Current collection circuit optimization method and system, program product and readable storage medium

The collecting line population and iterative optimization strategy are generated through the collecting line optimization method, which solves the problem of difficult-to-find global optimal solution in the wind farm, and achieves more efficient collecting line optimization, taking into account multiple optimization factors.

CN119067051APending Publication Date: 2024-12-03XEMC WINDPOWER CO LTD
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411183192.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The prior art is difficult to find a globally optimal collector line configuration solution in wind farms, especially in complex terrain or large wind farms, resulting in low computing efficiency and insufficient consideration of optimization goals.

Method used

By using the collector line optimization method, by generating the collector line population, each individual represents the optimal number of loops and the optimal collector line in the loop, using iterative processes and diversified individual optimization strategies, including individual movement and neighborhood search, comprehensive Considering the number of loops and line layout, the fitness function is optimized to evaluate multiple factors.

Benefits of technology

The ability to find a global optimal collector line configuration scheme in complex wind farms is improved, the optimization effect and efficiency are improved, the local optimal solution is avoided, and the algorithm exploration ability and convergence speed are enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067051A_ABST
    Figure CN119067051A_ABST
Patent Text Reader

Abstract

The invention provides a current collection line optimization method and system, a program product and a readable storage medium, and relates to the field of line optimization. Firstly, each individual is designed to represent a group of optimal loop number and an optimal current collection line in the loops, so that two key factors of the loop number and the line layout can be considered at the same time in the optimization process, and the problem of a sub-optimal solution possibly caused by step-by-step optimization in related technologies is avoided. Secondly, diversified individual optimization strategies are adopted; and when a better individual is found, generating a copy and moving to the optimal individual. The strategy is helpful for quickly approaching an optimal solution region in a solution space, and the convergence speed of the algorithm is improved. And when a better individual is not found, a random search behavior is realized through neighborhood search, so that the exploration capability of the algorithm is improved, and jumping out of a local optimal solution is facilitated. A current collection line configuration scheme close to global optimum can be found, and the optimization effect and efficiency are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of route optimization, and in particular to a collector line optimization method, system, program product and readable storage medium. Background Art

[0002] With the rapid development of renewable energy, wind power generation has become an important part of the global energy structure transformation. The construction and operation efficiency of wind farms directly affect the economic benefits and sustainable development of the wind power industry. In the process of wind farm construction, the design of the collector line is a key link, which not only affects the project cost, but is also closely related to the reliability and operation efficiency of the system. Therefore, how to optimize the layout of the collector line of the wind farm has become an important technical challenge facing the wind power industry.

[0003] At present, the wind power industry widely uses the minimum spanning tree method to design collector lines. This method first radially classifies the wind turbines in the wind farm with the booster station as the center. Then, the minimum spanning tree algorithm is applied in each classification to generate the shortest path. This method simplifies the calculation process by decomposing the complex wind farm layout problem into several sub-problems, so that a feasible collector line scheme can be quickly obtained in actual engineering. In addition, some researchers have also tried to apply artificial intelligence algorithms to collector line optimization. This method sets the range of the number of loops, performs cyclic solutions from 1 to the number of wind turbines, determines the interval of the number of loops with better results, and then performs further cyclic solutions within the interval. Finally, the number of loops with the best results is selected for multiple solutions in order to obtain a better collector line layout solution.

[0004] However, these existing technologies still have some limitations in practical applications. Although the minimum spanning tree method can find the shortest path in a single classification, the need for pre-classification may cause the global optimal solution to be ignored. Especially in complex terrain or large wind farms, it becomes extremely difficult to predetermine the optimal classification method. Although artificial intelligence algorithms have solved the classification problem to a certain extent, their computational complexity increases sharply with the increase in the scale of wind farms, and they may face the problem of low computational efficiency in large wind farms. In addition, these methods often only consider the single factor of line length in the optimization process, and it is difficult to take into account multiple optimization objectives such as cost, reliability, and electrical performance at the same time, which may lead to unexpected problems in actual operations.

[0005] In summary, the global optimal solution of the collector line cannot be obtained in the related art due to various reasons. Summary of the invention

[0006] The present application provides a collector line optimization method, system, program product and readable storage medium for finding a collector line configuration solution that is close to the global optimum in a complex wind farm environment.

[0007] In a first aspect, the present application provides a collector line optimization method, comprising: generating a collector line population, wherein each individual in the collector line population represents a set of optimal loop numbers and optimal collector lines in the loops; calculating the fitness value of each individual in the collector line population; using the collector line population as a loop variable, iterating according to preset rules until a termination condition is reached; the preset rules are: finding the nearest individual of a current individual, the current individual being any individual in the collector line population; determining whether the nearest individual is a better individual; if the nearest individual is a better individual, generating a copy of the current individual, and moving the copy to the nearest individual according to the individual movement rule; the fitness value of the moved copy is less than that of the current individual If the fitness value is greater than the nearest individual, the current individual is updated to the moved copy and the counter of the current individual is reset; if the nearest individual is not a better individual, a neighborhood search is performed on the current individual; it is determined whether the fitness value of the search result is less than the fitness value of the current individual; if it is less than the fitness value of the current individual, the current individual is updated to the search result and the counter of the current individual is reset; if it is not less than the fitness value of the current individual, the count of the counter is increased; if the count of the counter is greater than the counting threshold and the fitness value of the current individual is less than the fitness value of the global optimal solution, the global optimal solution is updated to the current individual; after the iteration is completed, the global optimal solution is output as the optimal collector line configuration scheme.

[0008] By adopting the above technical solution, firstly, by designing each individual to represent a set of optimal loop numbers and optimal collector lines in the loop, the optimization process can simultaneously consider the two key factors of the number of loops and line layout, avoiding the suboptimal solution problem that may be caused by step-by-step optimization in related technologies. This design enables the algorithm to search in a larger solution space, increasing the possibility of finding the global optimal solution. Secondly, a variety of individual optimization strategies are adopted. When a better individual is found, a copy is generated and moved to the best individual. This strategy helps to quickly approach the optimal solution area in the solution space and improves the convergence speed of the algorithm. When no better individual is found, random search behavior is achieved through neighborhood search, which increases the exploration ability of the algorithm and helps to jump out of the local optimal solution. The solution is continuously improved through an iterative process, including steps such as finding the nearest individual, judging whether it is a better individual, generating and moving copies, and performing neighborhood search. These steps cooperate with each other to ensure the breadth of the search and the improvement of the local optimal solution. The entire optimization process comprehensively considers multiple optimization factors, and can find a collector line configuration solution close to the global optimal in a complex wind farm environment, improving the optimization effect and efficiency.

[0009] In combination with some embodiments of the first aspect, in some embodiments, the individual code of the optimal loop number and the optimal collector line in the loop is: the individual code is a digital sequence, wherein the numbers in the digital sequence represent the numbers of the units or booster stations; the numbers of the booster stations in the individual code are used to separate different collector lines; the digital sequence in the collector line serial number segment represents the connection order of the units on the collector line.

[0010] By adopting the above technical solution, the individual code is designed as a digital sequence, in which the number represents the number of the unit or booster station. This design intuitively reflects the physical layout of the wind farm. Secondly, the number of the booster station is used as a separator for different collector lines. This feature solves the problem of loop classification, so that one code contains the information of multiple loops at the same time. Thirdly, in each collector line serial number segment, the order of numbers indicates the connection order of the unit. This design directly reflects the specific layout of the line. This coding method not only contains all the information on loop classification and unit connection order, but also provides convenience for the subsequent optimization algorithm, thereby improving the efficiency and accuracy of the entire optimization process.

[0011] In combination with some embodiments of the first aspect, in some embodiments, the step of generating a collector line population specifically includes: selecting an initial node in the collector line according to a first roulette selection strategy, the node being a unit or a booster station; the first roulette selection strategy is: generating a random number between 0 and 1, and sequentially subtracting the first selection probability of each node in the set from the random number; the first selection probability is: In the formula, P(x i ) is the node x i The probability of first choice, dist i,b Represents node x i The distance from booster station b; minDist b Indicates the minimum distance between all units and the booster station, dist j,b Represents node x j The distance between the booster station b and the booster station b; c1 is the artificial coefficient. When the difference is not greater than 0, the corresponding node is determined as the next node selected by the roulette wheel; and\or; select the next node of the current node in the collector line according to the second roulette selection strategy; the second roulette selection strategy is: generate a random number between 0 and 1, and use the random number to subtract the second selection probability of each node in the set in turn; the second selection probability is: In the formula, P(x i , x j ) is the node x i Select node x jThe second choice probability, maxDist i Represents all units and node x i The maximum distance between j,i Represents node x j With node x i The distance between; maxDist b Indicates the maximum distance between all units and booster station b; dist j,b Represents node x j The distance from the booster station b; c2 is the artificial coefficient; dist k,i Represents node x k With node x i The distance between k,b Represents node x k The distance between the booster station b and the booster station b; when the difference is not greater than 0, the corresponding node is determined as the next node selected by the roulette wheel.

[0012] By adopting the above technical solution, the first roulette selection strategy is introduced to select the initial node. This strategy gives a higher selection probability to the units farther away from the booster station by calculating the distance between each node and the booster station. This design is conducive to considering the access problem of long-distance units in the population initialization stage, avoiding the defects of the traditional method that may ignore the remote units. Secondly, the second roulette selection strategy is adopted to select the next node of the current node. This strategy considers the distance between the node and the current node and the distance from the booster station. This dual consideration not only ensures the continuity of the collector line, but also takes into account the rationality of the overall layout. Through the coordinated use of these two strategies, the algorithm can generate a more diverse and reasonable initial population, providing a good foundation for the subsequent optimization process.

[0013] In combination with some embodiments of the first aspect, in some embodiments, the individual movement rule is: when the number of the current position in the individual code of the copy is different from the number of the corresponding position in the individual code of the better individual, generate a random number between 0 and 1; when the random number is less than the preset selection probability, the number of the current position in the better individual has not appeared before the current position in the copy, and the number of the current position is not a booster station; replace the number of the current position in the individual code of the copy with the number of the corresponding position in the individual code of the better individual; remove the number of the current position from the unit selection probability vector, and calculate the second roulette selection probability of the current position for each unit in the unit selection probability vector; in the random When the number is less than the preset selection probability, the number of the current position in the better individual has not appeared before the current position in the copy, the number of the current position is not a booster station, the corresponding position of the better individual is a booster station, and the next position in the copy is not a booster station; the number of the current position in the individual code of the copy is replaced with a booster station, and the first roulette selection probability is calculated for each unit in the unit selection probability vector; when the number of the current position in the individual code of the copy is the same as the number of the corresponding position in the individual code of the better individual; the number of the current position is not a booster station, the number of the current position is removed from the unit selection probability vector, and the second roulette selection probability of the current position is calculated for each unit in the unit selection probability vector.

[0014] By adopting the above technical solution, firstly, by comparing the numbers of the current individual and the better individual at the corresponding position, it is decided whether to move, which ensures the targeted nature of the movement. Secondly, by introducing random numbers and preset selection probabilities, the random search behavior in the whale group algorithm is simulated, which increases the exploration ability of the algorithm. During the movement process, the algorithm also considers constraints such as whether the number is repeated and whether it is a booster station, to ensure that the generated new individuals meet the physical characteristics of the actual collector line. In addition, the algorithm dynamically updates the unit selection probability vector and calculates the probability of the first or second roulette selection according to the current state. This design enables each movement to adapt to the current search environment. Through this improved individual movement rule, the particularity of discrete coding can be effectively handled, thereby achieving better performance in the collector line optimization problem.

[0015] In combination with some embodiments of the first aspect, in some embodiments, the step of finding the nearest individual to the current individual specifically includes: calculating the number of characters required to replace the individual code of the current individual to transform it into the individual code of other individuals, the other individuals being individuals other than the current individual in the collector line population; and determining the other individual with the smallest number of characters required to be replaced as the nearest individual.

[0016] By adopting the above technical solution, the coding difference between the current individual and other individuals is calculated, that is, the number of characters that need to be replaced. This method is actually calculating the Hamming distance between individuals, because each whale individual has the same dimension and element composition. By comparing these differences, the algorithm can quickly determine the individual that needs to replace the least characters, that is, the closest individual. This method is not only simple and efficient in calculation, but also can accurately reflect the similarity between individuals.

[0017] In combination with some embodiments of the first aspect, in some embodiments, the neighborhood search is: generate a random number between 0 and 1; subtract the selection probabilities of different neighborhood search strategies in turn, and the selection probability is the ratio of the historical update and better number of neighborhood search strategies to the total historical update and better number of strategies; when the random number is less than or equal to 0, execute the corresponding neighborhood search strategy.

[0018] By adopting the above technical solution, a random number between 0 and 1 is generated, which introduces randomness to the strategy selection and avoids falling into the local optimum. Then, the algorithm subtracts the selection probabilities of different neighborhood search strategies in turn, which are dynamically calculated based on the historical update success rate. Specifically, the selection probability reflects the effectiveness of each strategy in the past optimization process. In this way, the algorithm can automatically tend to choose a more effective search strategy while still retaining the possibility of trying other strategies. This adaptive mechanism not only enhances the local search ability of the algorithm, but also improves the convergence speed.

[0019] In conjunction with some embodiments of the first aspect, in some embodiments, the fitness value is: minf(x)=K·A(x)+B·P(x)+K·C(x) In the formula, K is the conversion coefficient between present value and equivalent annual value; A(x) is the total line investment corresponding to individual x; B is the electricity price; P(x) is the annual line loss corresponding to individual x; C(x) is the switch cabinet price corresponding to individual x; Where A(x) is the total line investment corresponding to individual x, m represents the number of collector line loops; A i represents the unit investment of the ith line; l i represents the total length of the i-th line; Where P(x) is the annual line loss corresponding to individual x, l i represents the total length of the ith line, h is the annual operating hours; R i represents the resistance of the ith line; C(x)=m×c In the formula, m is the number of switch cabinets, and c is the price of a single switch cabinet.

[0020] By adopting the above technical solutions, the algorithm constructs a comprehensive and reasonable fitness evaluation system. First, the fitness function comprehensively considers three key factors: total line investment, annual loss cost, and switchgear price, which ensures the comprehensiveness of the evaluation.

[0021] In a second aspect, the present application provides a collector line optimization system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the collector line optimization system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0022] In a third aspect, the present application provides a computer program product comprising instructions, which, when executed on a collector line optimization system, enables the collector line optimization system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0023] In a fourth aspect, the present application provides a computer-readable storage medium comprising instructions, which, when executed on a collector line optimization system, enables the collector line optimization system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: First, by designing each individual to represent a set of optimal loop numbers and optimal collector lines in the loop, the optimization process can simultaneously consider the two key factors of loop number and line layout, avoiding the suboptimal solution problem that may be caused by step-by-step optimization in related technologies. This design enables the algorithm to search in a larger solution space, increasing the possibility of finding the global optimal solution. Secondly, a variety of individual optimization strategies are adopted. When a better individual is found, a copy is generated and moved to the best individual. This strategy helps to quickly approach the optimal solution area in the solution space and improves the convergence speed of the algorithm. When no better individual is found, random search behavior is achieved through neighborhood search, which increases the exploration ability of the algorithm and helps to jump out of the local optimal solution. The solution is continuously improved through an iterative process, including finding the nearest individual, judging whether it is a better individual, generating copies and moving, and performing neighborhood search. These steps work together to ensure the breadth of the search and the improvement of the local optimal solution. The entire optimization process comprehensively considers multiple optimization factors, and can find a collector line configuration solution close to the global optimal in a complex wind farm environment, improving the optimization effect and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of the collector line optimization method in the embodiment of the present application; Figure 2 is a schematic diagram of two individual whales in an embodiment of the present application; Figure 3 This is another partial flow chart of the method for optimizing the current collection line in an embodiment of the present application; Figure 4 This is another partial flow chart of the method for optimizing the current collection line in the embodiment of the present application; Figure 5 is another flow chart of the collector line optimization method in the embodiment of the present application; Figure 6 It is a flow chart of individual movement rules in the collector line optimization method in the embodiment of the present application; Figure 7 It is a flowchart of a specific selection strategy in the collector line optimization method in the embodiment of the present application; Figure 8 This is a schematic diagram of a neighborhood search strategy for the collector line optimization method in an embodiment of the present application; Fig. 9 Schematic diagram of another neighborhood search strategy of the collector line optimization method in an embodiment of the present application; Fig.10 Schematic diagram of another neighborhood search strategy of the collector line optimization method in an embodiment of the present application; Fig.11 It is an exemplary hardware structure diagram of the collection line optimization system in the embodiment of the present application. DETAILED DESCRIPTION

[0026] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more listed items.

[0027] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0028] See also Figure 1 , Figure 1 It is a flow chart of the collector line optimization method in the embodiment of the present application.

[0029] S101, generating a collector line population, wherein each individual in the collector line population represents a set of optimal loop numbers and optimal collector lines in the loops; The collector line population represents a set of potential solutions, an individual refers to a single solution in the population, and the optimal number of loops is used to represent the optimal number of loops in the collector line system. The optimal collector line in the loop refers to the most optimally configured collector line in each loop. For example, in wind farm design, an individual may represent a specific collector line layout scheme, including the number of loops and the specific line configuration of each loop.

[0030] In some embodiments, the population of collector lines represents a pod of whales and the individuals are whales.

[0031] In some embodiments, the generation of the collector line population can be achieved in a variety of ways: Optionally, a random generation method can be used: 1) determine the population size N; 2) for each individual, randomly generate the number of loops; 3) for each loop, randomly generate the collector line configuration; 4) repeat steps 2-3 until N individuals are generated; Optionally, a heuristic generation method can be used: 1) analyze the problem constraints and optimization goals; 2) design heuristic rules, such as the shortest path principle; 3) generate the initial population according to the heuristic rules. It is understandable that other methods can also be used to achieve the generation of the collector line population, which is not limited here.

[0032] It should be noted that in actual use, it is necessary to solve the optimal number of loops and the optimal collector line design scheme in the loop in one coding. Therefore, the individual coding needs to reflect the classification of collector line loops and the connection order of wind turbines in each collector line.

[0033] In some preferred embodiments, the individual code is a digital sequence, wherein the numbers in the digital sequence represent the numbers of the units or booster stations; the numbers of the booster stations in the individual code are used to separate different collector lines; and the digital sequence in the collector line serial number segment represents the connection order of the units on the collector line.

[0034] It should be noted that the code length is equal to twice the number of units minus 1.

[0035] In an exemplary embodiment, the coding example is: x=(6,1,7,2,3,5,7,4,7,7,7) in: 7 indicates a boost station (separator); 1-6 are wind turbine unit numbers; This code indicates that the wind farm contains 3 collector lines: Unit 6->Unit 1->Booster Station; Unit 2->Unit 3->Unit 5->Booster Station; Unit 4->Booster Station; It can be seen that the individual code is designed as a digital sequence, where the number represents the number of the unit or booster station. This design intuitively reflects the physical layout of the wind farm. Secondly, the number of the booster station is used as a separator for different collector lines. This feature solves the problem of loop classification, so that one code contains the information of multiple loops at the same time. Thirdly, in each collector line serial number segment, the order of numbers indicates the connection order of the unit. This design directly reflects the specific layout of the line. This coding method not only contains all the information of loop classification and unit connection order, but also provides convenience for the subsequent optimization algorithm, thereby improving the efficiency and accuracy of the entire optimization process.

[0036] It should be noted that, in the actual use process, the optimization ability is closely related to the quality of the initial population: if there are individuals in the initial population that are close to the global optimal solution, the algorithm is likely to converge to the global optimal solution; if the initial population is far from the global optimal solution, it is difficult for the algorithm to find the global optimal solution. Therefore, the initialization of the population is an important part of applying metaheuristic algorithms to solve optimization problems.

[0037] Therefore, in some preferred embodiments, the design of wind farm collector lines is an engineering optimization design problem with a certain timeliness. The sooner the optimal or near-optimal solution is obtained, the better. This application proposes two roulette selection strategies to generate initial individuals, which can obtain a better initial solution and help improve the convergence speed. Thus, the solution efficiency is significantly improved.

[0038] In some preferred embodiments, following the above example, the first roulette selection strategy is to select the initial node of a collector line so that the unit farther from the booster station has a greater selection probability, which is more in line with most actual situations; the initial node in the collector line is selected according to the first roulette selection strategy, and the node is a unit or a booster station; The first roulette selection strategy is: Generate a random number between 0 and 1, Subtract the first selection probability of each node in the set from the random number in turn; The first choice probability is: In the formula, P(x i ) is the node x i The probability of first choice, dist i,b Represents node xi The distance from booster station b; minDist b Indicates the minimum distance between all units and the booster station, dist j,b Represents node x j The distance from the booster station b; c1 is the artificial coefficient; When the difference is not greater than 0, the corresponding node is determined as the next node selected by the roulette wheel.

[0039] In some preferred embodiments, the second roulette selection strategy is to make the units closer to the booster station and closer to the current node have a greater probability of selection when selecting the next node of a node in a collector line; Selecting the next node of the current node in the collector line according to the second roulette selection strategy; The second roulette selection strategy is: Generate a random number between 0 and 1, Subtract the second selection probability of each node in the set from the random number in turn; The second choice probability is: In the formula, P(x i , x j ) is the node x i Select node x j The second choice probability, maxDist i Represents all units and node x i The maximum distance between j,i Represents node x j With node x i The distance between; maxDist b Indicates the maximum distance between all units and booster station b; dist j,b Represents node x j The distance from the booster station b; c2 is the artificial coefficient; dist k,i Represents node x k With node x i The distance between k,b Represents node x k The distance from booster station b; When the difference is not greater than 0, the corresponding node is determined as the next node selected by the roulette wheel.

[0040] It should be noted that the first roulette wheel selection strategy and the second roulette wheel selection strategy may be used simultaneously or separately, and are not limited here.

[0041] It can be seen that the first roulette selection strategy is introduced to select the initial node. This strategy gives a higher selection probability to the units far away from the booster station by calculating the distance between each node and the booster station. This design is conducive to considering the access problem of long-distance units in the population initialization stage, avoiding the defect of traditional methods that may ignore the remote units. Secondly, the second roulette selection strategy is used to select the next node of the current node. This strategy considers the distance between the node and the current node and the distance to the booster station. This dual consideration not only ensures the continuity of the collector line, but also takes into account the rationality of the overall layout. Through the coordinated use of these two strategies, the algorithm can generate a more diverse and reasonable initial population, providing a good foundation for the subsequent optimization process.

[0042] S102, calculating the fitness value of each individual in the collector line population; The fitness value is a numerical indicator used to evaluate the quality of an individual. For example, in wind farm design, the fitness value may represent the total cost or efficiency of a collector line solution.

[0043] In some embodiments, each individual in the population is evaluated and its fitness value is calculated. The calculation of the fitness value is usually based on a predefined objective function and may consider multiple factors, such as economic cost, transmission efficiency, reliability, etc. This process provides a quantitative evaluation standard for subsequent selection and optimization operations, allowing the algorithm to distinguish the pros and cons of different solutions.

[0044] In some specific embodiments, the fitness value is minf(x)=K·A(x)+B·P(x)+K·C(x) In the formula, K is the conversion coefficient between present value and equivalent annual value; A(x) is the total line investment corresponding to individual x; B is the electricity price; P(x) is the annual line loss corresponding to individual x; C(x) is the switch cabinet price corresponding to individual x; Where A(x) is the total line investment corresponding to individual x, m represents the number of collector line loops; A i represents the i-th line; l i represents the total length of the i-th line; In the formula, r is the fund discount rate; N is the amortization period.

[0045] In the formula, A i represents the i-th line, S i represents the cross-sectional area of ​​the ith line; In the formula, S irepresents the cross-sectional area of ​​the i-th line, J represents the economic current density; I i represents the calculated working current of the ith circuit; In the formula, I i represents the calculated working current of the ith circuit, n i represents the number of wind turbines in the i-th collector line; P i,j It represents the rated power of the j-th wind turbine in the ith collector line; α is the power factor; U is the voltage level of the collector line.

[0046] In the formula, l i represents the total length of the i-th line, n represents the number of wind turbines in the field; (x0, y0) represents the coordinates of the inlet point of the booster station; (x i ,y i ) represents the coordinates of the i-th wind turbine.

[0047] Where P(x) is the annual line loss corresponding to individual x, l i represents the total length of the ith line, h is the annual operating hours; R i represents the resistance of the i-th line; Ri=ri×li In the formula, R i represents the resistance of the ith line, r i represents the resistance per kilometer of the line, l i represents the total length of the i-th line; In the formula, r i Represents the resistance per kilometer of the line, S i represents the cross-sectional area of ​​the ith line.

[0048] C(x)=m×c In the formula, m is the number of switch cabinets, and c is the price of a single switch cabinet.

[0049] It can be seen that the algorithm constructs a comprehensive and reasonable fitness evaluation system. First, the fitness function comprehensively considers three key factors: total line investment, annual loss cost, and switchgear price, which ensures the comprehensiveness of the evaluation.

[0050] S103, taking the collector line population as a loop variable and iterating according to a preset rule until a termination condition is reached; the termination condition is reaching a maximum number of iterations or finding a solution that meets a specific standard.

[0051] In some embodiments, the iteration process can be implemented in a variety of ways: Optionally, a fixed number of iterations method can be used: 1) set the maximum number of iterations N; 2) apply preset rules to the population for update; 3) add 1 to the number of iterations; 4) if the number of iterations is less than N, return to step 2, otherwise end; Optionally, a fitness convergence method can be used: 1) set a convergence threshold ε; 2) apply preset rules to the population for update; 3) calculate the difference between the current optimal fitness and the previous generation's optimal fitness; 4) if the difference is greater than ε, return to step 2, otherwise end. It is understandable that other methods can also be used to implement the iteration process, which are not limited here.

[0052] S104, the preset rule is to find the nearest individual of the current individual, the current individual being any individual in the collector line population; It should be noted that the preset rule is step S104 to step S112, rather than just step S104.

[0053] In some embodiments, the search for the nearest individual can be achieved in a variety of ways: Optionally, the coding distance method can be used: 1) define a coding distance metric, such as the Hamming distance; 2) calculate the distance between the current individual and each other individual in the population; 3) select the individual with the smallest distance as the nearest individual; Optionally, the feature space distance method can be used: 1) map the individual to the feature space; 2) calculate the Euclidean distance between the current individual and other individuals in the feature space; 3) select the individual with the smallest distance as the nearest individual. It is understandable that other methods can also be used to achieve the search for the nearest individual, which are not limited here.

[0054] In some preferred embodiments, following the above example; Calculate the number of characters required to replace the individual code of the current individual to transform it into the individual code of other individuals, where the other individuals are other individuals in the collector line population except the current individual; The other individual with the smallest number of characters to be replaced is determined as the nearest individual.

[0055] refer to Figure 2 , Figure 2 Schematic diagram of two whale individuals in the embodiment of the present application. Specifically, each whale individual has the same dimension and the element composition of each whale individual is the same; based on these characteristics, the Hamming distance is used to calculate the distance between different whale individuals.

[0056] It can be seen that the difference in encoding between the current individual and other individuals is the number of characters that need to be replaced. This method is actually calculating the Hamming distance between individuals, because each whale individual has the same dimension and element composition. By comparing these differences, the algorithm can quickly determine the individual that needs to replace the least characters, that is, the closest individual. This method is not only simple and efficient in calculation, but also can accurately reflect the similarity between individuals.

[0057] S105, judging whether the nearest individual is a better individual; A better individual is an individual whose fitness value is less than the fitness value of the current individual.

[0058] S106, if the nearest individual is a better individual, generate a copy of the current individual, and move the copy to the nearest individual according to the individual movement rule; A replica of the current individual is an exact copy of the current individual. The individual movement rule is used to guide how to adjust the replica to get closer to the nearest individual. For example, the movement rule may involve making partial modifications in the encoding space in the direction of the nearest individual.

[0059] It should be noted that this process is intended to explore the solution space between the current individual and the better individual, and to find a possible better solution. By generating a copy instead of directly modifying the current individual, the algorithm retains the original solution, increasing the safety of exploration. Therefore, in other embodiments, if the better solution is like the original solution, the original solution is retained, otherwise, the original solution is deleted.

[0060] It should be noted that the individual movement rule can be, using linear interpolation: 1) calculate the difference vector between the current individual and the nearest individual in the encoding space; 2) select an interpolation coefficient α (0 < α < 1); 3) multiply the difference vector by α to obtain the movement vector; 4) add the movement vector to the copy of the current individual; or, the feature borrowing method can be used: 1) identify the dominant features in the nearest individual; 2) replace the corresponding features in the copy of the current individual with a certain probability; 3) make local adjustments to the replaced features, which are not limited here.

[0061] It should be noted that in the subsequent embodiments, preferred individual movement rules are provided, which will not be described in detail here.

[0062] S107, when the fitness value of the moved copy is less than the fitness value of the current individual, the current individual is updated to the moved copy, and the counter of the current individual is reset; The moved copy refers to the current individual copy after the individual movement rule is adjusted.

[0063] Specifically, this step first compares the fitness values ​​of the moved replica and the original current individual. If the replica performs better, the current individual is replaced with the replica, and the counter associated with the individual is reset to zero. This process updates the population, retains the better solution, and provides the individual with new improvement opportunities by resetting the counter. This mechanism not only ensures the improvement of population quality, but also maintains the dynamics and adaptability of the algorithm.

[0064] It should be noted that the existence of the counter accepts inferior solutions to a certain extent and avoids falling into local optimality, which increases the ability of the algorithm to jump out of local optimality.

[0065] In this embodiment (not specifically referring to step S107), the counter mechanism uses two triggering modes to control the algorithm's search process: Other opportunities based on current solution: When the algorithm explores around the current solution, a counter will record the number of attempts. If no better solution is found within a certain number of attempts, a jump-out mechanism will be triggered, prompting the algorithm to move to a new search area. This ensures that the algorithm does not linger too much near the local optimal solution.

[0066] Exceeding the preset threshold: At the same time, the counter also accumulates the number of iterations of the overall search. Once this count exceeds the pre-set threshold, the update and exit operations will be triggered regardless of the quality of the current solution. This mechanism ensures the global search capability of the algorithm and prevents it from staying in a certain solution space area for a long time.

[0067] S108. If the nearest individual is not a better individual, perform a neighborhood search on the current individual; Neighborhood search is used to refer to the process of finding a better solution in the local space of the current solution. For example, a neighborhood search may involve making small random modifications to the current individual.

[0068] Neighborhood search usually involves making small modifications or perturbations to the current solution to find possible improvements. This process aims to find a local optimal solution near the current solution, especially when global search cannot provide improvements. Through neighborhood search, the algorithm can make better use of the information of the current solution and improve the efficiency and accuracy of the search.

[0069] In some embodiments, neighborhood search can be implemented in a variety of ways: Optionally, a random perturbation method can be used: 1) define the perturbation range and probability; 2) for each coding bit of the current individual, randomly perturb with a certain probability; 3) generate multiple candidate solutions after perturbation; 4) evaluate the fitness of these candidate solutions; Optionally, a systematic neighborhood search method can be used: 1) define the neighborhood structure; 2) systematically generate all neighborhood solutions of the current solution; 3) evaluate the fitness of all neighborhood solutions; 4) select the optimal neighborhood solution. It is understandable that other methods can also be used to implement neighborhood search, such as variable neighborhood search or adaptive neighborhood structure, which are not limited here.

[0070] It should be noted that in the subsequent embodiments, preferred neighborhood search rules are provided, which will not be described in detail here.

[0071] S109, determining whether the fitness value of the search result is less than the fitness value of the current individual; S110, if it is less than the fitness value of the current individual, updating the current individual as the search result, and resetting the counter of the current individual; S111, if it is not less than the fitness value of the current individual, then increase the count of the counter; This step increments the counter associated with the current individual by 1. This process keeps track of the number of times the individual has failed to improve in a row. The increase in the counter reflects the difficulty the current individual is experiencing in the local search, and may indicate that the individual is close to a local optimum. By tracking this count, the algorithm can decide in subsequent steps whether it needs to adopt a more aggressive search strategy or re-evaluate the current individual's position.

[0072] S112: when the count of the counter is greater than the count threshold and the fitness value of the current individual is less than the fitness value of the global optimal solution, updating the global optimal solution to the current individual; This step is performed at the end of each iteration. Specifically, this step first checks whether the counter of the current individual exceeds the preset threshold, which indicates that the individual has not been improved for a long time. Then, the fitness value of the current individual is compared with the global optimal solution. If the current individual performs better, it will be updated to the new global optimal solution even though it has stagnated. This mechanism allows the algorithm to update the global optimal solution after a long search, which increases the possibility of the algorithm finding a better solution, especially in complex multi-modal problems.

[0073] S113. After the iteration is completed, the global optimal solution is output as the optimal collector line configuration solution.

[0074] It can be seen that, firstly, by designing each individual to represent a set of optimal loop numbers and optimal collector lines in the loop, the optimization process can simultaneously consider the two key factors of the number of loops and line layout, avoiding the suboptimal solution problem that may be caused by step-by-step optimization in related technologies. This design enables the algorithm to search in a larger solution space, increasing the possibility of finding the global optimal solution. Secondly, a variety of individual optimization strategies are adopted. When a better individual is found, a copy is generated and moved to the best individual. This strategy helps to quickly approach the optimal solution area in the solution space and improves the convergence speed of the algorithm. When no better individual is found, random search behavior is achieved through neighborhood search, which increases the exploration ability of the algorithm and helps to jump out of the local optimal solution. The solution is continuously improved through an iterative process, including steps such as finding the nearest individual, judging whether it is a better individual, generating copies and moving, and performing neighborhood search. These steps cooperate with each other to ensure the breadth of the search and the improvement of the local optimal solution. The entire optimization process comprehensively considers multiple optimization factors, and can find a collector line configuration solution close to the global optimal in a complex wind farm environment, improving the optimization effect and efficiency.

[0075] It should be noted that the above embodiment is only a partial description of a complete embodiment, and the following embodiment will fully describe the collector line optimization method; refer to Figure 3 , Figure 3 ; Figure 3 This is another partial flow chart of the method for optimizing the current collection line in an embodiment of the present application; Figure 4 This is another partial flow chart of the method for optimizing the current collection line in the embodiment of the present application; It should be noted that some steps of this embodiment have been described in detail in the above embodiments, so the explanation of the steps in the above embodiments may be referred to and are not limited here.

[0076] S31, initialization parameters; It should be noted that the initialization parameters here refer to setting appropriate initial conditions for the algorithm to ensure that the algorithm can run correctly and converge to a valid solution.

[0077] S32, initializing a group of whales, wherein each whale in the group of whales represents a set of optimal number of loops and optimal collector lines in the loops; It should be noted that in order to create a diverse set of initial solutions, a good starting point is provided for subsequent optimization. This helps the algorithm explore a wider solution space and increases the possibility of finding the global optimal solution; S33, calculating the fitness value of each whale; S34, traverse each whale in the whale group; It should be noted that ensuring that each solution has a chance to be optimized maintains the diversity of the population. This helps the algorithm fully explore the solution space and avoid premature convergence; S341, finding a better and closest whale to the current whale; It should be noted that the current solution is guided to move towards a better solution while considering the distance between solutions. This helps to balance local search and global search and improve the efficiency of the algorithm; If a better and closest whale is found, execute step S342; If no better and closest whale is found, execute step S343; S342. If a better and closest whale is found; It should be noted that the algorithm is allowed to perform a guided search to move closer to known good solutions. This helps to speed up the convergence process and quickly improve the quality of the solution; S3421. Generate a copy of the current whale; S3422. According to the individual movement rule, the replica moves to the better and closest whale; S3423. Calculate the fitness value of the replica, and determine whether the fitness value of the replica is less than the fitness value of the current whale; If the fitness value of the replica is less than the fitness value of the current whale, execute step S3424; If the fitness value of the replica is greater than or equal to the fitness value of the current whale, execute step S3425; It should be noted that better solutions are accepted and the quality of solutions is continuously improved. This ensures that the algorithm always moves towards a better solution; S3424, if the fitness value of the replica is less than the fitness value of the current whale; S34241, replace the current whale with a copy; S34242, reset the current whale's counter to 0; Return to step S24; S3425. If the fitness value of the replica is greater than or equal to the fitness value of the current whale, it is determined whether the counter of the current whale is less than the counting threshold; Whether the current whale counter is less than the counting threshold, execute step S34251; Whether the current whale counter is less than the counting threshold, execute step S34252; It should be noted that the algorithm is allowed to accept inferior solutions to a certain extent to avoid falling into the local optimum, which increases the ability of the algorithm to jump out of the local optimum.

[0078] S34251. If the current whale's counter is less than the counting threshold; S342511, the current whale's counter is increased by 1; Return to step S24; S34252: If the current whale's counter is greater than or equal to the counting threshold; S342521, reinitialize the current whale; S342522. Calculate the fitness value of the new current whale; Return to step S24; S343. If no better and closest whale is found; It should be noted that when no better solution can be found, the algorithm is allowed to perform local search. This helps to explore around the current solution and may find new high-quality solutions; S3431, generate a copy of the current whale; S3432, performing a neighborhood search on the replica to obtain a new replica, and determining whether the fitness value of the new replica is less than the fitness value of the current whale; Whether the fitness value of the new copy is less than the fitness value of the current whale, execute step S3433; Whether the fitness value of the new copy is greater than or equal to the fitness value of the current whale, step S3434 is executed; S3433. If the fitness value of the new replica is less than the fitness value of the current whale; S34331, replace the current whale with a new copy; S34332, reset the current whale's counter to 0; Return to step S24; S3434: If the fitness value of the new replica is greater than or equal to the fitness value of the current whale, and it is determined that the counter of the current whale is less than the counting threshold; If the current whale counter is less than the counting threshold, execute step S34341; If the current whale's counter is greater than or equal to the counting threshold, execute step S34342; It should be noted that when no improvement is achieved for a long time, consider updating the global optimal solution or reinitializing. This helps the algorithm maintain a balance between global and local searches; S34341. If the current whale's counter is less than the counting threshold; S343411, the current whale's counter is increased by 1; It should be noted that the current solution is allowed to continue to try to improve within a certain number of times. This gives each solution enough opportunities to prove its potential; Return to step S24; S34342. If the counter of the current whale is greater than or equal to the counting threshold, determine whether the fitness value of the current whale is less than the fitness value of the updated global optimal individual; If the fitness value of the current whale is less than the fitness value of the updated global optimal individual, execute step S343421; If the fitness value of the current whale is greater than or equal to the fitness value of the updated global optimal individual, execute step S343422; S343421. If the current whale's fitness value is less than the fitness value of the updated global optimal individual; It should be noted that the global optimal solution is updated and the best solution found by the algorithm is saved. This ensures that the algorithm can record and output the optimal solution found during the entire search process. S3434211. Update the global optimal solution to the fitness value of the current whale; S3434212. Update the global optimal individual to the current whale; Return to step S24; S343422. If the fitness value of the current whale is greater than or equal to the fitness value of the updated global optimal individual, reinitialize the current whale; It should be noted that when the global optimal solution cannot be improved, the current solution is reinitialized. This increases the chances of the algorithm exploring new solution space areas and helps to escape from the local optimum; S343423. Calculate the fitness value of the new current whale; Return to step S24; S35. Check whether there is a whale in the last generation of population that is better than the global optimal individual. If so, update the global optimal individual. It should be noted that the output is the optimal solution found during the entire search process. This ensures that the final output of the algorithm is the best possible result. It should be noted that S35 should be started after the entire iteration is completed, so it can be executed after step S32. Of course, since the result is the output global optimal individual, it can also be executed after step S3434212, which is not limited here.

[0079] S36. After the iteration is completed, the global optimal individual is output as the optimal collector line configuration solution.

[0080] refer to Figure 5 , Figure 5 is another flow chart of the collector line optimization method in the embodiment of the present application; In some preferred embodiments, the individual movement rules are: S501, generating a random number between 0 and 1 when the number of the current position in the individual code of the copy is different from the number of the corresponding position in the individual code of the better individual; Continuing from the above example, wind turbines or booster stations are represented by numbers; S502: When the random number is less than the preset selection probability, the number of the current position in the better individual has not appeared before the current position in the copy, and the number of the current position is not a booster station; Specifically, the algorithm checks three conditions: (1) whether the generated random number is less than the preset selection probability, (2) whether the number of the current position in the better individual has not appeared in the copy before, and (3) whether the number of the current position is not a booster station. These three conditions jointly determine whether to update. This design takes into account both randomness (through random numbers) and the validity of the solution (avoiding repeated numbers and improper replacement of booster stations).

[0081] S503, replacing the number of the current position in the individual code of the copy with the number of the corresponding position in the individual code of the better individual; S504, removing the number of the current position from the unit selection probability vector, and calculating the second roulette selection probability of the current position for each unit in the unit selection probability vector; The unit selection probability vector includes the probability that all units and booster stations select the next unit and booster station respectively.

[0082] Specifically, the algorithm first removes the number at the current position from the unit selection probability vector, because the number has been used and should not be selected again. Then, the algorithm recalculates the second roulette selection probability of the remaining units. This process ensures that when selecting the next node, the probability distribution will be dynamically adjusted according to the current selection situation, thereby improving the adaptability and efficiency of the algorithm.

[0083] S505. When the random number is less than the preset selection probability, the number of the current position in the better individual has not appeared before the current position in the copy, the number of the current position is not a booster station, the position corresponding to the better individual is a booster station, and the next position in the copy is not a booster station; replace the number of the current position in the individual code of the copy with a booster station, and calculate the first roulette selection probability for each unit in the unit selection probability vector; Specifically, the algorithm first checks multiple conditions: the random number is less than the preset selection probability, the number of the current position in the better individual has not appeared before in the replica, the current position is not a booster station, the corresponding position of the better individual is a booster station, and the next position in the replica is not a booster station. When all these conditions are met, the algorithm changes the number of the current position of the replica to a booster station and recalculates the first roulette selection probability of each unit in the unit selection probability vector. This step allows the algorithm to insert booster stations at appropriate locations while maintaining the validity of the solution.

[0084] S506: When the number of the current position in the individual code of the copy is the same as the number of the corresponding position in the individual code of the better individual; S507: If the number of the current position is not a booster station, remove the number of the current position from the unit selection probability vector, and calculate the second roulette selection probability of the current position for each unit in the unit selection probability vector.

[0085] The principle and process of this step are the same as those of step S504. For relevant details, please refer to step S504 and no limitation is made here.

[0086] It can be seen that, firstly, by comparing the numbers of the current individual and the better individual at the corresponding position, it is decided whether to move, which ensures the targeted nature of the movement. Secondly, the introduction of random numbers and preset selection probabilities simulates the random search behavior in the whale group algorithm and increases the exploration ability of the algorithm. During the movement process, the algorithm also considers constraints such as whether the number is repeated and whether it is a booster station to ensure that the generated new individuals meet the physical characteristics of the actual collector line. In addition, the algorithm dynamically updates the unit selection probability vector and calculates the probability of the first or second roulette selection according to the current state. This design enables each movement to adapt to the current search environment. Through this improved individual movement rule, the particularity of discrete coding can be effectively handled, thereby achieving better performance in the collector line optimization problem.

[0087] It should be noted that the above embodiment is only a partial description of a complete embodiment, and the following embodiment will fully describe the collector line optimization method; refer to Figure 6 It is a flow chart of individual movement rules in the collector line optimization method in the embodiment of the present application; S6.1 Input: copy X of whale individual Ωu, "better and closest" whale individual Y of whale Ωu, crew selection probability vector S.

[0088] S6.2 calculates the first roulette wheel selection probability for each unit in the probability vector S; S6.3 Main loop; S6.3.1 For position i=1 to whale copy length |X|, perform a loop: S6.3.2 Determine the length of the probability vector S: If the probability vector length |S| == 1, execute step 6.3.2.1; Otherwise, proceed to step 6.3.3.

[0089] S6.3.2.1 Whale copy element Xi = Whale copy length |X| + 1; S6.3.3 Compare the whale copy element Xi with the better whale element Yi: If the whale copy element Xi ≠ the better whale element Yi, then execute step 6.3.3.1; Otherwise, proceed to step 6.3.3.2.

[0090] S6.3.3.1 If the random number r < probability threshold λ, execute step 6.3.3.1.1; Otherwise, proceed to step 6.3.3.1.2.

[0091] S6.3.3.1.1 Performing probabilistic moves: S6.3.3.1.1.1 Determine the effectiveness of the better whale element Yi: If a better whale element Yi exists and Yi≠whale copy length |X|+1, perform the following steps: Temporarily store the value of the whale copy element Xi; Whale copy element Xi = better whale element Yi; Remove the better whale element Yi from the probability vector S; Calculate the second roulette selection probability of the whale replica element Xi for each unit in the probability vector S; Otherwise, proceed to step 6.3.3.1.1.2.

[0092] S6.3.3.1.1.2 Handling of special situations: If the better whale element Yi == whale copy length |X|+1, then perform the following judgment: If the whale copy element Xi-1 ≠ whale copy length |X|+1, then: Whale copy element Xi = whale copy length |X| + 1; Calculate the first roulette wheel selection probability for each unit in the probability vector S; Otherwise: execute a specific selection strategy; If the above conditions are not met, a specific selection strategy is implemented; S6.3.3.1.2 Untriggered probabilistic move: Execute a specific selection strategy; S6.3.3.2 When Xi and Yi are equal: If the whale replica element Xi ≠ whale replica length |X|+1, then perform the following steps: Remove the whale copy element Xi from the probability vector S; Calculate the second roulette selection probability of the whale replica element Xi for each unit in the probability vector S; otherwise, if the whale replica element Xi-1 == the whale replica length |X|+1, then execute the specific selection strategy; S6.4 Results are returned; S6.4.1 returns the moved whale copy X; Figure 7It is a flowchart of a specific selection strategy in the collector line optimization method in the embodiment of the present application; In some preferred embodiments, the specific selection strategy is: S7.1. Input: moving whale replica X, crew selection probability vector S, element number j in X that executes a specific selection strategy.

[0093] S7.2, set the flag bit flag to true; S7.3, when the flag is true, execute the following loop: S7.3.2. Set the internal flag bit _flag to false S7.3.3. Generate a random number r between 0 and 1 S7.3.4. For each element in the crew selection probability vector S (from 1 to the length of S), execute a loop: S7.3.4.1. Subtract the selection probability P(Si) of the current unit from the random number r S7.3.4.2 If the updated random number r is less than or equal to 0, then: S7.3.4.2.1 If the current crew number is equal to the whale copy length plus 1, then: If the element before j in the whale copy is equal to the length of the whale copy plus 1, set the internal flag bit _flag to true; S7.3.4.2.2 Record the current unit number as id; S7.3.4.2.3 Break out of loop; S7.3.5 If after completing the cycle the random number r is still greater than 0, then: S7.3.5.1 If the element before j in the whale copy is equal to the length of the whale copy plus 1, and the probability of selection of the backup unit is not 0, then set the internal flag bit _flag to true; S7.3.5.2 Otherwise: If the probability of selecting the standby unit is not 0, set id to the length of S minus 1; Otherwise, set id to the length of S minus 2; S7.3.6 If the internal flag bit _flag is not true, set the external flag bit flag to false; S7.4 Select execution; S7.4.1 If the selected unit number is equal to the jth element in the whale copy, then: S7.4.1.1 If the length of the whale copy plus 1 is equal to the jth element in the whale copy, then: Remove the selected unit from the unit selection probability vector S; Calculate the second roulette selection probability of the jth element in the whale copy for each unit in S; S7.4.1.2 Otherwise: Calculate the first roulette wheel selection probability for each unit in S; S7.4.2 Otherwise: S7.4.2.1 If the whale copy length plus 1 is not equal to the selected unit number, then: The value of X equals the value of the element at position i in Y, which is changed to the value of the jth element in the whale copy. Update the jth element in the whale copy to the selected unit number; Remove the selected unit from the unit selection probability vector S; Calculate the second roulette selection probability of the jth element in the whale copy for each unit in S; S7.4.2.2 Otherwise: Update the jth element in the whale copy to the selected unit number; Calculate the first roulette wheel selection probability for each unit in S; S7.5 result return; S7.5.1 Return the moved whale copy X; In some preferred embodiments, the neighborhood search is: Generate a random number between 0 and 1; Subtract the selection probability of different neighborhood search strategies in turn. The selection probability is the ratio of the number of historical updates of the neighborhood search strategy to the total number of historical updates. When the random number is less than or equal to 0, the corresponding neighborhood search strategy is executed.

[0094] refer to Figure 8 , Figure 8 It is a schematic diagram of a neighborhood search strategy of the collector line optimization method in an embodiment of the present application; the first neighborhood search strategy is: randomly select two elements in the solution and exchange their positions.

[0095] refer to Fig. 9 , Fig. 9 It is a schematic diagram of another neighborhood search strategy of the collector line optimization method in the embodiment of the present application; the second neighborhood search strategy is: randomly select two elements in the solution and insert one of them into the position of the other.

[0096] refer to Fig.10 , Fig.10 Schematic diagram of another neighborhood search strategy of the collector line optimization method in the embodiment of the present application; the third neighborhood search strategy is: Execute the first neighbor search strategy twice in a row; It can be seen that a random number between 0 and 1 is generated, which introduces randomness to the strategy selection and avoids falling into the local optimum. Then, the algorithm subtracts the selection probabilities of different neighborhood search strategies in turn, which are dynamically calculated based on the historical update success rate. Specifically, the selection probability reflects the effectiveness of each strategy in the past optimization process. In this way, the algorithm can automatically tend to choose a more effective search strategy while still retaining the possibility of trying other strategies. This adaptive mechanism not only enhances the local search ability of the algorithm, but also improves the convergence speed.

[0097] The following introduces an exemplary collector line optimization system 1100 provided in an embodiment of the present application. Figure 5 It is a schematic diagram of an exemplary hardware structure of the collector line optimization system 1100 provided in an embodiment of the present application.

[0098] In some embodiments, the collector line optimization system 1100 is a computer device or the collector line optimization system 1100 includes a computer device. The computer device includes a processor, a memory and a network interface connected via a system bus. Among them, 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. The network interface of the computer device is used to communicate with other external terminals or servers through a network connection. In some embodiments, the network interface can be a wired network interface, and in some embodiments, the network interface can also be a wireless network interface. When the computer program is executed by the processor, the method in the embodiment of the present application is implemented.

[0099] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0100] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0101] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0102] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media integration. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0103] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A method for optimizing a collector line, characterized in that: include: Generate a collector line population, wherein each individual in the collector line population represents a set of optimal loop numbers and optimal collector lines in the loops; Calculating the fitness value of each individual in the collector line population; The collector line population is used as a loop variable and iterated according to a preset rule until a termination condition is reached; The preset rules are: Find the nearest individual of the current individual, where the current individual is any individual in the collector line population; Determine whether the nearest individual is a better individual; If the nearest individual is a better individual, a copy of the current individual is generated, and the copy is moved to the nearest individual according to the individual movement rule; When the fitness value of the moved copy is less than the fitness value of the current individual, the current individual is updated to the moved copy, and the counter of the current individual is reset; If the nearest individual is not a better individual, a neighborhood search is performed on the current individual; Determine whether the fitness value of the search result is less than the fitness value of the current individual; If it is less than the fitness value of the current individual, the current individual is updated as the search result, and the counter of the current individual is reset; If it is not less than the fitness value of the current individual, then increase the count of the counter; When the count of the counter is greater than the count threshold and the fitness value of the current individual is less than the fitness value of the global optimal solution, updating the global optimal solution to the current individual; After the iteration is completed, the global optimal solution is output as the optimal collector line configuration solution.

2. The method according to claim 1, characterized in that The individual codes of the optimal number of loops and the optimal collector lines in the loops are: The individual code is a digital sequence, wherein the numbers in the digital sequence represent the numbers of the units or booster stations; the numbers of the booster stations in the individual code are used to separate different collector lines; and the order of numbers in the collector line serial number segment represents the connection order of the units on the collector line.

3. The method according to claim 2, characterized in that The step of generating a population of collector lines specifically includes: selecting an initial node in the collector line according to a first roulette selection strategy, wherein the node is a unit or a booster station; The first roulette wheel selection strategy is: Generate a random number between 0 and 1, Subtracting the first selection probability of each of the nodes in the set from the random number in turn; The first selection probability is: In the formula, P(x i ) is the node x i The probability of first choice, dist i,b Represents node x i The distance from booster station b; minDist b Indicates the minimum distance between all units and the booster station, dist j,b Represents node x j The distance from the booster station b; c1 is the artificial coefficient; If the difference is not greater than 0, the corresponding node is determined as the next node selected by the roulette wheel. and\or; Selecting a next node of the current node in the collector line according to a second roulette selection strategy; The second roulette selection strategy is: Generate a random number between 0 and 1, Subtracting the second selection probability of each of the nodes in the set from the random number in turn; The second selection probability is: In the formula, P(x i , x j ) is the node x i Select node x j The second choice probability, maxDist i Represents all units and node x i The maximum distance between j,i Represents node x j With node x i The distance between; maxDist b Indicates the maximum distance between all units and booster station b; dist j,b Represents node x j The distance from the booster station b; c2 is the artificial coefficient; dist k,i Represents node x k With node x i The distance between k,b Represents node x k The distance from booster station b; When the difference is not greater than 0, the corresponding node is determined as the next node selected by the roulette wheel.

4. The method according to claim 3, characterized in that The individual movement rules are: In the case where the number of the current position in the individual code of the copy is different from the number of the corresponding position in the individual code of the better individual, generating a random number between 0 and 1; In the case where the random number is less than the preset selection probability, the number of the current position in the preferred individual has not appeared before the current position in the copy, and the number of the current position is not a booster station; Replace the number of the current position in the individual code of the copy with the number of the corresponding position in the individual code of the better individual; Remove the number of the current position from the unit selection probability vector, and calculate the second roulette selection probability of the current position for each unit in the unit selection probability vector; In the case where the random number is less than the preset selection probability, the number of the current position in the better individual has not appeared before the current position in the copy, the number of the current position is not a booster station, the position corresponding to the better individual is a booster station, and the next position in the copy is not a booster station; the number of the current position in the individual code of the copy is replaced with a booster station, and the first roulette selection probability is calculated for each unit in the unit selection probability vector; In the case where the number of the current position in the individual code of the copy is the same as the number of the corresponding position in the individual code of the better individual; The number of the current position is not a booster station, the number of the current position is removed from the unit selection probability vector, and the second roulette selection probability of the current position is calculated for each unit in the unit selection probability vector.

5. The method according to claim 2, characterized in that: The step of finding the nearest individual to the current individual specifically includes: Calculate the number of characters required to replace the individual code of the current individual to transform it into the individual code of other individuals, where the other individuals are other individuals in the collector line population except the current individual; The other individual with the smallest number of characters to be replaced is determined as the nearest individual.

6. The method according to claim 2, characterized in that The neighborhood search is: Generate a random number between 0 and 1; Subtract the selection probabilities of different neighborhood search strategies in turn, where the selection probability is the ratio of the number of historical updates of the neighborhood search strategy to the total number of historical updates; When the random number is less than or equal to 0, a corresponding neighborhood search strategy is executed.

7. The method according to claim 1, characterized in that The fitness value is: minf(x)=K·A(x)+B·P(x)+K·C(x) In the formula, K is the conversion coefficient between present value and equivalent annual value; A(x) is the total line investment corresponding to individual x; B is the electricity price; P(x) is the annual line loss corresponding to individual x; C(x) is the switch cabinet price corresponding to individual x; Where A(x) is the total line investment corresponding to individual x, m represents the number of collector line loops; A i represents the unit investment of the ith line; l i represents the total length of the i-th line; Where P(x) is the annual line loss corresponding to individual x, l i represents the total length of the ith line, h is the annual operating hours; R i represents the resistance of the ith line; C(x)=m×c In the formula, m is the number of switch cabinets, and c is the price of a single switch cabinet.

8. A collector line optimization system, characterized in that: The collector line optimization system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the collector line optimization system to execute the method described in any one of claims 1-7.

9. A computer program product comprising instructions, characterized in that When the computer program product runs on a collector line optimization system, the collector line optimization system is enabled to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a collector line optimization system, the collector line optimization system is caused to execute the method according to any one of claims 1 to 7.

Citation Information

Cited By

  • Method and system for realizing energy efficiency optimization of photovoltaic system based on big data analysis algorithm

    CN119558204A

  • Large-scale design-oriented high-efficiency area and line length optimization macro layout method

    CN119886035A

  • Efficient area and wirelength optimization macro placement method for large-scale design

    CN119886035B