Topological Optimization Method and System of a Collector System Based on a Particle Swarm Genetic Hybrid Algorithm

Abstract

The present invention relates to the field of new energy power generation technology, and specifically discloses a method and system for optimizing the topology of a collector system based on a particle swarm genetic hybrid algorithm. The method includes: establishing an objective function for optimizing the topology of the collector system with the lowest total life cycle cost as the goal, and at the same time considering the constraints of the cable current-carrying capacity, the number of wind turbines connected by the cable, and the cable overlap as constraint conditions to obtain a mathematical model of the topology structure of the collector system for an offshore wind farm; using an adaptive particle swarm genetic hybrid algorithm to solve the mathematical model to obtain the optimal topology structure of the collector system; when solving the model, the present invention adopts an adaptive particle swarm genetic hybrid algorithm, the coding adopts a hybrid coding method, and the reciprocal of the square of the total life cycle cost is selected as the fitness function to highlight the differences between individuals. The decoded result is not only an economically optimal topology scheme, but also can intuitively and clearly represent the topology structure.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy power generation, and particularly to a topology optimization method and system for a collector system based on a particle swarm genetic hybrid algorithm. Background Art

[0002] The statements in this part only provide background technical information related to the present invention, and do not necessarily constitute prior art.

[0003] Wind power generation is a way to convert wind energy into electrical energy. Compared with onshore wind power, offshore wind power has richer and more stable wind resources, is low-carbon and environmentally friendly, and saves land occupation area. In recent years, offshore wind power has developed rapidly and is gradually becoming the future development direction of wind power. An offshore wind farm mainly consists of three parts: a wind turbine group, a collector system, and a power transmission system. Among them, the collector system is the connection part between the wind turbine group and the power transmission system. The optimized design of this part has a crucial impact on the economy and reliability of the operation of the offshore wind farm. The electrical equipment in the collector system is numerous and complex, and its economic construction cost accounts for a relatively large proportion in the construction of the offshore wind farm. Moreover, compared with onshore wind farms, the offshore environment is more severe, and its operation and maintenance costs are higher.

[0004] Once a failure occurs in the collector system, it may cause large-scale power losses in the entire wind farm, thereby affecting economic benefits. For an offshore wind farm with a selected site, once the model of the wind turbine is selected, the investment cost of the wind turbine is relatively fixed. If we want to save economic costs, there will be more flexibility starting from the line topology structure of the collector system. Optimizing the line topology structure can reduce the economic cost of the offshore wind farm and effectively improve the operation efficiency of the system. The topology optimization problem of the offshore wind power collector system usually takes economy and reliability as the objective functions.

[0005] The topology optimization problem of the collector system is a multi-dimensional non-linear optimization problem. Existing methods mostly use intelligent optimization algorithms such as genetic algorithms, ant colony algorithms, and fuzzy clustering algorithms, or algorithms based on graph theory such as the DMST algorithm and the Dijkstra algorithm for solution. However, the optimization and convergence effects of the above intelligent optimization algorithms are relatively lacking. For example, genetic algorithms are easily affected by the initial population and parameter settings, deviating from the optimal solution direction or converging prematurely; ant colony algorithms are easily trapped in local optima and have a slow convergence speed due to the limitations of the pheromone update strategy; it is difficult to select clustering centers and determine membership functions for fuzzy clustering algorithms; algorithms based on graph theory have limitations in network structure assumptions, high complexity, limitations in single-source shortest paths, and poor dynamic adaptability, and their optimization and convergence effects are relatively lacking. Summary of the Invention

[0006] To solve the above problems, the present invention proposes a topology optimization method and system for a collector system based on a particle swarm genetic hybrid algorithm, taking the life cycle cost as the objective function, and at the same time introducing the cable current-carrying capacity, the number of wind turbines connected, and the cable intersection as constraint conditions. When solving the objective, an adaptive particle swarm genetic hybrid algorithm is introduced to find the optimal parameters, a hybrid coding is used to represent the connection situation and position of the wind turbines, a suitable fitness function is selected, and the design and dynamic update mechanism of its genetic operator are used. The decoded result is not only the economically optimal topology scheme, but also can intuitively and clearly represent the topological structure.

[0007] In some embodiments, the following technical solutions are adopted:

[0008] A topology optimization method for a collector system based on a particle swarm genetic hybrid algorithm, comprising:

[0009] Taking the lowest life cycle cost as the objective, establishing an objective function for the topology optimization of the collector system, and at the same time considering the cable current-carrying capacity constraint, the number of wind turbines connected by the cable, and the cable overlap as constraint conditions, to obtain a mathematical model of the topology structure of the offshore wind farm collector system;

[0010] Using an adaptive particle swarm genetic hybrid algorithm to solve the mathematical model to obtain the optimal topology structure of the collector system;

[0011] Wherein, the process of using an adaptive particle swarm genetic hybrid algorithm to solve the mathematical model is specifically as follows:

[0012] Using a hybrid coding method to encode the topology structure of the collector system, randomly generating an initial population to represent different topology structures, and each topology structure represents an individual; designing a fitness function to evaluate the fitness of each individual, and updating and recording the current global optimal solution of the population;

[0013] Taking the individuals in the current population as particles, using an adaptive weighted particle swarm algorithm to iteratively update the particles. The particles record their own optimal positions during the search process to obtain the current individual optimal solution of each particle; taking the updated all individual optimal solutions as a new population, performing genetic operations to generate a new generation of individuals; calculating the fitness values of the new generation of individuals and sorting them together with the fitness value of the current global optimal solution of the population to obtain a new global optimal solution of the population; repeating this process until the termination condition is met, obtaining the global optimal solution of the population and decoding it, so as to obtain the optimal topology structure of the collector system.

[0014] As a further solution, taking the lowest life cycle cost as the objective, establishing an objective function for the topology optimization of the collector system, specifically:

[0015] ;

[0016] Among them, is the life cycle cost, is the equipment procurement and construction cost in the infrastructure stage, is the operation cost in the operation stage , maintenance cost and power outage loss cost sum, is the equipment disposal cost in the recovery stage; is the discount factor.

[0017] As a further solution, the specific constraint on the ampacity of the submarine cable is:

[0018] ;

[0019] Among them, is the maximum continuous load current flowing through a certain section of the submarine cable, is the total correction factor of the allowable ampacity of the submarine cable, is the ampacity of the submarine cable.

[0020] As a further solution, the specific constraint on the number of wind turbines connected to the submarine cable is:

[0021] ;

[0022] ;

[0023] Among them, is the number of wind turbines connected to a single submarine cable, is the maximum number of wind turbines that the submarine cable can connect, is the floor function, is the maximum wind turbine capacity that the submarine cable can withstand, is the rated capacity of a single wind turbine.

[0024] As a further solution, the specific constraint on the overlap of the submarine cable is:

[0025] ;

[0026] Among them, , respectively represent two sections of submarine cable between four wind turbines, is the point , and cross product.

[0027] As a further solution, a hybrid coding method is used to code the topology structure of the collection system, specifically:

[0028] Use binary bits to represent the connection relationship between the collector submarine cable and the wind turbines. 1 represents connection, and 0 represents non-connection; the connection method between each wind turbine and the submarine cable corresponds to a binary bit, which serves as the connection code.

[0029] Use real values to represent the positions of the wind turbines. Represent the positions of each wind turbine with two-dimensional coordinates, and the real number code obtained by combining the positions of all wind turbines in sequence serves as the position code.

[0030] Combine the connection code and the position code into an individual, representing a topological structure of the collector system.

[0031] As a further solution, design a fitness function to evaluate the fitness of each individual. The specific fitness function is as follows:

[0032] ;

[0033] Among them, is the life cycle cost.

[0034] In some other embodiments, the following technical solutions are adopted:

[0035] A collector system topology optimization system based on a particle swarm genetic hybrid algorithm, including:

[0036] A model construction module, which is used to establish an objective function for collector system topology optimization with the goal of minimizing the life cycle cost, and at the same time consider the submarine cable current-carrying capacity constraint, the number of wind turbines connected to the submarine cable, and the submarine cable overlap as constraint conditions to obtain a mathematical model of the topological structure of the offshore wind farm collector system.

[0037] A model solving module, which is used to solve the mathematical model using an adaptive particle swarm genetic hybrid algorithm to obtain the optimal topological structure of the collector system.

[0038] Among them, the process of solving the mathematical model using an adaptive particle swarm genetic hybrid algorithm is specifically as follows:

[0039] Use a hybrid coding method to encode the topological structure of the collector system, randomly generate an initial population to represent different topological structures, and each topological structure represents an individual; design a fitness function to evaluate the fitness of each individual, update and record the current global optimal solution.

[0040] Taking the individuals in the current population as particles, an adaptive weighted particle swarm optimization algorithm is used to iteratively update the particles. The particles record the optimal positions they have experienced during the search process to obtain the current individual optimal solutions of each particle. The updated all individual optimal solutions are used as a new population, and genetic operations are performed to generate a new generation of individuals. The fitness values of the new generation of individuals are calculated and sorted together with the fitness value of the current global optimal solution to obtain a new global optimal solution. Repeat this process until the termination condition is met, and the global optimal solution is obtained and decoded to obtain the optimal topological structure of the collector system.

[0041] In some other embodiments, the following technical solutions are adopted:

[0042] A terminal device includes a processor and a memory. The processor is used to implement instructions; the memory is used to store multiple instructions, and the instructions are adapted to be loaded and executed by the processor to perform the above-mentioned collector system topology optimization method based on the particle swarm genetic hybrid algorithm.

[0043] In some other embodiments, the following technical solutions are adopted:

[0044] A computer-readable storage medium stores multiple instructions, and the instructions are adapted to be loaded and executed by the processor of the terminal device to perform the above-mentioned collector system topology optimization method based on the particle swarm genetic hybrid algorithm.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0046] (1) The present invention comprehensively considers the equipment procurement and construction costs, operation costs, maintenance costs, power outage loss costs, and equipment disposal costs of the collector system during the whole life cycle of the offshore wind farm. A target function is constructed with the lowest whole life cycle cost as the goal. At the same time, the constraint condition of cable intersection is particularly considered. By introducing an intersection detection algorithm to detect and judge whether there is an intersection of the cables, it is possible to avoid the situation of intersection in the planned cable topology, reduce the failure probability caused by physical damage caused by cable intersection, reduce the electromagnetic interference generated by cable intersection, reduce power loss, ensure stable power transmission, and thus improve system reliability; at the same time, it can reduce the maintenance cost and repair complexity caused by cable intersection.

[0047] (2) When solving the model of the present invention, an adaptive particle swarm genetic hybrid algorithm is adopted. The encoding uses a hybrid encoding method. The reciprocal of the square of the life cycle cost is selected as the fitness function to highlight the differences between individuals. The adaptive particle swarm algorithm is combined for dynamic optimization. Genetic operators are designed to increase the population diversity until the algorithm terminates. The decoded optimal solution is the optimal topological structure of the collector system. This algorithm combines the advantages of the genetic algorithm and the adaptive particle swarm algorithm. Moreover, the result obtained by the hybrid encoding method can accurately represent the connection situation and position information between each fan, and can intuitively and clearly represent the topological structure under the economically optimal topological scheme.

[0048] Other features and advantages of the additional aspects of the present invention will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of this aspect. Brief Description of the Drawings

[0049] Figure 1 It is a flow chart of the topological optimization method of the collector system based on the particle swarm genetic hybrid algorithm in the embodiment of the present invention;

[0050] Figure 2 It is a schematic diagram of the solution process of the adaptive particle swarm genetic hybrid algorithm in the embodiment of the present invention;

[0051] Figure 3 It is a schematic diagram of the calculation of condition 1 in the design principle of the genetic selection operator in the embodiment of the present invention;

[0052] Figure 4 It is a schematic diagram of the calculation of condition 2 in the design principle of the genetic selection operator in the embodiment of the present invention;

[0053] Figure 5 It is a distribution map of the fan positions of a wind farm in the embodiment of the present invention;

[0054] Figure 6 It is an optimal topological structure diagram of the fans of a wind farm in the example of the present invention. Detailed Description of the Embodiment

[0055] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs.

[0056] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should also be understood that when the terms "comprise" and / or "include" are used in this specification, they specify the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0057] Embodiment 1

[0058] In one or more embodiments, a topology optimization method for a collector system based on a particle swarm genetic hybrid algorithm is disclosed, which specifically includes the following processes:

[0059] (1) Taking the lowest life cycle cost as the goal, establishing an objective function for the topology optimization of the collector system, and at the same time considering the cable current-carrying capacity constraint, the number of wind turbines connected to the cable, and the cable overlap as constraint conditions, to obtain a mathematical model of the topology structure of the offshore wind farm collector system;

[0060] (2) Using an adaptive particle swarm genetic hybrid algorithm to solve the mathematical model to obtain the optimal topology structure of the collector system.

[0061] Among them, the process of using an adaptive particle swarm genetic hybrid algorithm to solve the mathematical model is specifically as follows:

[0062] Encoding the topology structure of the collector system using a hybrid encoding method, randomly generating an initial population to represent different topology structures, and each topology structure represents an individual; designing a fitness function to evaluate the fitness of each individual, updating and recording the current global optimal solution of the population;

[0063] Taking the individuals in the current population as particles, using an adaptive weighted particle swarm algorithm to iteratively update the particles. The particles record their own optimal positions during the search process to obtain the current individual optimal solution of each particle; taking all the updated individual optimal solutions as a new population, performing genetic operations to generate a new generation of individuals; calculating the fitness values of the new generation of individuals and sorting them together with the fitness value of the current global optimal solution of the population to obtain a new global optimal solution of the population; repeating this process until the termination condition is met, decoding the global optimal solution of the population to obtain the optimal topology structure of the collector system.

[0064] As a specific embodiment, in combination with Figure 1 , the method of this embodiment specifically includes the following processes:

[0065] S1: Starting from the perspective of the life cycle cost, defining the objective function for the topology optimization of the collector system:

[0066] The life of an offshore wind farm is generally 25 years, denoted as Considering its 25-year operation life and various costs during the operation cycle, such as equipment procurement and construction costs in the infrastructure construction stage, operation costs, maintenance costs, power outage loss costs in the operation stage, and equipment disposal costs in the recovery stage, etc.

[0067] S11: Define as the equipment procurement and construction costs in the infrastructure construction stage, and its calculation formula is:

[0068] ;

[0069] Among them, is the total number of different types of submarine cables, is the total number of the i th type of submarine cable, is the length of the i th root of the j th type of submarine cable, is the unit purchase cost of the i th type of submarine cable equipment, is the unit construction cost of the i th type of submarine cable equipment.

[0070] S12: Define as the sum of operation costs , maintenance costs and power outage loss costs in the operation stage, and its calculation formula is:

[0071] ;

[0072] Among them, the operation cost is ; is the working current of the i th type of submarine cable, is the unit length resistance of the i th type of submarine cable, is the on-grid electricity price of offshore wind power, taking 0.75 yuan / , is the annual usage fee per unit sea area of the sea area where this offshore wind farm is located, is the total sea area used by the offshore wind farm.

[0073] Maintenance cost ; is the number of failures of the i th type of submarine cable in one year, is the cost required for single repair of the i th type of submarine cable. Since the failure types are different, the repair costs are also different, and here is represented by the annual expected value.

[0074] The power outage loss cost is , 8760 per year h , is the probability of failure of the i th type of submarine cable, is the i th type of the number of wind turbines connected to the th submarine cable, is the average power of the wind turbine, is the probability of outage of

[0075] S13: Define as the equipment processing cost in the recovery stage, and its calculation formula is:

[0076] ;

[0077] is the cost of ships and manual loading and unloading mainly consumed in the recovery process. Since the materials of equipment such as submarine cable conductors are metals and have a certain recovery value, so is the value of secondary utilization after recovery.

[0078] However, the value of secondary utilization after equipment recovery is not high, and ships and labor costs are also required in the infrastructure construction stage to lay equipment such as submarine cables. The equipment processing cost in the recovery stage of this embodiment can be offset against the ships and labor costs required in the infrastructure construction stage. Therefore

[0079] S14: The objective function of the topology optimization of the collection system is:

[0080] ;

[0081] Among them, is the life cycle cost, and the discount factor , r is the discount rate taken as 8%.

[0082] S2: Considering the implementation method of cable connection in actual operation, define the constraint conditions of the topology optimization problem and establish a mathematical model of the topology structure of the offshore wind farm:

[0083] Considering the actual operation situation, the constraint conditions of the collection system topology optimization mainly include submarine cable current-carrying capacity constraint, the number of wind turbines connected to the submarine cable, submarine cable overlap constraint, etc.

[0084] S21: Considering the submarine cable current-carrying capacity constraint, it can be expressed as:

[0085] ;

[0086] Among them, is the maximum continuous load current flowing through a certain section of submarine cable, is the total correction factor of the allowable current-carrying capacity of the submarine cable, is the current-carrying capacity of the submarine cable.

[0087] S22: When designing the topology of the collection system, due to the limitation of the current-carrying capacity of the submarine cable, the number of wind turbines connected to a single submarine cable is limited. When connecting the submarine cable, it should be ensured that the number of wind turbines connected to a single submarine cable is not more than the maximum number of wind turbines that the submarine cable can connect, which can be expressed as:

[0088] ;

[0089] ;

[0090] Among them, is the number of wind turbines connected to a single submarine cable, is the maximum number of wind turbines that the submarine cable can connect, is the floor function, is the maximum wind turbine capacity that the submarine cable can withstand, is the rated capacity of a single wind turbine.

[0091] S23: In order to avoid the situation where the submarine cables connecting the wind turbines intersect in the topology planning of the submarine cable, in this embodiment, the given wind turbine data is used to detect and judge whether there is an intersection of the submarine cables. Specifically, the positions of the wind turbines are regarded as the coordinates of points in the coordinate system, and the submarine cable connecting two wind turbines is regarded as a line segment, and it is judged whether the two line segments intersect.

[0092] Suppose the coordinates of the 4 points corresponding to 4 wind turbines are respectively , , , , represents the line segment with points and as endpoints, represents the line segment with points and as endpoints.

[0093] Condition 1: Vectors , are respectively on the left and right sides of vector ;

[0094] Condition 2: Vectors , are respectively on the left and right sides of vector ;

[0095] When both Condition 1 and Condition 2 are satisfied, it can be determined that the line segment and the line segment intersect.

[0096] To determine the orientation of one vector relative to another vector, the specific process is as follows:

[0097] Taking the vector and the vector as an example, the cross - product formula for the points , , is: . If the result of the cross - product of the vector and the vector is greater than 0, that is , then the vector is on the left side of the vector ; if , then the vector is on the right side of the vector ; if , then the vector is parallel to the vector . The calculation schematic diagram of Condition 1 is as shown in Figure 3 , and the calculation schematic diagram of Condition 2 is as shown in Figure 4 .

[0098] To determine whether there is an intersection between submarine cables using the fan coordinates, the constraint conditions are:

[0099] ;

[0100] Among them, , represent two submarine cables between four fans, and is the cross - product of the points , , .

[0101] S23: The integrated offshore wind farm collection system aims at the lowest economic index LCC, and takes the submarine cable current - carrying capacity constraint, the number of fans connected by the submarine cable and other conditions as constraints, and establishes the collection system topology optimization model as:

[0102] ;

[0103] .

[0104] S3: Combining Figure 2 , use the hybrid coding method to encode the topology structure of the collection system and randomly generate the initial population:

[0105] In this embodiment, the use of a hybrid coding method can improve the convergence speed of the algorithm, help find the optimal solution faster, enhance the expression ability of the solution, capture more features and information, and thus enhance the search effect.

[0106] S31: The process of coding the topology structure of the collector system using the hybrid coding method is as follows:

[0107] S311: Use binary bits to represent the connection relationship between the collector submarine cable and the wind turbines. 1 represents connection, and 0 represents non-connection. Suppose there are wind turbines and collector submarine cables. The connection method between each wind turbine and the submarine cable corresponds to a binary bit as the connection code, that is, the connection situation of each wind turbine is represented by a binary string with a length of . For example, if there are 5 wind turbines and 3 collector submarine cables, the connection method of wind turbine 1 is 101 (a binary string with a length of 3), indicating that wind turbine 1 is connected to collector submarine cable 1 and collector submarine cable 3, but not connected to collector submarine cable 2.

[0108] S312: Use real values to represent the positions of the wind turbines. Represent the position of each wind turbine with coordinates, that is, use a two-dimensional coordinate system to represent the position of each wind turbine. After coding, the position of each wind turbine is represented by a real number array with a length of 2 as the position code; the first element represents the position of the wind turbine on the x-axis, and the second element represents the position of the wind turbine on the y-axis. For example: if the position coordinates of 3 wind turbines are , , , the real number coding obtained by combining the positions of all wind turbines in order is . This coding method can intuitively represent the position distribution of the wind turbines and make flexible adjustments to the positions during the optimization process.

[0109] S32: Combine the connection code and the position code into an individual. An individual contains both connection information and position information, and randomly generate an initial population to represent different topological structures.

[0110] S4: Design a fitness function to evaluate each individual and update to obtain the current optimal solution:

[0111] The fitness function is a function that evaluates the quality of an individual during the algorithm iteration. The higher the function value of an individual, that is, the higher the fitness score, the higher the quality of the solution, the easier it is to be selected for reproduction, and its traits will appear in the next generation, producing more excellent offspring. As the algorithm iterates, the quality of the solution will improve, and the fitness will increase until the optimal solution is found.

[0112] S41: This embodiment aims to find a topological optimization solution that minimizes the total life cycle cost of the power collection system, i.e., the economically optimal solution. The lower the total life cycle cost of an individual, the higher its fitness, and the higher the quality of the solution corresponding to this individual. There is a negative correlation between the two. Therefore, the reciprocal is taken and then squared. Squaring can make the fitness differences corresponding to different solutions more obvious, facilitating further screening. Thus, the reciprocal of the square of the total life cycle cost is selected as the fitness function, which can better show the differences between individuals and can be expressed as:

[0113] ;

[0114] S42: According to the defined fitness function, calculate the fitness value of each individual, and perform a sorting operation on the fitness values of each individual to obtain the current optimal solution.

[0115] S43: The algorithm termination condition can be set as: when the error of the fitness of each individual is within the 2% error band, the algorithm terminates.

[0116] S44: If the current optimal solution of the algorithm is the optimal solution among the global optimal solution and the individual optimal solution, then determine whether the algorithm meets the termination condition. If not, continue to the next step. If it meets, output the current optimal solution as the global optimal solution to step S7;

[0117] Otherwise, collect the current optimal solution and continue to the next step.

[0118] S5: Utilize the idea of the adaptive particle swarm optimization algorithm to update the velocity and position of each particle and find the optimal solution:

[0119] Using the dynamic update mechanism of the particle swarm algorithm can improve the search efficiency. This embodiment adopts an adaptive particle swarm optimization algorithm. The position of each particle represents a potential solution, i.e., a topological scheme. The individuals in the population in the above steps can be replaced by the positions of the particles in this step. The individual optimal solution mentioned in the above steps enables the particles to perform local fine search around their own previous optimal positions, and the global optimal solution attracts the particles to move in the optimal direction of the entire population, avoiding the particles from being overly trapped in local optima. The continuous update and optimization of the individual optimal solution may discover new and better topological structures, thereby promoting the update of the global optimal solution and finding the global optimal solution.

[0120] The specific process is as follows:

[0121] S51: Adopt an adaptive weighted particle swarm algorithm. The iteration formula of the algorithm is:

[0122] ;

[0123] ;

[0124] Where: is the velocity of the particle, is the position of the particle, is the inertia weight for dynamic adjustment, and are the learning factors, and are random numbers used to increase randomness, is the optimal position of the particle, i.e., the individual optimal solution, is the global optimal position of the particle, i.e., the global optimal solution.

[0125] S511: Initialize the velocity and position of each particle;

[0126] S512: Initialize the parameters, including the inertia weight , the learning factors and , the number of algorithm iterations ;

[0127] S513: In each iteration, calculate the new velocity and new position of the particle;

[0128] S514: Dynamically adjust the inertia weight according to the change of the particle objective function value, to enhance or reduce the search ability.

[0129] S52: Repeat the current step and record the currently obtained individual optimal solution.

[0130] S6: Use the updated all individual optimal solutions as a new population and perform genetic operations to generate a new generation of individuals:

[0131] Regard the positions of all updated particles, i.e., all individual optimal solutions, as a new population. Assume the population size is , and perform selection, crossover, and mutation operations on this population.

[0132] S61: Initialize the crossover probability , , , the mutation probability ;

[0133] S62: Design a selection operator, whose purpose is to directly inherit the optimized individuals to the next generation or generate new individuals through crossover and inherit them to the next generation. Calculate the fitness values corresponding to all individuals in the initialized population, and sort all fitness values in descending order;

[0134] Specifically, the selection operator used in the present invention is the exponential sorting method. The individuals are sorted from the highest to the lowest fitness value, and then a selection probability is assigned through an exponential function, so that individuals with high fitness have a higher probability of being selected, while individuals with low fitness have a certain chance of survival. Select ( ) individuals for reproduction and generation of the next generation, which are called parental individuals.

[0135] S621: Calculate the fitness value of each individual in the new population, sort all the individuals in the population from the highest to the lowest fitness, and the sorted sequence is , where is the individual with the highest fitness in the population.

[0136] S622: Calculate the selection probability using the exponential sorting method. is the fitness value of individual . Let the intermediate variable , is the parameter used to adjust and control the selection pressure. The larger the value, the greater the selection pressure, and the greater the probability that an individual with high fitness is selected. The selection probability corresponding to individual is calculated by the formula . Obviously, the higher the fitness of an individual, the larger its corresponding value, and the larger the proportion it occupies in the sum of corresponding to all individuals. Therefore, the selection probability corresponding to individual is higher.

[0137] S623: Perform the selection operation using the roulette wheel selection method. Denote the cumulative selection probability corresponding to each individual as . Initially, . Generate a random number in the interval [0, 1]. When , individual is selected and copied into the next generation population. Repeat the above steps until parental individuals are selected. Each time an individual is selected and copied into the next generation population, a random number needs to be regenerated to ensure that each individual has a chance to be selected according to its selection probability in each selection round. Each time an individual is selected and copied into the new population, and finally a next generation population containing individuals is obtained.

[0138] S63: Design a crossover operator, the purpose of which is to retain the characteristic information of parental individuals and increase population diversity to promote population evolution;

[0139] The present invention selects an adaptive crossover probability formula, adjusts the crossover probability according to the relative magnitudes of the individual fitness values, so that individuals with high fitness values perform crossover with a lower probability to protect excellent genes, while individuals with lower fitness values perform crossover with a higher probability to increase their improvement opportunities. For the fitness value of an individual, and for the average fitness of the population, the adjustment formula for the adaptive crossover probability is as follows, where are all constants:

[0140] ;

[0141] This crossover operator can retain the relevant characteristics of the parent generation, making the genetic process tend to be optimal.

[0142] S64: Design a mutation operator to randomly change the genes of some individuals, with the aim of introducing new gene information, preventing premature convergence and helping the algorithm jump out of the local optimal solution;

[0143] The present invention adopts the mutation method of basic bit mutation, randomly selects two gene positions in the binary part of an individual, performs an inversion operation on them, changing 0 to 1 and 1 to 0 in the binary encoding.

[0144] S641: Randomly generate a random number within the interval [0, 1] , if , then proceed to the next mutation operation; if , do not perform the mutation operation and jump to step S65.

[0145] S642: The number of wind turbines in the wind farm is h , the length of the binary encoding of an individual is h , randomly select an integer within the interval d as the mutation position, and invert the encoding value at this position. Randomly selecting a certain gene position for flipping in the binary encoding increases the population diversity.

[0146] S65: Return to step S42 to sort the population obtained after the genetic operation, update and record the new group optimal solution, and determine whether the new group optimal solution is the global optimal solution. If so, continue to determine whether the termination condition is satisfied. If the termination condition is satisfied, output the current group optimal solution as the global optimal solution and end the algorithm; otherwise, return to step S5 and continue to execute the algorithm.

[0147] S7: Decode the obtained global optimal solution to obtain the optimal topological structure of the collector system.

[0148] In this embodiment, a 1.5-km square wind farm is taken as an example, with 20 wind turbines installed and an offshore substation. 35-kV submarine cables are selected. The wind turbine distribution is drawn according to the wind turbine positions as Figure 5 shown. Debugging is performed using the adaptive particle swarm genetic hybrid algorithm of the present invention, and the result of the wind turbine connection is:

[0149]

[0150] Decoding it obtains an optimal topological structure under the conditions of current-carrying capacity, number of wind turbine connections, and submarine cable intersection constraints as Figure 6 shown.

[0151] The method of this embodiment takes the life cycle cost as the objective function, considers the submarine cable current-carrying capacity, the number of wind turbines connected by the submarine cable, and the submarine cable intersection constraints, and then considers the topological structure of the collection system and establishes the objective function of the optimization problem. When solving, an adaptive particle swarm genetic hybrid algorithm is introduced to find the optimal parameters. The hybrid coding is used to represent the wind turbine connection situation and position, and a suitable fitness function is selected. With the design of its genetic operator and dynamic update mechanism, the decoded result is not only an economically optimal topological scheme but also can intuitively and clearly represent the topological structure.

[0152] Embodiment 2

[0153] In one or more embodiments, a collection system topology optimization system based on a particle swarm genetic hybrid algorithm is disclosed, including:

[0154] A model construction module, which is used to establish the objective function of the collection system topology optimization with the lowest life cycle cost as the goal, and at the same time consider the submarine cable current-carrying capacity constraint, the number of wind turbines connected by the submarine cable, and the submarine cable overlap as constraints, to obtain the mathematical model of the collection system topology structure of the offshore wind farm;

[0155] A model solving module, which is used to solve the mathematical model using an adaptive particle swarm genetic hybrid algorithm to obtain the optimal topological structure of the collection system;

[0156] Among them, the process of solving the mathematical model using an adaptive particle swarm genetic hybrid algorithm is specifically as follows:

[0157] Encoding the collection system topology structure using a hybrid coding method, randomly generating an initial population to represent different topological structures, and each topological structure represents an individual; designing a fitness function to evaluate the fitness of each individual, updating and recording the current global best solution;

[0158] Taking the individuals in the current population as particles, an adaptive weighted particle swarm optimization algorithm is used to iteratively update the particles. The particles record the optimal positions they have experienced during the search process to obtain the current individual optimal solutions of each particle. The updated all individual optimal solutions are used as a new population, and genetic operations are performed to generate a new generation of individuals. Calculate the fitness values of the new generation of individuals and sort them together with the fitness value of the current global optimal solution of the population to obtain a new global optimal solution of the population. Repeat this process until the termination condition is met, obtain the global optimal solution of the population and decode it, so as to obtain the optimal topological structure of the collector system.

[0159] It should be noted that the specific implementation methods of the above modules are the same as those in Embodiment 1 and will not be elaborated here.

[0160] Embodiment 3

[0161] In one or more embodiments, a terminal device is disclosed, which includes a processor and a memory. The processor is used to implement instructions; the memory is used to store multiple instructions, and the instructions are adapted to be loaded and executed by the processor to perform the topology optimization method of the collector system based on the particle swarm genetic hybrid algorithm described in Embodiment 1.

[0162] It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0163] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0164] In the implementation process, each step of the above method may be completed by the integrated logic circuit in the hardware of the processor or the instructions in the form of software.

[0165] Embodiment 4

[0166] In one or more embodiments, a computer-readable storage medium is disclosed, in which multiple instructions are stored, and the instructions are adapted to be loaded and executed by the processor of the terminal device to perform the topology optimization method of the collector system based on the particle swarm genetic hybrid algorithm described in Embodiment 1.

[0167] Although the specific implementation manners of the present invention have been described above in conjunction with the accompanying drawings, it is not a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications or deformations that can be made without creative efforts on the basis of the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A topology optimization method for a collector system based on a particle swarm genetic hybrid algorithm, characterized in that including: Taking the minimum total life cycle cost as the goal, establish the objective function for the topology optimization of the collection system. At the same time, considering the constraints of the cable current-carrying capacity, the number of wind turbines connected to the cable, and the cable overlap as constraints, obtain the mathematical model of the topology structure of the collection system in the offshore wind farm; The specific cable overlap constraint is: ； Among them, and respectively represent two sections of submarine cables between four fans; Use an adaptive particle swarm genetic hybrid algorithm to solve the mathematical model to obtain the optimal topology structure of the collection system; Among them, the process of using the adaptive particle swarm genetic hybrid algorithm to solve the mathematical model is specifically: Use a hybrid coding method to encode the topology structure of the collection system, randomly generate an initial population to represent different topology structures, and each topology structure represents an individual; design a fitness function to evaluate the fitness of each individual, update and record the current global best solution; Using a hybrid coding method to encode the topology structure of the collection system is specifically: Use binary bits to represent the connection relationship between the collection cable and the wind turbine, 1 represents connection, 0 represents non-connection, and the connection method between each wind turbine and the cable corresponds to a binary bit, which is used as the connection code; Use real values to represent the positions of the wind turbines, represent the position of each wind turbine with two-dimensional coordinates, and the real number code obtained by combining the positions of all wind turbines in sequence is used as the position code; Combine the connection code and the position code into an individual to represent a topology structure of the collection system; Take the individuals in the current population as particles, and use an adaptive weighted particle swarm algorithm to iteratively update the particles. The particles record their own best positions during the search process to obtain the current individual best solution of each particle; use the updated global best solutions of all individuals as a new population, perform genetic operations to generate a new generation of individuals; calculate the fitness values of the new generation of individuals and sort them together with the fitness value of the current global best solution to obtain a new global best solution; repeat this process until the termination condition is met, obtain the global best solution and decode it to obtain the optimal topology structure of the collection system; Using the idea of the adaptive particle swarm algorithm, update the velocity and position of each particle to find the optimal solution. The specific process includes: Initialize the velocity and position of each particle; Initialize the parameters, including the inertia weight, learning factor, and the number of algorithm iterations; In each iteration, calculate the new velocity and new position of the particle; According to the change of the particle objective function value, dynamically adjust the inertia weight to enhance or reduce the search ability; Repeat the current step and record the current individual best solution obtained; Use the updated global best solutions of all individuals as a new population, perform genetic operations to generate a new generation of individuals; Design a crossover operator to retain the characteristic information of the parent individuals and increase the population diversity to promote the evolution of the population; Design a mutation operator to randomly change the genes of some individuals; Design a fitness function to evaluate the fitness of each individual. The specific fitness function is: ； Among them, is the life cycle cost.

2. The topology optimization method of a collector system based on a particle swarm genetic hybrid algorithm according to claim 1, characterized in that Taking the minimum total life cycle cost as the goal, establish the objective function for the topology optimization of the collection system, specifically: ； Among them, is the life cycle cost, is the equipment procurement and construction cost in the infrastructure stage, is the operation cost in the operation stage , maintenance cost and power outage loss cost sum, is the equipment disposal cost in the recovery stage, R is the discount factor.

3. A topology optimization method for a collector system based on a particle swarm genetic hybrid algorithm according to claim 2, characterized in that, The specific cable current-carrying capacity constraint is: ； Among them, is the maximum continuous load current flowing through a certain section of the submarine cable, is the total correction factor of the allowable current-carrying capacity of the submarine cable, is the current-carrying capacity of the submarine cable.

4. The topology optimization method of a collector system based on a particle swarm genetic hybrid algorithm according to claim 2, wherein The specific constraint on the number of wind turbines connected to the cable is: ； ； Among them, is the number of wind turbines connected by a single submarine cable, is the maximum number of wind turbines that the submarine cable can connect, is the floor function, is the maximum wind turbine capacity that the submarine cable can withstand, is the rated capacity of a single wind turbine.

5. A topology optimization system for a collector system based on a particle swarm genetic hybrid algorithm, characterized in that, including: A model construction module is used to establish an objective function for the topological optimization of the collector system with the goal of minimizing the life-cycle cost. At the same time, considering the constraints of the cable current-carrying capacity, the number of wind turbines connected to the cable, and the cable overlap as constraint conditions, a mathematical model of the topological structure of the offshore wind farm collector system is obtained; The specific cable overlap constraint is as follows: ； Among them, and respectively represent two sections of submarine cables between four fans; A model solving module is used to solve the mathematical model using an adaptive particle swarm genetic hybrid algorithm to obtain the optimal topological structure of the collector system; Among them, the process of solving the mathematical model using an adaptive particle swarm genetic hybrid algorithm is specifically as follows: Use a hybrid coding method to encode the topological structure of the collector system, randomly generate an initial population to represent different topological structures, and each topological structure represents an individual; design a fitness function to evaluate the fitness of each individual, update and record the current global best solution of the population; Using a hybrid coding method to encode the topological structure of the collector system is specifically as follows: Use binary bits to represent the connection relationship between the collector cable and the wind turbine, 1 means connected, 0 means not connected, and the connection method between each wind turbine and the cable corresponds to a binary bit, which is used as the connection coding; Use real values to represent the positions of the wind turbines, represent the position of each wind turbine with two-dimensional coordinates, and the real number coding obtained by combining the positions of all wind turbines in order is used as the position coding; Combine the connection coding and the position coding into an individual, representing a topological structure of the collector system; Use the individuals in the current population as particles, and use an adaptive weighted particle swarm algorithm to iteratively update the particles. The particles record their own best positions during the search process to obtain the current individual best solution of each particle; use the updated individual best solutions of all individuals as a new population, and perform genetic operations to generate a new generation of individuals; calculate the fitness values of the new generation of individuals and sort them together with the fitness value of the current global best solution of the population to obtain a new global best solution; repeat this process until the termination condition is met, obtain the global best solution of the population and decode it, so as to obtain the optimal topological structure of the collector system; Using the idea of the adaptive particle swarm algorithm, update the velocity and position of each particle to find the optimal solution. The specific process includes: Initialize the velocity and position of each particle; Initialize parameters, including the inertia weight, learning factor, and the number of algorithm iterations; In each iteration, calculate the new velocity and new position of the particle; According to the change of the particle objective function value, dynamically adjust the inertia weight to enhance or reduce the search ability; Repeat the current step and record the current individual best solution obtained; Use the updated individual best solutions of all individuals as a new population, and perform genetic operations to generate a new generation of individuals; Design a crossover operator to retain the characteristic information of the parent individuals and increase the population diversity to promote the evolution of the population; Design a mutation operator to randomly change the genes of some individuals; Design a fitness function to evaluate the fitness of each individual. The specific fitness function is as follows: ； Among them, is the life cycle cost.

6. A terminal device, which includes a processor and a memory, where the processor is used to implement instructions; the memory is used to store multiple instructions, and is characterized in that, The instructions are suitable for being loaded and executed by a processor for the collector system topological optimization method based on the particle swarm genetic hybrid algorithm according to any one of claims 1-4.

7. A computer-readable storage medium storing multiple instructions, characterized in that, The instructions are adapted to be loaded and executed by a processor of a terminal device for the topology optimization method of a collector system based on a particle swarm genetic hybrid algorithm according to any one of claims 1-4.

Images