An offshore wind farm power collection system topology optimization method based on immune algorithm

Abstract

The present application belongs to the field of wind farm power collection system topology optimization technology, and provides a kind of offshore wind farm power collection system topology optimization method based on immune algorithm: according to the rated power of offshore wind farm booster station, the rated voltage of power collection system and the threshold interval of the carrying capacity of selected sea cable, the threshold interval of feeder number and the threshold interval of the number of wind turbines allowed to be connected to single feeder are calculated;With offshore wind farm booster station as pole point, polar coordinate system is established, and according to the above threshold interval, a plurality of initial wind turbine grouping schemes are generated with different grouping numbers and different starting wind turbines;The minimum spanning tree algorithm is used to connect the wind turbines in each group to form an initial power collection system topology scheme, and the initial antibody population is screened out by fitness evaluation function and stored in the optimal antibody;Iterative optimization is carried out by using immune algorithm, and by updating memory bank, retaining high-quality antibodies, replacing poor-quality antibodies and randomly generating new antibodies, the antibody population is continuously optimized, and finally the optimal power collection system topology scheme is output.

Description

Technical Field

[0001] This invention relates to the field of topology optimization technology for wind farm power collection systems, specifically a topology optimization method for offshore wind farm power collection systems based on an immune algorithm. Background Technology

[0002] As the global energy structure shifts towards cleaner and lower-carbon energy sources, offshore wind power, as an important form of renewable energy, is experiencing continuous expansion in its development and utilization. The offshore wind farm's data collection system is responsible for collecting and transmitting the electricity generated by distributed wind turbines to offshore substations. The topology optimization design of this system directly impacts investment costs, operational reliability, and maintenance economics. With the increasing distance of offshore wind farms from shore and the expansion of installed capacity, the investment cost of the data collection system has risen significantly, with the purchase and laying of submarine cables being particularly expensive. Furthermore, the harsh marine environment and complex engineering conditions increase the risk of system failures and make maintenance difficult and costly. Traditional wind farm data collection system topology designs often rely on engineers' experience, making it difficult to minimize the total life-cycle cost while meeting electrical and marine engineering constraints.

[0003] To address the aforementioned issues, existing technologies have attempted to transform the topology design of power collection systems into a multi-objective combinatorial optimization problem using mathematical modeling and intelligent algorithms. This problem is characterized by diverse optimization objectives, complex constraints, and high problem dimensionality. Conventional optimization algorithms often suffer from drawbacks when handling such combinatorial optimization problems, including low optimization efficiency, susceptibility to getting trapped in optimal solutions, and difficulty in obtaining high-quality solutions within a reasonable timeframe. Therefore, significantly improving the solution efficiency and solution quality of the topology optimization process for offshore wind farm power collection systems while ensuring feasibility and economy has become a critical technical problem urgently needing to be solved in this field. Summary of the Invention

[0004] The purpose of this invention is to provide a topology optimization method for offshore wind farm power collection systems based on immune algorithms. This method has the advantages of efficiently exploring the solution space, avoiding getting trapped in local optima, and quickly obtaining high-quality topology schemes, thereby significantly reducing submarine cable investment costs, improving system reliability, and optimizing the economics of the entire life cycle.

[0005] This invention provides 1. a topology optimization method for offshore wind farm power collection systems based on an immune algorithm, comprising the following steps:

[0006] S1: Based on the rated power of the offshore wind farm booster station, the rated voltage of the power collection system, and the current carrying capacity threshold range of the submarine cable to be selected, calculate the threshold range of the number of feeders connecting the offshore wind farm booster station and the threshold range of the number of wind turbines that can be connected to a single feeder, and determine the grouping parameters.

[0007] S2: Establish a polar coordinate system with the offshore wind farm booster station as the pole. Based on the feeder number threshold range and the wind turbine number threshold range that a single feeder can connect to, group all wind turbines in the offshore wind farm multiple times with different grouping numbers and different starting wind turbines as the dividing starting point to obtain multiple initial wind turbine grouping schemes.

[0008] S3: For each initial wind turbine grouping scheme in S2, the minimum spanning tree algorithm is used to connect the wind turbines in each group, generating multiple initial power collection system topology schemes. The fitness evaluation function is used to evaluate the multiple initial power collection system topology schemes, and the initial power collection system topology schemes with the top 10% fitness ranking are selected as the initial antibody group. The high-quality antibody with the highest fitness is stored in the immune memory bank. The power collection system topology scheme includes at least the grouping result of the wind turbines, the connection relationship between the wind turbines in each group, the selection of each section of submarine cable, and the length of the submarine cable.

[0009] S4: The initial antibody population is iteratively optimized using an immune algorithm. The immune algorithm includes at least an immune memory mechanism, an immune concentration regulation mechanism, and an immune inoculation mechanism. In each iteration of optimization, the fitness of the current antibody population is evaluated and sorted, and the high-quality antibody with the highest fitness is stored in the immune memory bank.

[0010] S5: After iterative optimization until the termination condition is met, output the final power collection system topology scheme corresponding to the high-quality antibody with the highest fitness in the immune memory bank;

[0011] S6: Summarize the grouping results of wind turbines in the final power collection system topology scheme, the connection relationship between wind turbines in each group, the selection of each section of submarine cable and the length of the submarine cable, and generate the offshore wind farm power collection system topology model.

[0012] In a preferred implementation, the feeder quantity threshold range in S1 and the allowable number of wind turbines connected to each feeder are determined by the following formula:

[0013] (1)

[0014] (2)

[0015] In the formula, and These are the maximum and minimum values ​​for the feeders that can be connected to the offshore substation, respectively. The rated power of the offshore booster station; and These represent the maximum and minimum number of wind turbines that can be connected to a single feeder, respectively. This is the rated voltage of the power collection system for offshore wind farms. and These are the current carrying capacities of the submarine cables with the highest and lowest current carrying capacities among a variety of optional submarine cables. The power factor.

[0016] In a preferred implementation, step S2 involves grouping all wind turbines in the offshore wind farm multiple times based on the feeder quantity threshold range and the wind turbine number threshold range allowed to be connected to a single feeder, using different grouping numbers and different initial wind turbines as the dividing starting point, to obtain multiple initial wind turbine grouping schemes, specifically including:

[0017] S21: Within the aforementioned feeder quantity threshold range, select different feeder quantity values;

[0018] S22: For each selected feeder quantity value, within the threshold range of the number of wind turbines that can be connected to a single feeder, determine the number of wind turbines in different groups;

[0019] S23: For each group's number of fans, select different starting fans and sort them by polar angle to generate multiple initial fan grouping schemes, represented as follows:

[0020] (3)

[0021] In the formula, A collection of wind turbines for offshore wind farms; This is the first group in the offshore wind farm turbine grouping; This is the second group in the offshore wind farm turbine grouping; Grouping offshore wind turbines Group; Grouping of offshore wind turbines Group; for The first fan in the fan group; For wind turbine The second wind turbine in the group; For wind turbine The first in the group Typhoon machine.

[0022] In a preferred implementation, the expression for the fitness evaluation function in S3 is:

[0023] (4)

[0024] In the formula, The total cost of the power collection system for an offshore wind farm; For fitness evaluation function; Topology connection scheme for offshore wind farm power collection system; A collection of submarine cables that can be selected for offshore wind farms; This refers to the penalty for violating constraints in the topology connection scheme of the power collection system of an offshore wind farm.

[0025] In a preferred implementation, step S4 employs an immune algorithm to iteratively optimize the initial antibody population, specifically including:

[0026] S41: Construct a first-generation antibody population, wherein a portion of the first-generation antibody population consists of the remaining antibodies in the initial antibody population excluding the high-quality antibody with the highest fitness, and another portion of the first-generation antibody population consists of new antibodies generated by mutations of the remaining antibodies;

[0027] S42: Use the fitness evaluation function to evaluate and rank the first-generation antibody population and subsequent generations of antibody populations to distinguish between high-quality antibodies and low-quality antibodies.

[0028] S43: Update the optimal antibody with the highest fitness in the current antibody population to the immune memory bank;

[0029] S44: High-quality antibodies, excluding the best antibody, from the current antibody population are retained for the next generation; the inferior antibodies are replaced by high-quality antibodies from the immune memory bank; and some new antibodies are randomly generated through an immunization mechanism to maintain antibody diversity.

[0030] In a preferred implementation, the immunization mechanism randomly generates some new antibodies by randomly exchanging fans within different groups. The process of the immunization mechanism randomly generating some new antibodies is specifically represented as follows:

[0031] (5)

[0032] In the formula, and After the offshore wind turbines were grouped and vaccinated, and Group, , and These represent the groups of wind turbines. The first, second and third in the group Typhoon machine, , and These represent the groups of wind turbines. The first, second and third in the group Typhoon machine, and These represent the groups of wind turbines. Group 1 Typhoon generators and wind turbine groups Group 1 Typhoon machine, and It is a random number.

[0033] In a preferred implementation, the termination condition includes reaching a preset maximum number of iterations; and / or

[0034] No significant improvement in optimal fitness across multiple bands; and / or

[0035] The fitness value meets the preset cost threshold.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] This invention groups wind turbines based on their physical locations and combines this with an immune algorithm for iterative optimization. This decomposes the high-latitude topology problem of offshore wind farm power collection systems into a multi-stage problem of grouping, connection, and optimization, significantly reducing the optimization difficulty and unnecessary computation. Thus, while ensuring quality, it significantly improves the efficiency of optimization solutions. Attached Figure Description

[0038] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0039] Figure 1 This is the overall algorithm architecture diagram of the method of this invention.

[0040] Figure 2 This is the initial grouping part in the overall algorithm architecture of the method of this invention.

[0041] Figure 3 It is the iterative optimization part in the overall algorithm architecture of the method of this invention.

[0042] Figure 4 This is a flowchart of the fitness function evaluation method in the overall algorithm architecture of the present invention. Detailed Implementation

[0043] The present invention will be further described below through specific embodiments, but this is not a limitation of the present invention. Those skilled in the art can make various modifications or improvements based on the basic idea of ​​the present invention, but as long as they do not depart from the basic idea of ​​the present invention, they are all within the protection scope of the present invention.

[0044] Traditional offshore wind farm collector system topology design relies heavily on engineers' experience, making it difficult to minimize life-cycle costs while meeting electrical and marine engineering constraints. As offshore wind farms continue to increase their distance from shore and their installed capacity expands, the investment cost of collector systems is rising significantly, with the purchase and laying of submarine cables being particularly expensive. Existing optimization algorithms often suffer from low optimization efficiency, susceptibility to local optima, and difficulty in obtaining high-quality solutions within a reasonable timeframe when dealing with such multi-objective combined optimization problems.

[0045] To address this, this invention proposes a topology optimization method for offshore wind farm power collection systems based on an immune algorithm, see reference. Figures 1-4 This includes the following steps:

[0046] S1: Based on the rated power of the offshore wind farm booster station, the rated voltage of the power collection system, and the current carrying capacity threshold range of the submarine cable to be selected, calculate the threshold range of the number of feeders connecting the offshore wind farm booster station and the threshold range of the number of wind turbines that can be connected to a single feeder, and determine the grouping parameters.

[0047] S2: Establish a polar coordinate system with the offshore wind farm booster station as the pole. Based on the feeder number threshold range and the wind turbine number threshold range that a single feeder can connect to, group all wind turbines in the offshore wind farm multiple times with different grouping numbers and different starting wind turbines as the dividing starting point to obtain multiple initial wind turbine grouping schemes.

[0048] S3: For each initial wind turbine grouping scheme in S2, the minimum spanning tree algorithm is used to connect the wind turbines in each group, generating multiple initial power collection system topology schemes. The fitness evaluation function is used to evaluate the multiple initial power collection system topology schemes, and the initial power collection system topology schemes with the top 10% fitness ranking are selected as the initial antibody group. The high-quality antibody with the highest fitness is stored in the immune memory bank. The power collection system topology scheme includes at least the grouping result of the wind turbines, the connection relationship between the wind turbines in each group, the selection of each section of submarine cable, and the length of the submarine cable.

[0049] S4: The initial antibody population is iteratively optimized using an immune algorithm. The immune algorithm includes at least an immune memory mechanism, an immune concentration regulation mechanism, and an immune inoculation mechanism. In each iteration of optimization, the fitness of the current antibody population is evaluated and sorted, and the high-quality antibody with the highest fitness is stored in the immune memory bank.

[0050] S5: After iterative optimization until the termination condition is met, output the final power collection system topology scheme corresponding to the high-quality antibody with the highest fitness in the immune memory bank;

[0051] S6: Summarize the grouping results of wind turbines in the final power collection system topology scheme, the connection relationship between wind turbines in each group, the selection of each section of submarine cable and the length of the submarine cable, and generate the offshore wind farm power collection system topology model.

[0052] This invention effectively solves the complex combinatorial problem in the topology optimization of offshore wind farm power collection systems by combining an immune algorithm with a multi-stage optimization strategy. By systematically determining feeder parameters, generating diverse initial grouping schemes, and utilizing the memory, concentration adjustment, and seeding mechanisms of the immune algorithm for iterative optimization, this method significantly improves the solution efficiency of the optimization process, avoids getting trapped in local optima, and thus obtains a high-quality power collection system topology scheme, reduces the system's total lifecycle cost, and improves operational reliability.

[0053] In a preferred implementation, the feeder quantity threshold range in S1 and the allowable number of wind turbines connected to each feeder are determined by the following formula:

[0054] (1)

[0055] (2)

[0056] In the formula, and These are the maximum and minimum values ​​for the feeders that can be connected to the offshore substation, respectively. The rated power of the offshore booster station; and These represent the maximum and minimum number of wind turbines that can be connected to a single feeder, respectively. This is the rated voltage of the power collection system for offshore wind farms. and These are the current carrying capacities of the submarine cables with the highest and lowest current carrying capacities among a variety of optional submarine cables. The power factor.

[0057] The above technical solution avoids the uncertainties and irrationalities of traditional empirical estimations, ensuring the scientific validity and accuracy of the grouping parameters in step S1. Since the threshold ranges for the number of feeders and the threshold range for the number of wind turbines that can be connected to a single feeder are precisely defined, the initial wind turbine grouping scheme generated in step S2 will better conform to actual engineering constraints and electrical requirements, thereby reducing the number of invalid or low-quality initial schemes. Based on this, the initial collector system topology scheme generated by the minimum spanning tree algorithm in step S3 will have higher feasibility and better initial fitness, providing a high-quality initial antibody population for the subsequent immune algorithm iterative optimization in step S4. This significantly improves the convergence speed and optimization efficiency of the immune algorithm, enabling it to find a collector system topology scheme closer to the global optimum in a shorter time. The final output collector system topology model is effectively guaranteed in terms of both economy and reliability.

[0058] In a preferred implementation, step S2 involves grouping all wind turbines in the offshore wind farm multiple times based on the feeder quantity threshold range and the wind turbine number threshold range allowed to be connected to a single feeder, using different grouping numbers and different initial wind turbines as the dividing starting point, to obtain multiple initial wind turbine grouping schemes, specifically including:

[0059] S21: Within the aforementioned feeder quantity threshold range, select different feeder quantity values;

[0060] S22: For each selected feeder quantity value, within the threshold range of the number of wind turbines that can be connected to a single feeder, determine the number of wind turbines in different groups;

[0061] S23: For each group's number of fans, select different starting fans and sort them by polar angle to generate multiple initial fan grouping schemes, represented as follows:

[0062] (3)

[0063] In the formula, A collection of wind turbines for offshore wind farms; This is the first group in the offshore wind farm turbine grouping; This is the second group in the offshore wind farm turbine grouping; Grouping offshore wind turbines Group; Grouping of offshore wind turbines Group; for The first fan in the fan group; For wind turbine The second wind turbine in the group; For wind turbine The first in the group Typhoon machine.

[0064] This invention, through the aforementioned technical solution, systematically traverses the number of feeders, the number of fans within a group, and the selection of the initial fan, ensuring the comprehensiveness and diversity of the initial fan grouping scheme. First, S21 covers the overall configuration of the power collection system at different scales by selecting different feeder quantity values. Second, S22 further explores various combinations of the number of fans within a group for each feeder quantity value, making the load allocation of each feeder more flexible. Finally, S23 introduces spatial layout diversity by selecting different initial fans for polar angle sorting, avoiding the limitations caused by a single sorting starting point. This multi-level, multi-dimensional initial scheme generation strategy significantly increases the diversity of the initial antibody population, providing a broader optimization space for the subsequent global search of the immune algorithm, thereby effectively avoiding the algorithm getting trapped in local optima and improving the quality and economy of the final power collection system topology optimization results.

[0065] In a preferred implementation, the expression for the fitness evaluation function in S3 is:

[0066] (4)

[0067] In the formula, The total cost of the power collection system for an offshore wind farm; For fitness evaluation function; Topology connection scheme for offshore wind farm power collection system; A collection of submarine cables that can be selected for offshore wind farms; This refers to the penalty for violating constraints in the topology connection scheme of the power collection system of an offshore wind farm.

[0068] This fitness evaluation function aims to measure the merits of different topology schemes for offshore wind farm power collection systems. Its core idea is to integrate the economic efficiency (reflected by a cost evaluation function) with the engineering feasibility and compliance (reflected by penalty clauses for constraint violations). This combined approach ensures that the optimization process not only pursues the lowest possible cost but also meets all necessary electrical, marine engineering, and other constraints.

[0069] This invention effectively solves the problems of incomplete evaluation and difficulty in simultaneously considering economy and engineering feasibility in the process of topology optimization of collector systems by using the above-mentioned fitness evaluation function.

[0070] In a preferred implementation, step S4 employs an immune algorithm to iteratively optimize the initial antibody population, specifically including:

[0071] S41: Construct a first-generation antibody population, wherein a portion of the first-generation antibody population consists of the remaining antibodies in the initial antibody population excluding the high-quality antibody with the highest fitness, and another portion of the first-generation antibody population consists of new antibodies generated by mutations of the remaining antibodies;

[0072] S42: Use the fitness evaluation function to evaluate and rank the first-generation antibody population and subsequent generations of antibody populations to distinguish between high-quality antibodies and low-quality antibodies.

[0073] S43: Update the optimal antibody with the highest fitness in the current antibody population to the immune memory bank;

[0074] S44: High-quality antibodies, excluding the best antibody, from the current antibody population are retained for the next generation; the inferior antibodies are replaced by high-quality antibodies from the immune memory bank; and some new antibodies are randomly generated through an immunization mechanism to maintain antibody diversity.

[0075] Specifically, in S41, the remaining antibodies in the initial antibody swarm, excluding the high-fitness superior antibody, are combined with new antibodies generated by random mutation to form the first-generation antibody swarm. This introduces new solutions early in the optimization process, increasing the diversity of the antibody swarm and preventing the algorithm from prematurely converging to a local optimum.

[0076] In step S42, a fitness evaluation function is used to evaluate all antibodies in the current generation and distinguish between high-quality and low-quality antibodies. This evaluation process is the core of the immune algorithm, used to quantify the merits of each collector system topology scheme. The fitness evaluation function is typically the sum of a cost evaluation function and a constraint violation penalty term, as described in the above scheme.

[0077] In step S43, the optimal antibody with the highest fitness in the current generation is updated to the immune memory. The role of the immune memory is to store the best solution discovered by the algorithm so far, ensuring that high-quality information is not lost during iteration. When the current generation produces an antibody that is better than the antibody stored in the memory, this optimal antibody will be stored in the memory, thus ensuring that the memory always contains the globally optimal or near-globally optimal solution.

[0078] In step S44, high-quality antibodies from the current generation, excluding the optimal antibody, are retained for the next generation; low-quality antibodies are replaced by high-quality antibodies from the immune memory bank; and simultaneously, some new antibodies are randomly generated through an immunization mechanism to maintain antibody diversity. This step is crucial for achieving a balance between exploration and development in the algorithm. On one hand, retaining high-quality antibodies ensures that the algorithm evolves towards better solutions, utilizing discovered beneficial characteristics. On the other hand, replacing low-quality antibodies with high-quality antibodies from the immune memory bank effectively eliminates low-quality solutions and introduces validated excellent gene fragments, accelerating convergence. Furthermore, randomly generating some new antibodies through the immunization mechanism, such as by randomly exchanging fans within different groups or by locally recombining existing antibodies, continuously introduces new gene information, preventing the antibody population from getting trapped in local optima, thereby maintaining population diversity and enhancing the algorithm's global search capability.

[0079] Through the above technical solution, the present invention effectively solves the problems of insufficient antibody population diversity, slow convergence speed and easy getting trapped in local optima in the topology optimization process of offshore wind farm power collection system.

[0080] In a preferred implementation, the immunization mechanism randomly generates some new antibodies by randomly exchanging fans within different groups. The process of the immunization mechanism randomly generating some new antibodies is specifically represented as follows:

[0081] (5)

[0082] In the formula, and After the offshore wind turbines were grouped and vaccinated, and Group, , and These represent the groups of wind turbines. The first, second and third in the group Typhoon machine, , and These represent the groups of wind turbines. The first, second and third in the group Typhoon machine, and These represent the groups of wind turbines. Group 1 Typhoon generators and wind turbine groups Group 1 Typhoon machine, and It is a random number.

[0083] Specifically, the immunization mechanism of this application can generate novel antibodies. These new antibodies are not generated completely randomly, but rather with local adjustments based on existing grouping schemes. Specifically, when from two groups... and Randomly select a fan. and During the exchange, the new grouping scheme inherits some of the original grouping structure while introducing new combinations. This localized and purposeful mutation method allows the algorithm to more effectively escape local optima when exploring the solution space. This mechanism of randomly exchanging wind turbines effectively solves the problems of low optimization efficiency and insufficient solution quality that may be caused by traditional random generation of new antibodies. By making structural adjustments based on the existing grouping scheme, rather than generating it completely randomly, the algorithm can explore the solution space more intelligently and avoid getting trapped in local optima. This not only maintains the diversity of the antibody population and ensures the algorithm's global search capability, but also, because there is a certain "bloodline" relationship between the new antibody and the high-quality antibody, the new antibody is more likely to inherit the excellent characteristics of the high-quality antibody, thereby improving the efficiency of the optimization process and the quality of the final power collection system topology scheme. In this way, this application can obtain a more economical and feasible optimization scheme for the power collection system topology of offshore wind farms.

[0084] In a preferred implementation, the termination condition in S5 includes reaching a preset maximum number of iterations; and / or

[0085] No significant improvement in optimal fitness across multiple bands; and / or

[0086] The fitness value meets the preset cost threshold.

[0087] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A topology optimization method for offshore wind farm power collection systems based on an immune algorithm, characterized in that, Includes the following steps: S1: Based on the rated power of the offshore wind farm booster station, the rated voltage of the power collection system, and the current carrying capacity threshold range of the submarine cable to be selected, calculate the threshold range of the number of feeders connecting the offshore wind farm booster station and the threshold range of the number of wind turbines that can be connected to a single feeder, and determine the grouping parameters. S2: Establish a polar coordinate system with the offshore wind farm booster station as the pole. Based on the feeder number threshold range and the wind turbine number threshold range that a single feeder can connect to, group all wind turbines in the offshore wind farm multiple times with different grouping numbers and different starting wind turbines as the dividing starting point to obtain multiple initial wind turbine grouping schemes. S3: For each initial wind turbine grouping scheme in S2, the minimum spanning tree algorithm is used to connect the wind turbines in each group, generating multiple initial power collection system topology schemes. The fitness evaluation function is used to evaluate the multiple initial power collection system topology schemes, and the initial power collection system topology schemes with the top 10% fitness ranking are selected as the initial antibody group. The high-quality antibody with the highest fitness is stored in the immune memory bank. The power collection system topology scheme includes at least the grouping result of the wind turbines, the connection relationship between the wind turbines in each group, the selection of each section of submarine cable, and the length of the submarine cable. S4: The initial antibody population is iteratively optimized using an immune algorithm. The immune algorithm includes at least an immune memory mechanism, an immune concentration regulation mechanism, and an immune inoculation mechanism. In each iteration of optimization, the fitness of the current antibody population is evaluated and sorted, and the high-quality antibody with the highest fitness is stored in the immune memory bank. S5: After iterative optimization until the termination condition is met, output the final power collection system topology scheme corresponding to the high-quality antibody with the highest fitness in the immune memory bank; S6: Summarize the grouping results of wind turbines in the final power collection system topology scheme, the connection relationship between wind turbines in each group, the selection of each section of submarine cable and the length of the submarine cable, and generate the offshore wind farm power collection system topology model.

2. The method for topology optimization of offshore wind farm collection systems based on immune algorithms according to claim 1, characterized in that, The threshold range for the number of feeders in S1 and the threshold range for the number of wind turbines that can be connected to each feeder are determined by the following formula: （1） （2） In the formula, and These are the maximum and minimum values ​​for the feeders that can be connected to the offshore booster station, respectively. The rated power of the offshore booster station; and These represent the maximum and minimum number of wind turbines that can be connected to a single feeder, respectively. This is the rated voltage of the power collection system for offshore wind farms. and These are the current carrying capacities of the submarine cables with the highest and lowest current carrying capacities among a variety of optional submarine cables. The power factor.

3. The method for topology optimization of offshore wind farm collection systems based on immune algorithms according to claim 1, characterized in that, In step S2, based on the feeder quantity threshold range and the wind turbine quantity threshold range allowed to be connected to a single feeder, all wind turbines in the offshore wind farm are grouped multiple times using different grouping numbers and different initial wind turbines as the dividing starting point, resulting in multiple initial wind turbine grouping schemes, specifically including: S21: Within the aforementioned feeder quantity threshold range, select different feeder quantity values; S22: For each selected feeder quantity value, within the threshold range of the number of wind turbines that can be connected to a single feeder, determine the number of wind turbines in different groups; S23: For each group's number of fans, select different starting fans and sort them by polar angle to generate multiple initial fan grouping schemes, represented as follows: （3） In the formula, A collection of wind turbines for offshore wind farms; This is the first group in the offshore wind farm turbine grouping; This is the second group in the offshore wind farm turbine grouping; Grouping offshore wind turbines Group; Grouping of offshore wind turbines Group; for The first fan in the fan group; For wind turbine The second wind turbine in the group; For wind turbine The first in the group Typhoon machine.

4. The method for topology optimization of offshore wind farm collection systems based on immune algorithms according to claim 1, characterized in that, The expression for the fitness evaluation function in S3 is: （4） In the formula, The total cost of the offshore wind farm's power collection system; For fitness evaluation function; Topology connection scheme for offshore wind farm power collection system; A collection of submarine cables that can be selected for offshore wind farms; This refers to the penalty for violating constraints in the topology connection scheme of the power collection system of an offshore wind farm.

5. The method for topology optimization of offshore wind farm collection systems based on immune algorithms according to claim 1, characterized in that, In step S4, an immune algorithm is used to iteratively optimize the initial antibody population, specifically including: S41: Construct a first-generation antibody population, wherein a portion of the first-generation antibody population consists of the remaining antibodies in the initial antibody population excluding the high-quality antibody with the highest fitness, and another portion of the first-generation antibody population consists of new antibodies generated by mutations of the remaining antibodies; S42: Use the fitness evaluation function to evaluate and rank the first-generation antibody population and subsequent generations of antibody populations to distinguish between high-quality antibodies and low-quality antibodies. S43: Update the optimal antibody with the highest fitness in the current antibody population to the immune memory bank; S44: High-quality antibodies, excluding the best antibody, from the current antibody population are retained for the next generation; the inferior antibodies are replaced by high-quality antibodies from the immune memory bank; and some new antibodies are randomly generated through an immunization mechanism to maintain antibody diversity.

6. The method for topology optimization of offshore wind farm collection systems based on immune algorithms according to claim 5, characterized in that, The immunization mechanism randomly generates some new antibodies by randomly exchanging fans within different groups. The specific process of this immunization mechanism randomly generating some new antibodies is as follows: （5） In the formula, and After the offshore wind farm turbines were grouped and immunized, and Group, , and These represent the groups of wind turbines. The first, second and third in the group Typhoon machine, , and These represent the groups of wind turbines. The first, second and third in the group Typhoon machine, and These represent the groups of wind turbines. Group 1 Typhoon generators and wind turbine groups Group 1 Typhoon machine, and It is a random number.

7. The method for topology optimization of offshore wind farm collection systems based on immune algorithms according to claim 1, characterized in that, The termination condition includes reaching a preset maximum number of iterations; and / or No significant improvement in optimal fitness across multiple bands; and / or The fitness value meets the preset cost threshold.

Images