many challenges.

[0003] At present, the problem of renewable energy allocation optimization has become a research hotspot in the field of smart grid and energy Internet. Traditional optimization methods mainly include linear programming, nonlinear programming, dynamic programming and heuristic algorithms. However, when dealing with large-scale, high-dimensional and nonlinear complex problems, these methods often have problems such as high computational complexity, long solution time, and easy to fall into local optimal solutions, which makes it difficult to meet the real-time and high-efficiency requirements in practical applications.

[0004] As an optimization algorithm that simulates natural selection and genetic mechanisms, genetic algorithms are widely used in renewable energy allocation optimization problems because of their strong global search capabilities, good adaptability, and easy parallel processing. Genetic algorithms evolve individuals in a population through genetic operations such as selection, crossover, and mutation, gradually approaching the optimal solution. However, traditional genetic algorithms still have defects and shortcomings in practical applications.

[0005] First, the diversity of the initial population is crucial to the global search capability of the genetic algorithm. If the initial population is not diverse enough, the algorithm will easily fall into the local optimal solution during the search process, reducing the optimization effect. Secondly, the traditional fitness function often only considers a single goal, such as power output stability or cost-effectiveness, and it is difficult to take into account multiple optimization goals at the same time, resulting in poor overall performance of the final allocation scheme. In addition, although the traditional roulette selection method can increase the probability of selecting excellent individuals, it has deficiencies in the individual importance weight and fitness normalization processing, which affects the effectiveness and stability of the selection operation.

[0006] In terms of crossover operations, traditional single-point or multi-point crossover methods easily lead to insufficient population diversity, affecting global search capabilities. Although uniform crossover can increase diversity, it lacks a dynamic adjustment mechanism and is difficult to adjust according to actual optimization needs. Mutation operations usually use a fixed mutation probability and fail to dynamically adjust the mutation probability according to population diversity and evolutionary stage, resulting in low search efficiency and easy to fall into local optimal solutions. Most existing methods are based on static data for optimization, which is difficult to adapt to the dynamic changes in actual power demand and environmental conditions, affecting the response speed and adaptability of the system.

[0007] Therefore, how to provide a renewable energy distribution optimization method based on genetic algorithm is a problem that those skilled in the art need to solve urgently. Summary of the invention

[0008] One object of the present invention is to propose a renewable energy allocation optimization method based on genetic algorithm. The present invention optimizes the renewable energy allocation scheme by improving the selection, crossover and mutation operations of the genetic algorithm to achieve efficient and stable energy allocation. The method comprehensively considers the stability of power output, cost-effectiveness and environmental benefits, adopts a multi-objective fitness function, dynamically adjusts the algorithm parameters, and enhances the global search capability and system adaptability. At the same time, the real-time response mechanism ensures that the system can quickly adapt to environmental changes. It has the advantages of strong global search capability, good adaptability, significant optimization effect and fast real-time response, meeting the actual needs of modern smart grids and energy Internet.

[0009] A renewable energy distribution optimization method based on a genetic algorithm according to an embodiment of the present invention comprises the following steps:

[0010] S1. According to the requirements of the renewable energy allocation problem, an initial population is generated, wherein the initial population includes a plurality of individuals;

[0011] S2. Define the fitness function and evaluate the fitness of each individual in the population according to the power output stability, cost-effectiveness and environmental benefit indicators of the allocation scheme;

[0012] S3, using the improved roulette selection method to select individuals based on fitness;

[0013] S4, performing a crossover operation on the selected individuals, and generating new individuals by combining an improved multi-point crossover and a dynamic uniform crossover method;

[0014] S5, perform mutation operations on new individuals, adopt an adaptive mutation mechanism, and automatically adjust the mutation probability according to population diversity;

[0015] S6. Use new individuals generated by selection, crossover, mutation and dynamic adjustment to replace old individuals and form a new generation of population;

[0016] S7, repeating S2-S6 until a preset termination condition is met, where the termination condition includes a maximum number of iterations or the fitness of the population reaching a preset threshold;

[0017] S8. After the termination conditions are met, the individual with the highest fitness in the population is output as the final renewable energy allocation plan.

[0018] Furthermore, the S1 specifically includes:

[0019] S11. According to the specific requirements of the renewable energy allocation problem, a dataset containing different types of renewable energy, including wind, solar, and hydropower, is constructed;

[0020] S12. The allocation ratio of each renewable energy source is represented as a gene. All genes constitute the chromosome of an individual. The size of the initial population is set to N, and the length of the chromosome of each individual is set to L.

[0021] S13, randomly generate the initial value of each gene in the range [0,1];

[0022] S14, generating some initial individuals based on historical data and experience values, and mixing them with randomly generated individuals to increase the diversity and quality of the initial population;

[0023] S15. Verify the diversity of the initial population to ensure that there are no highly similar individuals in the population. The diversity is evaluated by calculating the genetic difference D of individuals in the population:

[0024]

[0025] Among them, N represents the population size, L represents the chromosome length, i and j represent the individual number, G ik represents the kth gene value of individual i, G jk represents the k-th gene value of individual j, represents the average value of the k-th gene in the population;

[0026] S16. The initial population that has passed the diversity verification is saved in a database and used as a starting point for genetic algorithm optimization.

[0027] Furthermore, the fitness function is specifically expressed as:

[0028]

[0029] Wherein, F(x) represents the fitness function, S(x) represents the stability of power output, C(x) represents the cost-effectiveness, E(x) represents the environmental benefit, w1 represents the weight coefficient of power output stability, w2 represents the weight coefficient of cost-effectiveness, and w3 represents the weight coefficient of environmental benefit, satisfying w1+w2+w3=1;

[0030] The calculation of power output stability S(x) is measured by statistically analyzing the variance of power output according to the allocation scheme represented by individual x:

[0031]

[0032] Among them, P t represents the power output at time t, represents the average power output in the time period, T represents the total duration of the time period, and ∈ represents a constant;

[0033] Calculation of cost-effectiveness C(x), calculate the total cost-effectiveness of the allocation scheme represented by individual x:

[0034]

[0035] Among them, c i represents the unit cost of the i-th renewable resource, r i (x) represents the proportion of the i-th renewable energy in individual x, k i represents the fixed cost of the i-th renewable energy source, δ represents a constant, and n represents the total number of renewable energy sources;

[0036] Calculation of environmental benefit E(x): Calculate the environmental benefit of the allocation scheme represented by individual x:

[0037]

[0038] Among them, e i represents the unit environmental benefit of the i-th renewable energy source, r i (x) represents the proportion of the i-th renewable energy in individual x, and ζ represents a constant.

[0039] Furthermore, the S3 specifically includes:

[0040] S31, set the selection probability, perform selection operations on individuals in the population, and select individuals with high fitness to enter the next generation;

[0041] S32. Normalize the fitness value of each individual in the population to ensure that the calculation of the selection probability is more accurate:

[0042]

[0043] Among them, f′i represents the normalized fitness value of individual i, f i represents the original fitness value of individual i, f min represents the minimum fitness value of individuals in the population, f max Indicates the maximum fitness value of individuals in the population;

[0044] S33. Calculate the selection probability of each individual according to the normalized fitness value and the importance weight of the individual:

[0045]

[0046] Among them, P i represents the selection probability of individual i, w i represents the importance weight of individual i, N represents the total number of individuals in the population, Represents the sum of all individuals’ weighted normalized fitness values;

[0047] S34. Using an improved roulette wheel selection method, selecting individuals to enter the next generation according to the calculated selection probability, including:

[0048] Calculate the cumulative selection probability for each individual:

[0049]

[0050] Among them, C i represents the cumulative selection probability of individual i, P k represents the selection probability of individual k;

[0051] Generate a random number r between 0 and 1;

[0052] Select the first one from the population that satisfies C i Individuals with ≥r enter the next generation;

[0053] S35. Perform selection operation on the entire population and repeat step S34 until a new population is generated.

[0054] Furthermore, the S4 specifically includes:

[0055] S41. Select multiple crossover points in the parent individuals and generate new offspring individuals by exchanging gene fragments at the crossover points:

[0056] Offspring individual 1 = parent individual 1 [1:i1] + parent individual 2 [i1 + 1:i2] + ... + parent individual 1 [i n-1 +1:n];

[0057] Offspring individual 2 = parent individual 2[1:i1] + parent individual 1[i1+1:i2] + ... + parent individual 2[i n-1 +1:n];

[0058] Among them, i1, i2, ..., i n-1 represents the crossover point position selected dynamically, parent individual 1 and parent individual 2 represent the two parent individuals involved in the crossover operation, and offspring individual 1 and offspring individual 2 represent the offspring individuals generated by the improved multi-point crossover;

[0059] S42. According to the dynamic uniform crossover strategy, the genes of the parent individuals are selected for exchange with uniform probability, and dynamic adjustments are made according to population diversity and optimization requirements:

[0060]

[0061] Among them, j represents the gene position, r represents the dynamically adjusted data uniformly distributed between [0,1], which is adjusted according to the fitness and diversity of the population;

[0062] S43. Generate a new individual with genetic information of two parent individuals through crossover operation.

[0063] Furthermore, the S5 specifically includes:

[0064] S51. Set the mutation probability p m , perform mutation operations on new individuals, increase population diversity by randomly changing individual gene values, and prevent the algorithm from falling into a local optimal solution;

[0065] S52, randomly selecting one or more gene positions in the individual gene sequence for mutation, wherein the selection of the mutation position is based on a random number generator;

[0066] S53, performing mutation operations on the gene at the selected mutation position, including bit flip mutation, uniform random mutation, Gaussian mutation and nonlinear mutation;

[0067] S54, dynamically adjust the mutation probability p according to the current population diversity, individual fitness, evolutionary stage and mutation history m :

[0068]

[0069] Among them, p m0 represents the initial mutation probability, σ represents the standard deviation of the current population, σ max represents the maximum population standard deviation, g represents the current evolutionary generation, and g max represents the maximum evolutionary generation, ΔF represents the difference between the current individual fitness and the average fitness of the population, and F max represents the maximum fitness of individuals in the population, H m represents the number of mutations of an individual in the mutation history, H max Indicates the maximum number of mutation history;

[0070] S55. Generate new individuals through mutation operations, with new gene values.

[0071] The beneficial effects of the present invention are:

[0072] First, by constructing a data set containing different types of renewable energy, and combining random generation with the generation of some initial individuals based on historical data and empirical values, the diversity of the initial population is ensured. The diversity of the initial population is crucial to the global search capability of the genetic algorithm. The present invention uses a diversity verification mechanism to evaluate the diversity by calculating the genetic differences of individuals in the population, thereby avoiding the initial population from being too concentrated in a local area, thereby enhancing the global search capability and improving the optimization effect.

[0073] Secondly, the present invention adopts a multi-objective fitness function, comprehensively considering the three key indicators of power output stability, cost-effectiveness and environmental benefits, and achieves a balance between different optimization objectives through weight adjustment. This method ensures that the overall performance of the optimization results is superior and better adapts to practical application needs. Traditional fitness functions often only consider a single objective and it is difficult to take into account multiple optimization objectives at the same time. The present invention significantly improves the optimization effect through comprehensive trade-offs.

[0074] In terms of selection operation, the present invention adopts an improved roulette selection method by introducing the importance weight of individuals and normalizing the fitness values, calculating the selection probability according to the normalized fitness values ​​and weights. This improved method improves the retention and reproduction probability of excellent individuals in the selection operation, ensuring the steady improvement of population quality. Although the traditional roulette selection method can improve the selection probability of excellent individuals, it has deficiencies in the individual importance weight and fitness normalization processing. The improvement measures of the present invention significantly improve the effectiveness and stability of the selection operation.

[0075] In terms of crossover operation, the present invention adopts a method combining improved multi-point crossover and dynamic uniform crossover, and increases the diversity of the population by dynamically adjusting the intersection position and crossover strategy, preventing the algorithm from falling into the local optimum. This method can not only be adjusted according to population diversity and optimization requirements, but also improves optimization efficiency and global search capabilities. Traditional single-point or multi-point crossover methods easily lead to insufficient population diversity and affect global search capabilities, while the crossover strategy of the present invention significantly enhances the diversity and adaptability of the population.

[0076] The present invention also introduces an adaptive mutation mechanism, which dynamically adjusts the mutation probability according to the current population diversity, individual fitness, evolutionary stage, and mutation history, thereby improving the algorithm's global search capability and convergence speed. By dynamically adjusting the mutation probability, the limitations brought by the fixed mutation probability are avoided, and the adaptability of the algorithm at different evolutionary stages is enhanced. This innovative mutation mechanism significantly improves the algorithm's search efficiency and prevents it from falling into a local optimal solution.

[0077] In addition, the present invention realizes the real-time response of the system to the actual power demand and environmental conditions by dynamically adjusting the fitness function and operating parameters, thereby improving the adaptability and response speed of the system. Most traditional methods are optimized based on static data, which is difficult to adapt to the dynamic changes of actual power demand and environmental conditions. The real-time response mechanism of the present invention enables the optimization process to quickly adapt to changes in the external environment, ensuring the effectiveness and flexibility of the renewable energy allocation scheme in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0079] Figure 1 A flowchart of a renewable energy distribution optimization method based on a genetic algorithm proposed by the present invention;

[0080] Figure 2 A flow chart of a method for combining multi-point crossover and dynamic uniform crossover in a renewable energy distribution optimization method based on a genetic algorithm proposed by the present invention;

[0081] Figure 3 This is a flow chart of the adaptive mutation mechanism of the renewable energy distribution optimization method based on genetic algorithm proposed by the present invention. DETAILED DESCRIPTION

[0082] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0083] refer to Figure 1-3 , a renewable energy allocation optimization method based on genetic algorithm, comprising the following steps:

[0084] S1. According to the requirements of the renewable energy allocation problem, an initial population is generated, wherein the initial population includes a plurality of individuals;

[0085] S2. Define the fitness function and evaluate the fitness of each individual in the population according to the power output stability, cost-effectiveness and environmental benefit indicators of the allocation scheme;

[0086] S3, using the improved roulette selection method to select individuals based on fitness;

[0087] S4, performing a crossover operation on the selected individuals, and generating new individuals by combining an improved multi-point crossover and a dynamic uniform crossover method;

[0088] S5, perform mutation operations on new individuals, adopt an adaptive mutation mechanism, and automatically adjust the mutation probability according to population diversity;

[0089] S6. Use new individuals generated by selection, crossover, mutation and dynamic adjustment to replace old individuals and form a new generation of population;

[0090] S7, repeating S2-S6 until a preset termination condition is met, where the termination condition includes a maximum number of iterations or the fitness of the population reaching a preset threshold;

[0091] S8. After the termination conditions are met, the individual with the highest fitness in the population is output as the final renewable energy allocation plan.

[0092] In this implementation, S1 specifically includes:

[0093] S11. According to the specific requirements of the renewable energy allocation problem, a dataset containing different types of renewable energy, including wind, solar, and hydropower, is constructed;

[0094] S12. The allocation ratio of each renewable energy source is represented as a gene. All genes constitute the chromosome of an individual. The size of the initial population is set to N, and the length of the chromosome of each individual is set to L.

[0095] S13, randomly generate the initial value of each gene in the range [0,1];

[0096] S14, generating some initial individuals based on historical data and experience values, and mixing them with randomly generated individuals to increase the diversity and quality of the initial population;

[0097] S15. Verify the diversity of the initial population to ensure that there are no highly similar individuals in the population. The diversity is evaluated by calculating the genetic difference D of individuals in the population:

[0098]

[0099] Among them, N represents the population size, L represents the chromosome length, i and j represent the individual number, G ik represents the kth gene value of individual i, G jk represents the k-th gene value of individual j, represents the average value of the k-th gene in the population;

[0100] S16. The initial population that has passed the diversity verification is saved in a database and used as a starting point for genetic algorithm optimization.

[0101] In this implementation, the fitness function is specifically expressed as:

[0102]

[0103] Wherein, F(x) represents the fitness function, S(x) represents the stability of power output, C(x) represents the cost-effectiveness, E(x) represents the environmental benefit, w1 represents the weight coefficient of power output stability, w2 represents the weight coefficient of cost-effectiveness, and w3 represents the weight coefficient of environmental benefit, satisfying w1+w2+w3=1;

[0104] The calculation of power output stability S(x) is measured by statistically analyzing the variance of power output according to the allocation scheme represented by individual x:

[0105]

[0106] Among them, P t represents the power output at time t, represents the average power output in the time period, T represents the total duration of the time period, and ∈ represents a constant;

[0107] Calculation of cost-effectiveness C(x), calculate the total cost-effectiveness of the allocation scheme represented by individual x:

[0108]

[0109] Among them, c i represents the unit cost of the i-th renewable resource, r i (x) represents the proportion of the i-th renewable energy in individual x, k i represents the fixed cost of the i-th renewable energy source, δ represents a constant, and n represents the total number of renewable energy sources;

[0110] Calculation of environmental benefit E(x): Calculate the environmental benefit of the allocation scheme represented by individual x:

[0111]

[0112] Among them, e i represents the unit environmental benefit of the i-th renewable energy source, r i (x) represents the proportion of the i-th renewable energy in individual x, and ζ represents a constant.

[0113] In this implementation, S3 specifically includes:

[0114] S31, set the selection probability, perform selection operations on individuals in the population, and select individuals with high fitness to enter the next generation;

[0115] S32. Normalize the fitness value of each individual in the population to ensure that the calculation of the selection probability is more accurate:

[0116]

[0117] Among them, f′ i represents the normalized fitness value of individual i, f i represents the original fitness value of individual i, f min represents the minimum fitness value of individuals in the population, f max Indicates the maximum fitness value of individuals in the population;

[0118] S33. Calculate the selection probability of each individual according to the normalized fitness value and the importance weight of the individual:

[0119]

[0120] Among them, P i represents the selection probability of individual i, w i represents the importance weight of individual i, N represents the total number of individuals in the population, Represents the sum of all individuals’ weighted normalized fitness values;

[0121] S34. Using an improved roulette wheel selection method, selecting individuals to enter the next generation according to the calculated selection probability, including:

[0122] Calculate the cumulative selection probability for each individual:

[0123]

[0124] Among them, C i represents the cumulative selection probability of individual i, P k represents the selection probability of individual k;

[0125] Generate a random number r between 0 and 1;

[0126] Select the first one from the population that satisfies C i Individuals with ≥r enter the next generation;

[0127] S35. Perform selection operation on the entire population and repeat step S34 until a new population is generated.

[0128] In this implementation manner, the S4 specifically includes:

[0129] S41. Select multiple crossover points in the parent individuals and generate new offspring individuals by exchanging gene fragments at the crossover points:

[0130] Offspring individual 1 = parent individual 1 [1:i1] + parent individual 2 [i1 + 1:i2] + ... + parent individual 1 [i n-1 +1:n];

[0131] Offspring individual 2 = parent individual 2[1:i1] + parent individual 1[i1+1:i2] + ... + parent individual 2[i n-1 +1:n];

[0132] Among them, i1, i2, ..., i n-1 represents the crossover point position selected dynamically, parent individual 1 and parent individual 2 represent the two parent individuals involved in the crossover operation, and offspring individual 1 and offspring individual 2 represent the offspring individuals generated by the improved multi-point crossover;

[0133] S42. According to the dynamic uniform crossover strategy, the genes of the parent individuals are selected for exchange with uniform probability, and dynamic adjustments are made according to population diversity and optimization requirements:

[0134]

[0135] Among them, j represents the gene position, r represents the dynamically adjusted data uniformly distributed between [0,1], which is adjusted according to the fitness and diversity of the population;

[0136] S43. Generate a new individual with genetic information of two parent individuals through crossover operation.

[0137] In this implementation manner, S5 specifically includes:

[0138] S51. Set the mutation probability p m , perform mutation operations on new individuals, increase population diversity by randomly changing individual gene values, and prevent the algorithm from falling into a local optimal solution;

[0139] S52, randomly selecting one or more gene positions in the individual gene sequence for mutation, wherein the selection of the mutation position is based on a random number generator;

[0140] S53, performing mutation operations on the gene at the selected mutation position, including bit flip mutation, uniform random mutation, Gaussian mutation and nonlinear mutation;

[0141] S54, dynamically adjust the mutation probability p according to the current population diversity, individual fitness, evolutionary stage and mutation history m :

[0142]

[0143] Among them, p m0 represents the initial mutation probability, σ represents the standard deviation of the current population, σ max represents the maximum population standard deviation, g represents the current evolutionary generation, and g max represents the maximum evolutionary generation, ΔF represents the difference between the current individual fitness and the average fitness of the population, and F max represents the maximum fitness of individuals in the population, H m represents the number of mutations of an individual in the mutation history, F max Indicates the maximum number of mutation history;

[0144] S55. Generate new individuals through mutation operations, with new gene values.

[0145] Embodiment 1:

[0146] In order to verify the feasibility of the present invention in implementation, the present invention is applied to a smart grid system in a certain province, aiming to achieve dynamic optimization allocation of renewable energy and improve the overall operation efficiency of the power system by improving the genetic algorithm.

[0147] In practical applications, we first constructed a data set containing multiple renewable energy sources such as wind, solar, and hydropower according to the specific requirements of the province's power system. The generation of the initial population not only relies on randomly generated individuals, but also combines some initial individuals generated based on historical data and experience values ​​to ensure the diversity of the initial population. By calculating the genetic differences of individuals in the population, diversity verification is performed to ensure that the population has a good initial distribution.

[0148] When defining the fitness function, we comprehensively considered three key indicators: stability of power output, cost-effectiveness, and environmental benefits. The fitness function achieves a balanced optimization of multiple objectives by adjusting the weights of each indicator. The specific fitness function combines factors such as the variance of power output, the unit cost and proportion of renewable energy, and the unit environmental benefit to ensure that the final allocation plan can achieve the best balance between multiple objectives.

[0149] In the selection operation, we introduced the importance weight of individuals and normalized the fitness value of each individual in the population. The selection probability is calculated based on the normalized fitness value and weight, and an improved roulette selection method is used to ensure the retention and reproduction probability of excellent individuals in the selection operation. The crossover operation adopts a combination of improved multi-point crossover and dynamic uniform crossover. By dynamically adjusting the crossover point position and crossover strategy, the diversity of the population is increased to prevent the algorithm from falling into the local optimum.

[0150] In terms of mutation operations, the mutation probability is dynamically adjusted according to the current population diversity, individual fitness, evolutionary stage, and mutation history. The adaptive mutation mechanism dynamically adjusts the mutation probability by calculating the population standard deviation, the current evolutionary generation, the difference between the individual fitness and the average fitness of the population, and the number of mutations of the individual in the mutation history in real time, avoiding the limitations of fixed mutation probability and enhancing the adaptability of the algorithm at different evolutionary stages.

[0151] Table 1 Comparison of renewable energy distribution optimization effect data

[0152]

[0153] By analyzing Table 1, we can clearly see the significant effect of the present invention in practical application. First, in terms of power output variance, the variance of power output before optimization was 10.5%, which was reduced to 3.8% after optimization, indicating that the stability of power output has been improved by 64%. This change significantly reduces the fluctuation of power output, which is of great significance to the stable operation of the power grid. In terms of cost-effectiveness, the average power generation cost was $0.12 per kilowatt-hour before optimization, which was reduced to $0.10 after optimization, and the cost-effectiveness was improved by 16.7%. This not only reduces the cost of power generation, but also improves the economic benefits of renewable energy. In terms of annual average carbon dioxide emissions, it was 5 million tons before optimization and reduced to 4.5 million tons after optimization, reducing carbon emissions by 10%. This data shows that the present invention has made a significant contribution to environmental protection and meets the current requirements of green development. In terms of system response time, the average system adjustment response time during the peak period of power demand before optimization was 15 minutes, which was shortened to 5 minutes after optimization, an increase of three times. The fast response time ensures that the system can respond to demand changes in a timely manner and improves the reliability of the power grid. In terms of optimization calculation time consumption, each allocation optimization calculation took about 2 hours before optimization, which was shortened to 30 minutes after optimization, and the calculation efficiency was improved four times. This means that the present invention can perform optimization calculations more quickly and improve the overall efficiency. In terms of the fluctuation range of power output during the maximum load period, the fluctuation range was 15% before optimization, and it was reduced to 5% after optimization, and the fluctuation range was reduced by 66.7%. This further proves the effectiveness of the optimization method in improving the stability of power output. In terms of system stability index, the index was 70 before optimization and increased to 90 after optimization, an increase of 28.6%. This shows that the optimization method significantly enhances the overall stability of the system. In terms of renewable energy utilization rate, the utilization rate was 75% before optimization and increased to 85% after optimization, an increase of 13.3%. This means that more renewable energy has been effectively utilized and energy utilization efficiency has been improved. In terms of peak power demand satisfaction rate, the satisfaction rate was 92% before optimization and increased to 98% after optimization, an increase of 6.5%. This result shows that the optimization method improves the power supply capacity during peak demand periods. In terms of total system cost, the total system cost was US$120 million per year before optimization, and it was reduced to US$100 million per year after optimization, and the total cost was reduced by 16.7%. This shows the significant effect of the optimization method in reducing system operating costs. In terms of environmental protection score, the score before optimization was 85 / 100, and it increased to 92 / 100 after optimization, an increase of 8.2%. This further shows the positive impact of the optimization method on environmental protection.

[0154] Based on the above data analysis, the present invention achieves efficient and stable distribution of renewable energy by improving the genetic algorithm, which not only improves the operating efficiency and stability of the power system, but also significantly reduces the cost of power generation and environmental impact. It has broad application prospects and practical value.

[0155] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A renewable energy allocation optimization method based on genetic algorithm, It is characterized in that The steps include: S1. According to the requirements of the renewable energy allocation problem, an initial population is generated, wherein the initial population includes a plurality of individuals; S2. Define the fitness function and evaluate the fitness of each individual in the population according to the power output stability, cost-effectiveness and environmental benefit indicators of the allocation scheme; S3, using the improved roulette selection method to select individuals based on fitness; S4, performing a crossover operation on the selected individuals, and generating new individuals by combining an improved multi-point crossover and a dynamic uniform crossover method; S5, perform mutation operations on new individuals, adopt an adaptive mutation mechanism, and automatically adjust the mutation probability according to population diversity; S6. Use new individuals generated by selection, crossover, mutation and dynamic adjustment to replace old individuals and form a new generation of population; S7, repeating S2-S6 until a preset termination condition is met, where the termination condition includes a maximum number of iterations or the fitness of the population reaching a preset threshold; S8. After the termination conditions are met, the individual with the highest fitness in the population is output as the final renewable energy allocation plan.

2. The renewable energy distribution optimization method based on genetic algorithm according to claim 1, It is characterized in that The S1 specifically includes: S11. According to the specific requirements of the renewable energy allocation problem, a dataset containing different types of renewable energy, including wind, solar, and hydropower, is constructed; S12. The allocation ratio of each renewable energy source is represented as a gene. All genes constitute the chromosome of an individual. The size of the initial population is set to N, and the length of the chromosome of each individual is set to L. S13, randomly generate the initial value of each gene in the range [0,1]; S14, generating some initial individuals based on historical data and experience values, and mixing them with randomly generated individuals; S15. Verify the diversity of the initial population and evaluate the diversity by calculating the genetic difference D of individuals in the population: Among them, N represents the population size, L represents the chromosome length, i and j represent the individual number, G ik represents the kth gene value of individual i, G jk represents the k-th gene value of individual j, represents the average value of the k-th gene in the population; S16. The initial population that has passed the diversity verification is saved in a database and used as a starting point for genetic algorithm optimization.

3. The renewable energy distribution optimization method based on genetic algorithm according to claim 1, It is characterized in that The specific expression of the fitness function is: Among them, F(x) represents the fitness function, S(x) represents the power output stability, C(x) represents the cost benefit, E(x) represents the environmental benefit, and w 1 The weight coefficient representing the stability of power output, w 2 Represents the weight coefficient of cost-effectiveness, w 3 Represents the weight coefficient of environmental benefits, satisfying w 1 +w 2 +w 3 =1; The calculation of power output stability S(x) is measured by statistically analyzing the variance of power output according to the allocation scheme represented by individual x: Among them, P t represents the power output at time t, represents the average power output in the time period, T represents the total duration of the time period, and ∈ represents a constant; Calculation of cost-effectiveness C(x), calculate the total cost-effectiveness of the allocation scheme represented by individual x: Among them, c i represents the unit cost of the i-th renewable resource, r i (x) represents the proportion of the i-th renewable energy in individual x, k i represents the fixed cost of the i-th renewable energy source, δ represents a constant, and n represents the total number of renewable energy sources; Calculation of environmental benefit E(x): Calculate the environmental benefit of the allocation scheme represented by individual x: Among them, e i represents the unit environmental benefit of the i-th renewable energy source, r i (x) represents the proportion of the i-th renewable energy in individual x, and ζ represents a constant.

4. The renewable energy distribution optimization method based on genetic algorithm according to claim 1, It is characterized in that The S3 specifically includes: S31, set the selection probability, perform selection operations on individuals in the population, and select individuals with high fitness to enter the next generation; S32. Normalize the fitness value of each individual in the population: Among them, f′ i represents the normalized fitness value of individual i, f i represents the original fitness value of individual i, f min represents the minimum fitness value of individuals in the population, f max Indicates the maximum fitness value of individuals in the population; S33. Calculate the selection probability of each individual according to the normalized fitness value and the importance weight of the individual: Among them, P i represents the selection probability of individual i, w i represents the importance weight of individual i, N represents the total number of individuals in the population, Represents the sum of all individuals’ weighted normalized fitness values; S34. Using an improved roulette wheel selection method, selecting individuals to enter the next generation according to the calculated selection probability, including: Calculate the cumulative selection probability for each individual: Among them, C i represents the cumulative selection probability of individual i, P k represents the selection probability of individual k; Generate a random number r between 0 and 1; Select the first one from the population that satisfies C i Individuals with ≥r enter the next generation; S35. Perform selection operation on the entire population and repeat step S34 until a new population is generated.

5. The renewable energy distribution optimization method based on genetic algorithm according to claim 1, It is characterized in that The S4 specifically includes: S41. Select multiple crossover points in the parent individuals and generate new offspring individuals by exchanging gene fragments at the crossover points: Offspring individual 1 = parent individual 1 [1:i 1 ]+parent individual 2[i 1 +1:i 2 ]+…+ Parent individual 1[i n-1 +1:n]; Offspring individual 2 = parent individual 2 [1:i 1 ]+parent individual 1[i 1 +1:i 2 ]+…+ Parent individual 2[i n-1 +1:n]; Among them, i 1 ,i 2 ,……,i n-1 represents the crossover point position selected dynamically, parent individual 1 and parent individual 2 represent the two parent individuals involved in the crossover operation, and offspring individual 1 and offspring individual 2 represent the offspring individuals generated by the improved multi-point crossover; S42. According to the dynamic uniform crossover strategy, the genes of the parent individuals are selected for exchange with uniform probability, and dynamic adjustments are made according to population diversity and optimization requirements: Among them, j represents the gene position, and r represents the dynamically adjusted data uniformly distributed between [0,1]; S43. Generate a new individual with genetic information of two parent individuals through crossover operation.

6. The method for optimizing renewable energy distribution based on genetic algorithm according to claim 1, It is characterized in that The S5 specifically includes: S51. Set the mutation probability p m , perform mutation operation on the new individuals; S52, randomly selecting one or more gene positions in the individual gene sequence for mutation, wherein the selection of the mutation position is based on a random number generator; S53, performing mutation operations on the gene at the selected mutation position, including bit flip mutation, uniform random mutation, Gaussian mutation and nonlinear mutation; S54, dynamically adjust the mutation probability p according to the current population diversity, individual fitness, evolutionary stage and mutation history m : Among them, p m0 represents the initial mutation probability, σ represents the standard deviation of the current population, σ max represents the maximum population standard deviation, g represents the current evolutionary generation, and g max represents the maximum evolutionary generation, ΔF represents the difference between the current individual fitness and the average fitness of the population, and F max represents the maximum fitness of individuals in the population, H m represents the number of mutations of an individual in the mutation history, H max Indicates the maximum number of mutation history; S55. Generate new individuals through mutation operations, with new gene values.

Images