Water-light complementary optimization method and system based on competitive reproduction and adaptive vector

Abstract

The invention discloses a competitive reproduction and adaptive vector-based water-light complementation optimization method and system. The method comprises the following steps: determining an optimization target, constructing a multi-objective function, and establishing a comprehensive multi-objective function model; initializing a production population, wherein each individual in the population represents a potential solution; by executing competitive breeding operation, selecting superior individuals for breeding according to fitness, dynamically adjusting a search strategy in combination with a self-adaptive adjustment mechanism, and iteratively improving the solution quality of the whole population; evaluating the current iterative solution set by using a Pareto optimal standard, and adaptively adjusting a reference vector according to an evaluation result; and when iteration reaches a predetermined termination condition, outputting an optimal solution set meeting a multi-objective optimization requirement. The complementary characteristics of hydraulic power generation and photovoltaic power generation are effectively integrated, and the power generation efficiency and the system stability are optimized. The operation cost is reduced, the energy utilization efficiency is improved, and the adaptive capacity and robustness of the system in an uncertain environment are remarkably enhanced.

Description

Technical Field

[0001] The present invention relates to the field of high-dimensional multi-objective optimization and cross-technology of water-photovoltaic complementary power systems, and in particular to a water-photovoltaic complementary optimization method and system based on competitive reproduction and adaptive vectors. Background Art

[0002] The hydro-photovoltaic hybrid system is a hybrid energy system that combines hydropower and photovoltaic power generation. This system takes advantage of the complementary characteristics of the two renewable energy sources: photovoltaic power generation is mainly used during sunny days, while hydropower generation is used to supplement electricity demand at night or on cloudy days. The system aims to improve energy efficiency, reduce the instability of power supply, and reduce dependence on fossil fuels. When optimizing the hydro-photovoltaic hybrid system, multiple objectives usually need to be considered, such as power generation cost, system reliability, environmental impact, and energy efficiency. There may be conflicts between these objectives, so it is necessary to find a balance point in multi-objective optimization. In addition, as the number of objectives increases, the optimization problem becomes more complex and faces the following challenges. When optimizing the hydro-photovoltaic hybrid system, multiple objectives usually need to be considered, such as power generation cost, system reliability, environmental impact, and energy efficiency. There may be conflicts between these objectives, so it is necessary to find a balance point in multi-objective optimization. In addition, as the number of objectives increases, the optimization problem becomes more complex and faces the following challenges: (1) The curse of dimensionality: As the number of objectives increases, the search space grows exponentially, and traditional optimization algorithms are difficult to efficiently search for the optimal solution. (2) Diversity and uniformity of solutions: In high-dimensional multi-objective optimization, it is very difficult to find a Pareto frontier solution set that is uniformly distributed and diverse. (3) Computational resource requirements: High-dimensional multi-objective optimization usually requires a lot of computing resources and time, especially when dealing with complex systems such as water-photovoltaic complementary systems. The high-dimensional multi-objective optimization method based on adaptive reference vectors establishes a mathematical model of the water-photovoltaic complementary system, defines optimization objectives such as total power generation cost, system reliability, environmental impact and energy utilization efficiency, uses uniformly generated reference vectors and randomly generated initial populations, dynamically adjusts the position and direction of the reference vectors, performs evolutionary operations such as crossover, mutation and non-dominated sorting, gradually optimizes the population, and finally obtains the Pareto frontier solution set. This method effectively improves the overall performance and sustainability of the water-photovoltaic complementary system. Summary of the invention

[0003] In view of the above-mentioned problems, the present invention is proposed.

[0004] Therefore, the technical problem solved by the present invention is: how to effectively integrate and optimize multiple conflicting objectives in high-dimensional multi-objective optimization problems, especially how to achieve an optimal balance between cost efficiency and energy utilization in a water-photovoltaic complementary system.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a water-light complementarity optimization method based on competitive reproduction and adaptive vectors, comprising: establishing optimization goals, constructing multi-objective functions, and establishing a comprehensive multi-objective function model;

[0006] Initialize the production population, where each individual in the population represents a potential solution to a multi-objective function model;

[0007] By executing competitive reproduction operations, the winning individuals are selected for reproduction according to fitness, and the search strategy is dynamically adjusted in combination with the adaptive adjustment mechanism to iteratively improve the solution quality of the entire population;

[0008] In each iteration, the current solution set is evaluated using the Pareto optimality criterion, and the reference vector is adaptively adjusted based on the evaluation results;

[0009] When the iteration reaches the predetermined termination condition, the optimal solution set that meets the multi-objective optimization requirements is output.

[0010] As a preferred solution of the water-photovoltaic complementary optimization method based on competitive reproduction and adaptive vector described in the present invention, wherein: the optimization objectives include minimizing cost minF1(x), maximizing power generation efficiency MaxF2(x), maximizing the robustness of power generation of the water-photovoltaic complementary system maxF3(x), and maximizing photovoltaic power generation and reducing thermal power generation maxF4(x);

[0011] The establishment of a comprehensive multi-objective function model:

[0012]

[0013] P min <P<P max

[0014] α≥0; β≥0

[0015] -P(x) 储能容里max ≤P(x) 需求 ≤P(x) 储能容里max

[0016] W min (x)≤W(x)≤W max (x)

[0017] E 储能t+1 (x) = E 储能t (x)+τP

[0018] Among them, OC(x)1 represents the operating cost of the thermal power unit, SC(x)1 represents the start-up and shutdown cost of the thermal power unit, PC(x)1 represents the penalty cost of abandonment of photovoltaic power generation system, and SC(x)2 represents the start-up and shutdown cost of pumped storage power station; while ensuring the maximum power generation efficiency, the constraints must be met: P 火 =P 需求 -P 总 , P 总 =P 光伏 +P 蓄水 ;P 总 Represents the total power generation of the system, P 光伏 Represents the power of photovoltaic power generation, P 蓄水 Indicates the power of hydroelectric power; P 需求 represents the total power generation demand, P 火 represents the power generation required by the thermal power unit, η and μ represent the weights of hydropower generation and photovoltaic power generation; P 水力发电 represents the amount of hydroelectric power, P 光伏发电 represents photovoltaic power generation; λ represents the weight coefficient, and σ represents the uncertainty and volatility in system operation;

[0019] Meeting power generation efficiency requirements constraints:

[0020] P 光伏 +P 水力 +P 火力 =P 需求

[0021] Photovoltaic power generation capacity constraints:

[0022] 0≤P 光伏 ≤P 光伏max

[0023] Hydropower capacity constraints:

[0024] 0≤P 水力 ≤P 水力max

[0025] Thermal power generation capacity constraints:

[0026] 0≤P 火カ ≤P 火力max

[0027] Energy equipment charging and discharging restrictions:

[0028] -P 储能容量max ≤P 需求 ≤P 储能容量max

[0029] Energy storage balance equation:

[0030] E 储能 ( t+1 )=E储能 ( t )+л * P 储电 -g * P 放电

[0031] Among them, E 储能 ( t ) and E 储能 ( t+1 ) represent the energy storage capacity of the current and next time step respectively; л and g represent the weights of charging efficiency and discharging efficiency respectively; P 储电 Indicates the power stored by the energy storage device, P 放电 Indicates the power of the energy storage device at the discharge point.

[0032] As a preferred solution of the water-photovoltaic complementary optimization method based on competitive reproduction and adaptive vector described in the present invention, wherein: the initialization of the production population includes initializing the population:

[0033] O={x1,x2,x3,...,x n}

[0034] Where n represents the number of individuals in the population:

[0035] x i =[x I ,x2,x3,...,x m ]

[0036] Among them, x i Represents an individual in a population.

[0037] As a preferred solution of the water-light complementary optimization method based on competitive reproduction and adaptive vector described in the present invention, wherein: the performing competitive reproduction operation includes randomly selecting two parent populations O p x1, x2 in the parent population O, and remove x1, x2 from the parent population O P Deleted;

[0038] By evaluating the fitness functions of the two individuals, the individual with high fitness is selected to compete and become the winner x w , failed individual x l They will learn from the winning individuals, and both will mutate and enter the offspring population O f , the formula is:

[0039] x 变异后 =x 变异前 +Δ j

[0040]

[0041] Among them, x变异前 Represents the decision variables of the winner and the loser after learning, and the increased mutation offset Δ is determined according to the mutation probability j Then generate x 变异后 ;μ j represents the random generation probability of a random number, and η represents the parameter of mutation intensity;

[0042] The x formed after the competition operation w , x l The decision variables of each position in the , based on the probability premise, are mutated within a given range, and two individuals are selected cyclically to compete until the parent population O P All individuals in the offspring population O f ;

[0043] The parent population O P With the offspring population O f Merge to generate a merged population Q;

[0044] A competitive selection operation is performed, where the reference vector of V is associated with all individuals in the population Q, i.e., the reference vectors with the smallest angle between the individual and the reference vector in V, according to the angle between the individual and the reference vector;

[0045] Traverse all reference vectors associated with individuals, and for each reference vector, select the individual with the smallest APD value from the associated individuals to enter the next generation population O P ;

[0046] Then traverse the reference vectors that are not associated with individuals. For the reference vectors that are not associated with individuals, select the individual with the smallest angle from the remaining individuals in the population to enter the next generation, where the APD distance is defined as:

[0047] Among them, d t,i,j represents the APD distance from node i to node j at time t; P represents the penalty function about θ, which is expressed as:

[0048]

[0049] γ νt,j =min<ν t,i ,ν t,j >

[0050] Among them, α represents a user-defined parameter that controls the intensity of the time factor’s influence on the penalty, t represents the current number of iterations, and t max represents the maximum number of iterations; γ νt,j represents the normalization factor, γ represents the t , j forms the angle of the other reference vectors with the smallest angle; the angle γ is used to normalize the angle given by ν t, the angle in the subspace specified by j, generates a uniform Pareto-optimal solution set; M represents a constant; θ t,i,j Represents with ν t,i The angle of the other parameter vectors that forms the minimum angle.

[0051] As a preferred solution of the water-light complementary optimization method based on competitive reproduction and adaptive vector described in the present invention, the adaptive adjustment mechanism includes the calculation of the next reference vector generation method by the maximum and minimum values ​​of the objective function in the individuals in the high-quality population, and the formula is as follows:

[0052]

[0053] Among them, V t represents the parameter vector of the tth iteration, r(x) max represents the maximization objective function, f(x) min represents the minimization objective function;

[0054] The maximization and minimization of the objective function is used to guide the reference vector to change and then guide the population to evolve in the direction of the objective function.

[0055] As a preferred scheme of the water-photovoltaic complementary optimization method based on competitive reproduction and adaptive vector described in the present invention, wherein: the use of the Pareto optimality criterion to evaluate the current solution set includes repeating the cycle of initializing the production population and the competitive reproduction operation until the termination condition is met and the number of fitness evaluations reaches the maximum number.

[0056] As a preferred solution of the water-photovoltaic complementary optimization method based on competitive reproduction and adaptive vectors described in the present invention, wherein: the output of the optimal solution set that meets the multi-objective optimization requirements includes judging whether the number of fitness evaluations of the population meets a threshold, and if the requirement for the number of solutions of the output water-photovoltaic complementary system optimization solution set is met, outputting the result;

[0057] Otherwise, the competitive breeding operation is repeated until the termination condition is met.

[0058] The water-light complementary optimization system based on competitive reproduction and adaptive vector is characterized by: including:

[0059] Objective function construction module: establish optimization objectives, construct multi-objective functions, and establish a comprehensive multi-objective function model;

[0060] Population initialization module: initializes the production population, each individual in the population represents a potential solution;

[0061] Reproduction and fitness evaluation module: By performing competitive reproduction operations, the winning individuals are selected for reproduction according to fitness, and the search strategy is dynamically adjusted in combination with the adaptive adjustment mechanism to iteratively improve the solution quality of the entire population;

[0062] Adaptive reference vector adjustment module: In each iteration, the current solution set is evaluated using the Pareto optimality criterion, and the reference vector is adaptively adjusted based on the evaluation results;

[0063] Iteration termination and solution set output module: When the iteration reaches the predetermined termination condition, the optimal solution set that meets the multi-objective optimization requirements is output.

[0064] The beneficial effects of the present invention are as follows: by utilizing a high-dimensional competitive group algorithm to optimize the water-photovoltaic complementary power generation mode, a comprehensive, diverse and high-quality solution set can be obtained, thereby providing important decision-making support for the planning, operation and management of the water-photovoltaic complementary system. By comprehensively considering multiple objectives for modeling, the conflicts between multiple optimization objectives that need to be considered in the operation of water-photovoltaic complementary are resolved. Multi-objective optimization can better solve the complexity and dynamics of the water-photovoltaic complementary system, thereby improving the robustness of the system. The water-photovoltaic complementary operation mode faces uncertainty and complex variability, and multi-objective optimization can help the system better adapt to and respond to changes in the external environment. Therefore, the method of the present invention not only improves the utility of the water-photovoltaic complementary system, but also provides important decision-making support for the planning, operation and management of the water-photovoltaic complementary power generation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the figures required for use in the description of the embodiments will be briefly introduced below. Obviously, the figures described below are only some embodiments of the present invention. For ordinary technicians in this field, other figures can be obtained based on these figures without paying creative work.

[0066] Figure 1 A flow chart of a water-photovoltaic complementarity optimization method based on competitive breeding and adaptive vectors provided in the first embodiment of the present invention;

[0067] Figure 2 An explanation diagram of the solution process of a high-dimensional competition optimization algorithm of a water-light complementary optimization method based on competition reproduction and adaptive vectors provided in the first embodiment of the present invention;

[0068] Figure 3 Comparison of the effects of the water-light complementary optimization method based on competitive propagation and adaptive vector provided by the first embodiment of the present invention and the greedy benchmark method considering comprehensive evaluation indicators Figure 1 ;

[0069] Figure 4 Comparison of the effects of the water-light complementary optimization method based on competitive propagation and adaptive vector provided by the first embodiment of the present invention and the greedy benchmark method considering comprehensive evaluation indicators Figure 2 ;

[0070] Figure 5 Comparison of the effects of the water-light complementary optimization method based on competitive propagation and adaptive vector provided by the first embodiment of the present invention and the greedy benchmark method considering comprehensive evaluation indicators Figure 3 ;

[0071] Figure 6 Comparison effect of the water-light complementary optimization method based on competitive breeding and adaptive vector provided by the first embodiment of the present invention with the current mainstream optimization algorithm Figure 1 ;

[0072] Figure 7 Comparison effect of the water-light complementary optimization method based on competitive breeding and adaptive vector provided by the first embodiment of the present invention with the current mainstream optimization algorithm Figure 2 ;

[0073] Figure 8 Comparison effect of the water-light complementary optimization method based on competitive breeding and adaptive vector provided by the first embodiment of the present invention with the current mainstream optimization algorithm Figure 3 . DETAILED DESCRIPTION

[0074] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the specification drawings. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0075] Example 1, reference Figures 1-2 , which is an embodiment of the present invention, provides a water-light complementarity optimization method based on competitive breeding and adaptive vectors, including:

[0076] S1: Establish optimization goals, construct multi-objective functions, and establish a comprehensive multi-objective function model.

[0077] The multi-objective function includes cost minimization:

[0078] Min F1(x)=OC(x)1+SC(x)1+PC(x)1+SC(x)2

[0079] Among them, OC(x)1 represents the operating cost of the thermal power unit, SC(x)1 represents the start-up and shutdown cost of the thermal power unit, PC(x)1 represents the penalty cost of abandonment of photovoltaic power generation system, and SC(x)2 represents the start-up and shutdown cost of the pumped storage power station;

[0080] Maximize power generation efficiency:

[0081] MaxF2(x)=ηP 光伏 (x)+μP 蓄水 (x)

[0082] To ensure maximum power generation efficiency, constraints must be met: P 火 =P 需求 -P 总 , P 总 =P 光伏 +P 蓄水 ;

[0083] Among them, P 总 Represents the total power generation of the system, P 光伏 Represents the power of photovoltaic power generation, P 蓄水 Indicates the power of hydroelectric power; P 需求 represents the total power generation demand, P 火 represents the power generation required to be supplemented by the thermal power unit, η and μ represent the weights of hydropower generation and photovoltaic power generation;

[0084] Maximizing the robustness of power generation in the water-photovoltaic hybrid system:

[0085]

[0086] Among them, P 水力发电 represents the amount of hydroelectric power, P 光伏发电 represents photovoltaic power generation; λ represents the weight coefficient, and σ represents the uncertainty and volatility in system operation. The formula is:

[0087]

[0088] Maximize photovoltaic power generation and reduce thermal power generation:

[0089]

[0090] Establish a mathematical problem model for multi-objective optimization:

[0091]

[0092] P min <P<P max

[0093] α≥0; β≥0

[0094] -P(x) 储能容里max ≤P(x) 需求 ≤P(x) 储能容里max

[0095] W min (x)≤W(x)≤W max (x)

[0096] E储能t+1 (x) = E 储能t (x)+τP

[0097] Meeting power generation efficiency requirements constraints:

[0098] P 光伏 +P 水力 +P 火力 =P 需求

[0099] Photovoltaic power generation capacity constraints:

[0100] 0≤P 光伏 ≤P 光伏max

[0101] Hydropower capacity constraints:

[0102] 0≤P 水力 ≤P 水力max

[0103] Thermal power generation capacity constraints:

[0104] 0≤P 火カ ≤P 火力max

[0105] Energy equipment charging and discharging restrictions:

[0106] -P 储能容量max ≤P 需求 ≤P 储能容量max

[0107] Energy storage balance equation:

[0108] E 储能(t+1) =E 储能(t0 +л * P 储电 -g * P 放电

[0109] Among them, E 储能(t0 and E 储能(t+1) Represent the energy storage capacity of the current and next time step respectively; л and g represent the weights of charging efficiency and discharging efficiency respectively; P 储电 Indicates the power stored by the energy storage device, P 放电 Indicates the power of the energy storage device at the discharge point.

[0110] S2: Initialize the production population. Each individual in the population represents a potential solution to a multi-objective function model. First, initialize the population in the target space:

[0111] O={x1,x2,x3,...,x n}

[0112] Each individual in the population is a candidate solution, representing a possible solution to the water-photovoltaic complementary system.

[0113] Where n represents the number of individuals in the population:

[0114] x i =[x1,x2,x3,...,x m ]

[0115] Among them, x i Represents individuals in the population, including m-dimensional decision variables, representing parameters in the hydro-photovoltaic complementary power generation system.

[0116] S3: By performing competitive reproduction operations, the winning individuals are selected for reproduction according to fitness, and the search strategy is dynamically adjusted in combination with the adaptive adjustment mechanism to iteratively improve the solution quality of the entire population.

[0117] Randomly select 2 parent populations O p x1, x2 in the parent population O, and remove x1, x2 from the parent population O P Deleted.

[0118] By evaluating the fitness functions of the two individuals, the individual with high fitness is selected to compete and become the winner x w , failed individual x l They will learn from the winning individuals, and both will mutate and enter the offspring population O f , expressed as:

[0119]

[0120] The variation of winning individuals and losing individuals is shown in the following formula:

[0121] x 变异后 =x 变异前 +Δ j

[0122]

[0123] Among them, x 变异前 Represents the decision variables of the winner and the loser after learning, and the increased mutation offset Δ is determined according to the mutation probability j Then generate x 变异后 ;μ j represents the probability of random generation of a random number, and η represents the parameter of mutation intensity.

[0124] By improving the diversity of the solution set through competitive selection and polynomial mutation, the optimal solution can be found in a shorter time. This is because in each generation of evolution, only the best individuals or some of the better individuals will be selected as parents, thus ensuring the rapid spread and accumulation of excellent genes in the population. It provides a more efficient operation solution for the operation of the light complementary system.

[0125] The x formed after the competition operation w , x l The decision variables of each position in the , based on the probability premise, are mutated within a given range, and two individuals are selected cyclically to compete until the parent population O P All individuals in the offspring population O f ;

[0126] The parent population O P With the offspring population O f Merge to generate a merged population Q;

[0127] A competitive selection operation is performed, where the reference vector of V is associated with all individuals in the population Q, i.e., the reference vectors with the smallest angle between the individual and the reference vector in V, according to the angle between the individual and the reference vector;

[0128] Traverse all reference vectors associated with individuals, and for each reference vector, select the individual with the smallest APD value from the associated individuals to enter the next generation population O P ;

[0129] Then traverse the reference vectors that are not associated with individuals. For the reference vectors that are not associated with individuals, select the individual with the smallest angle from the remaining individuals in the population to enter the next generation, where the APD distance is defined as:

[0130] Among them, d t,i,j represents the APD distance from node i to node j at time t; P represents the penalty function about θ, which is expressed as:

[0131]

[0132] γ νt,j =min<ν t,i ,ν t,j

[0133] Among them, α represents a user-defined parameter that controls the intensity of the time factor’s influence on the penalty, t represents the current number of iterations, and t max represents the maximum number of iterations; γ νt,j represents the normalization factor, γ represents the t , j forms the angle of the other reference vectors with the smallest angle; the angle γ is used to normalize the angle given by ν t, the angle in the subspace specified by j, generates a uniform Pareto-optimal solution set; M represents a constant; θ t,i,j Represents and ν t,i The angle of the other parameter vectors that forms the minimum angle.

[0134] Generally speaking, multi-objective optimization emphasizes convergence in the early stage and diversity in the later stage. The penalty function is designed based on this problem. In the early stage, t≤tmax, so P≈0, which means d≈||f′||, ||f′|| emphasizes convergence. In the later stage, t gradually increases, and the influence of P also increases accordingly, so as to achieve the purpose of emphasizing diversity in the later stage.

[0135] Furthermore, the angle γ is used to normalize the t , the angle in the subspace specified by j. When the distribution of some reference vectors is too dense (or too sparse), resulting in an extremely small (or large) angle between the candidate solution and the reference vector, it is easy to fall into the dilemma of local optimal solution. Therefore, normalization can generate a uniform Pareto quality solution set.

[0136] The next reference vector generation method is calculated by the maximum and minimum values ​​of the objective function among individuals in the high-quality population. The formula is as follows:

[0137]

[0138] Among them, V t represents the parameter vector of the tth iteration, r(x) max represents the maximization objective function, f(x) min represents the minimization objective function;

[0139] The maximization and minimization of the objective function is used to guide the reference vector to change and then guide the population to evolve in the direction of the objective function.

[0140] Realize dynamic and adaptive adjustment of the reference vector to guide the reference vector close to the direction vector optimized by the objective function.

[0141] It should be noted that evolutionary operation is one of the core steps in the multi-objective optimization algorithm. By performing operations such as crossover and mutation on individuals in the population, new individuals are generated, thereby continuously improving the quality and diversity of the current solution. In each round of iteration, the CSO algorithm divides the individuals in the population into several pairs for competition. The winning individuals are selected by evaluating the fitness of the population, and the winning individual population guides the evolution of individuals in the population to achieve dynamic adaptive adjustment.

[0142] S4: In each iteration, the current solution set is evaluated using the Pareto optimality criterion, and the reference vector is adaptively adjusted based on the evaluation results.

[0143] The repeated cycle is the initialization of the production population and the competitive breeding operation until the termination condition is met and the number of fitness evaluations reaches 100,000 times.

[0144] Furthermore, the iterative loop determines whether the solution set satisfies the maximum number of iterations and whether the number of solutions in the solution set satisfies the Pareto solution set requirements of the water-photovoltaic complementary system.

[0145] S5: When the iteration reaches the predetermined termination condition, the optimal solution set that meets the multi-objective optimization requirements is output.

[0146] Determine whether the number of fitness evaluations of the population meets the threshold. If it meets the requirements for the number of solutions of the output water-photovoltaic complementary system optimization solution set, output the result; otherwise, the competition reproduction operation is circulated until the termination condition is met.

[0147] It should be noted that Figure 2 This is an explanation diagram of the CSO algorithm in the solution process. The present invention proposes a multi-objective optimization method based on a competitive group optimization algorithm for the power grid operation of a water-photovoltaic complementary system. The competitive group optimization algorithm introduces a competition mechanism. The competition mechanism is used to eliminate the inferior solution set, and the candidate solution is updated through evolutionary operations such as competition and mutation to generate the next generation of individuals. Then, the optimal solution set that meets the cost, efficiency, and thermal power generation power is obtained.

[0148] The above embodiments also include a water-light complementary optimization system based on competitive reproduction and adaptive vectors, specifically:

[0149] Objective function construction module: establish optimization objectives, construct multi-objective functions, and establish a comprehensive multi-objective function model;

[0150] Population initialization module: initializes the production population, each individual in the population represents a potential solution;

[0151] Reproduction and fitness evaluation module: By performing competitive reproduction operations, the winning individuals are selected for reproduction according to fitness, and the search strategy is dynamically adjusted in combination with the adaptive adjustment mechanism to iteratively improve the solution quality of the entire population;

[0152] Adaptive reference vector adjustment module: In each iteration, the current solution set is evaluated using the Pareto optimality criterion, and the reference vector is adaptively adjusted based on the evaluation results;

[0153] Iteration termination and solution set output module: When the iteration reaches the predetermined termination condition, the optimal solution set that meets the multi-objective optimization requirements is output.

[0154] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of the power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a water-light complementary optimization method based on competitive breeding and adaptive vectors is implemented.

[0155] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0156] Example 2, reference Figures 3 to 8 , which is an embodiment of the present invention, provides a water-photovoltaic complementarity optimization method based on competitive reproduction and adaptive vectors. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation / comparative experiments.

[0157] Figure 3 and Figure 4 The water-photovoltaic complementary system optimization method of the present invention is more efficient than the greedy algorithm which only considers the current state. Figure 5 It is shown that the cost of optimizing the water-photovoltaic complementary system of the present invention is lower than that of the greedy algorithm which only considers the current state. Figure 6 It is shown that the robustness of the present invention in optimizing the water-photovoltaic complementary system is superior to the benchmark method. Figure 7-8It shows that compared with other single-objective optimization algorithms, this paper is superior to other comparison algorithms in terms of cost, power generation efficiency, and power generation robustness. It shows that the high-dimensional multi-objective water-photovoltaic complementary system optimization method of the present invention can effectively reduce carbon emissions, save power grid power generation costs, and ensure that the new energy power generation system can stably and continuously generate electricity.

[0158] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A water-light complementary optimization method based on competitive reproduction and adaptive vector, characterized in that: include: Establish optimization goals, construct multi-objective functions, and establish a comprehensive multi-objective function model; Initialize the production population, where each individual in the population represents a potential solution to a multi-objective function model; By executing competitive reproduction operations, the winning individuals are selected for reproduction according to fitness, and the search strategy is dynamically adjusted in combination with the adaptive adjustment mechanism to iteratively improve the solution quality of the entire population; In each iteration, the current solution set is evaluated using the Pareto optimality criterion, and the reference vector is adaptively adjusted based on the evaluation results; When the iteration reaches the predetermined termination condition, the optimal solution set that meets the multi-objective optimization requirements is output.

2. The water-light complementary optimization method based on competitive reproduction and adaptive vector according to claim 1, characterized in that: The optimization objectives include minimizing cost minF1(x), maximizing power generation efficiency MaxF2(x), maximizing the robustness of power generation of the water-photovoltaic complementary system maxF3(x), and maximizing photovoltaic power generation and reducing thermal power generation maxF4(x); The establishment of a comprehensive multi-objective function model: P min <P<P max α≥0； β≥0 -P(x) 储能容里max ≤P(x) 需求 ≤P(x) 储能容里max W min (x)≤W(x)≤W max (x) E 储能t+1 (x)=E 储能t (x)+τP Among them, OC(x)1 represents the operating cost of the thermal power unit, SC(x)1 represents the start-up and shutdown cost of the thermal power unit, PC(x)1 represents the penalty cost of abandonment of photovoltaic power generation system, and SC(x)2 represents the start-up and shutdown cost of pumped storage power station; while ensuring the maximum power generation efficiency, the constraints must be met: P 火 =P 需求 -P 总 , P 总 =P 光伏 +P 蓄水 ;P 总 Represents the total power generation of the system, P 光伏 Represents the power of photovoltaic power generation, P 蓄水 Indicates the power of hydroelectric power; P 需求 represents the total power generation demand, P 火 represents the power generation required by the thermal power unit, η and μ represent the weights of hydropower generation and photovoltaic power generation; P 水力发电 represents the amount of hydroelectric power, P 光伏发电 represents photovoltaic power generation; λ represents the weight coefficient, and σ represents the uncertainty and volatility in system operation; Meeting power generation efficiency requirements: P 光伏 +P 水力 +P 火力 =P 需求 Photovoltaic power generation capacity constraints: 0≤P 光伏 ≤P 光伏max Hydropower capacity constraints: 0≤P 水力 ≤P 水力max Thermal power generation capacity constraints: 0≤P 火カ ≤P 火力max Energy equipment charging and discharging restrictions: -P 储能容量max ≤P 需求 ≤P 储能容量max Energy storage balance equation: HAVE BEEN 储能 ( t+1 )=E 储能 ( t )+l * Q 储电 -g * Q 放电 Among them, E 储能 ( t ) and E 储能 ( t+1 ) represent the energy storage capacity of the current and next time step respectively; л and g represent the weights of charging efficiency and discharging efficiency respectively; P 储电 Indicates the power stored by the energy storage device, P 放电 Indicates the power of the energy storage device at the discharge point.

3. The water-light complementary optimization method based on competitive breeding and adaptive vector according to claim 2, characterized in that: The initialization of the production population includes initializing the population: O={x1,x2,x3,...,x n } Where n represents the number of individuals in the population: x i =[x I ,x2,x3,...,x m ] Among them, x i Represents an individual in a population.

4. The water-light complementary optimization method based on competitive breeding and adaptive vector according to claim 3, characterized in that: The competitive breeding operation includes randomly selecting two parent populations O p x1, x2 in the parent population O, and remove x1, x2 from the parent population O P Deleted in; By evaluating the fitness functions of the two individuals, the individual with high fitness is selected to compete and become the winner x w , failed individual x l They will learn from the winning individuals, and both will mutate and enter the offspring population O f , the formula is: x 变异后 =x 变异前 +Δ j Among them, x 变异前 Represents the decision variables of the winner and the loser after learning, and the increased mutation offset Δ is determined according to the mutation probability j Then generate x 变异后 ;μ j represents the random generation probability of a random number, and η represents the parameter of mutation intensity; The x formed after the competition operation w , x ι The decision variables of each position in the , based on the probability premise, are mutated within a given range, and two individuals are selected cyclically to compete until the parent population O P All individuals in the offspring population O f ; The parent population O P With the offspring population O f Merge to generate a merged population Q; A competitive selection operation is performed, where the reference vector of V is associated with all individuals in the population Q, i.e., the reference vectors with the smallest angle between the individual and the reference vector in V, according to the angle between the individual and the reference vector; Traverse all reference vectors associated with individuals, and for each reference vector, select the individual with the smallest APD value from the associated individuals to enter the next generation population O P ; Then traverse the reference vectors that are not associated with individuals. For the reference vectors that are not associated with individuals, select the individual with the smallest angle from the remaining individuals in the population to enter the next generation, where the APD distance is defined as: d t,i,j =(1+P(θ t,i,j ))||f′ t,i || Among them, d t,i,j represents the APD distance metric with penalty term, ||f′ t,i || represents the norm of the current objective function value; P represents the penalty function about θ, which is expressed as: c νt,j =min<ν t,i ,n t,j > Among them, α represents a user-defined parameter that controls the intensity of the time factor’s influence on the penalty, t represents the current number of iterations, and t max represents the maximum number of iterations; γ νt,j represents the normalization factor, γ represents the t , j forms the angle of the other reference vectors with the smallest angle; the angle γ is used to normalize the angle given by ν t , the angle in the subspace specified by j, generates a uniform Pareto-optimal solution set; M represents a constant; θ t,i,j Represents with ν t,i The angle of the other parameter vectors that forms the minimum angle.

5. The water-light complementary optimization method based on competitive reproduction and adaptive vector according to claim 4, characterized in that: The adaptive adjustment mechanism includes the following formula for calculating the next reference vector generation method by calculating the maximum and minimum values ​​of the objective function among individuals in the high-quality population: Among them, V t represents the parameter vector of the tth iteration, r(x) max represents the maximization objective function, f(x) min represents the minimization objective function; The maximization and minimization of the objective function is used to guide the reference vector to change and then guide the population to evolve in the direction of the objective function.

6. The water-light complementarity optimization method based on competitive propagation and adaptive vector according to claim 5, characterized in that: The use of the Pareto optimality criterion to evaluate the current solution set includes repeating a cycle of initializing the production population and the competitive breeding operation until a termination condition is met and the number of fitness evaluations reaches a maximum number.

7. The water-light complementary optimization method based on competitive propagation and adaptive vector according to claim 6, characterized in that: Outputting the optimal solution set that meets the multi-objective optimization requirements includes determining whether the number of fitness evaluations of the population meets a threshold, and outputting the result if the number of solutions required for outputting the optimal solution set of the water-photovoltaic complementary system is met; Otherwise, the competitive breeding operation is repeated until the termination condition is met.

8. A water-light complementary optimization system based on competitive breeding and adaptive vectors using the method according to any one of claims 1 to 7, characterized in that: Objective function construction module: establish optimization objectives, construct multi-objective functions, and establish a comprehensive multi-objective function model; Population initialization module: initializes the production population, each individual in the population represents a potential solution; Reproduction and fitness evaluation module: By performing competitive reproduction operations, the winning individuals are selected for reproduction according to fitness, and the search strategy is dynamically adjusted in combination with the adaptive adjustment mechanism to iteratively improve the solution quality of the entire population; Adaptive reference vector adjustment module: In each iteration, the current solution set is evaluated using the Pareto optimality criterion, and the reference vector is adaptively adjusted based on the evaluation results; Iteration termination and solution set output module: When the iteration reaches the predetermined termination condition, the optimal solution set that meets the multi-objective optimization requirements is output.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images