Offshore wind plant cable topology connection optimization method considering cable trench sharing

Abstract

The invention discloses an offshore wind power plant cable topology connection optimization method considering cable trench sharing, and solves the problems that the cost is increased due to the fact that a non-feasible solution exists in a cross algorithm under the existing large unit investment condition, and a cable topology optimization method considering multiple cables in the same cable trench does not exist. The method comprises the following steps: randomly generating a same-trench cable topology tree structure; constructing a same-trench cable topological structure model; calculating the cost of each same-trench cable topological structure model; and by taking the minimum cost as a target, optimizing each model by adopting a particle swarm optimization algorithm to obtain an optimal cable topological structure. According to the model construction method for laying the cables in the same trench, the optimal cable topology connection structure is obtained through optimization calculation of the model, the rationality of laying the cables in the same trench is guaranteed, intersection is avoided, and meanwhile the cost is minimized. The model accurately reflects the condition of laying the cables in the same trench, the construction method is simple and easy to implement, and a calculation basis is provided for subsequent optimization.

Description

Technical Field

[0001] The invention relates to the technical field of offshore wind farm design, and in particular to an offshore wind farm cable topology connection optimization method considering cable trench sharing. Background Art

[0002] With the development trend of offshore wind farms in large scale and offshore, wind turbines have now expanded from the early 4MW and 7MW to the current 16MW, and recent research has moved towards 20MW. This has caused a problem. The technology of cables is relatively mature, so the current current carrying capacity of cables is basically unchangeable due to the limitations of insulation materials and other reasons. This means that the thickest cable that could carry the energy transmitted by 7 wind turbines can now only carry 2-3 units, which has caused the problem of cable crossing. The crossing of cables will cause a significant increase in installation costs, harm the operation of cables, and affect the conditions for cable operation. Therefore, it should be avoided as much as possible. In previous studies, the number of cables outgoing from offshore substations is generally limited, so the crossing problem can be avoided by algorithms. However, after the commissioning of large units, this situation will form an infeasible solution through the original algorithm, which will also increase costs. However, current work all assumes that there can only be one cable on a path, and does not consider the cable topology optimization method for the existence of multiple cables in the same cable trench. Summary of the invention

[0003] The present invention mainly solves the problems of avoiding the existence of infeasible solutions in the crossover algorithm under the condition of the existing large-scale unit commissioning, resulting in increased costs, and the lack of a cable topology optimization method considering multiple cables in the same cable trench. It provides an offshore wind farm cable topology connection optimization method considering cable trench sharing.

[0004] The above technical problem of the present invention is mainly solved by the following technical solution: A method for optimizing cable topology connection in an offshore wind farm considering cable trench sharing, comprising the following steps: Randomly generate the topological tree structure of the same-channel cables according to the wind turbine nodes; Based on the generation process, the tree structure edges, the weight of each edge and the number of wind turbines carried by each edge are obtained to build a topological structure model of the same-channel cable; Calculate the cost of each co-channel cable topology model; With the goal of minimizing cost, the particle swarm optimization algorithm is used to optimize each model, and the cable topology corresponding to the optimal solution is obtained as the optimal cable topology.

[0005] The present invention proposes a model construction method for laying cables in the same trench. By optimizing the model, the optimal cable topology connection structure is obtained, which not only ensures the rationality of laying cables in the same trench and avoids crossing, but also minimizes the cost and saves installation costs. The constructed topological structure model of cables in the same trench can accurately reflect the situation of laying cables in the same trench. The construction method is simple and easy to implement, which provides a calculation basis for subsequent optimization.

[0006] As a preferred solution, Establishing a first optimization variable and a second optimization variable according to a generation process of a topological tree structure of cables in the same trench; The first optimization variable represents the current node number and the node number selected by the cable selection in each step of the cable connection process; The second optimization variable represents whether the cables are laid in the same trench along the same path when the same node number is selected in each step of the cable connection process.

[0007] In this scheme, the first optimization variable and the second optimization variable jointly represent the selection of each cable connection step in the cable connection step, including the selection of the connection line and the selection of the assumed cable in the same trench. All the first optimization variables and the second optimization variables jointly describe the process of generating the cable topology tree structure. n Indicates the current node number and the selected node number of the cable connection step in the nth step, which contains the current node number and the selected node number information. The first optimization variable x n is an integer variable. The second optimization variable y n Indicates whether the cables are laid in the same trench when the same node number is selected in the nth step of the cable connection step. The second optimization variable y n It is a binary variable, where 0 means there is no hypothetical cable in the same trench and 1 means the cable is installed in the same trench.

[0008] As a preferred solution, the wind turbine nodes include actual nodes and virtual nodes. The actual nodes respectively represent the wind turbine positions, and the actual nodes are numbered according to a set order; the virtual nodes respectively represent the nodes at the same positions of the same trench cables, and the virtual nodes are numbered according to a set order based on the actual node numbers; the number of actual nodes is consistent with that of virtual nodes, and the positions correspond one to one.

[0009] In this solution, actual nodes are used to represent the positions of wind turbines, and the actual nodes are numbered in sequence. Virtual nodes are set corresponding to actual nodes, and each actual node is equipped with a virtual node, which represents the node connected to the same position of the cable in the same trench. The virtual nodes are numbered in sequence after the actual node numbers.

[0010] As a preferred solution, the steps of constructing the same-channel cable topology model include: S21. Set a first set, a second set, a third set and a fourth set, and a distance neighbor matrix; The first set contains the nodes that have been visited so far. The second set contains nodes that have not yet been visited. The third set contains the weights of the edges of the currently formed tree structure. The fourth set includes the number of wind turbines carried by all edges of the tree structure. Distance proximity matrix, a two-dimensional matrix containing the mutual distance information between all nodes; S22. Obtain the first variable and the second variable of the current co-channel cable topology tree structure; S23. Generate a tree structure starting from the substation location, delete the current node from the second set and add it to the first set; S24. According to the first variable of the current connection process, the selection node number is obtained, the selection node is deleted from the second set and added to the first set, and the weight of the corresponding edge is found from the distance proximity matrix according to the current node and the selection node number, the weight is the distance between the two nodes, and the weight of the edge is added to the third set, and the number of wind turbines carried by the currently connected edge is cumulatively analyzed; S25. According to the second variable of the current connection process, information is obtained on whether the connection needs to be laid with cables in the same trench. If the cables need to be laid in the same trench, the node numbers of the two ends of the edge connected in the same trench are obtained according to the edge of the connection, and the two end nodes are deleted from the second set and added to the first set. At the same time, the weight of the edge connected in the same trench is obtained, and the weight of the edge is added to the third set. The number of wind turbines carried by the edge of the current connection is cumulatively analyzed; S26. Enter the next connection process, repeat S24-S25 until all connection processes are completed, and obtain the first set, the third set and the fourth set as the current co-channel cable topology structure model. This scheme establishes each set and the distance neighbor matrix, and determines the current node, the selected node of the cable connection, the edge between the nodes and the edge of the same channel, and the weight corresponding to the edge and the number of wind turbines according to each step of the cable connection process in the generation process of the co-channel cable topology tree structure. These data representing the cable topology structure are stored in the corresponding set, and the first set, the third set and the fourth set finally obtained construct the current co-channel cable topology structure model. According to the steps of this scheme, the co-channel cable topology structure model is constructed for each co-channel cable topology tree structure. At the beginning of the construction, the first set, the third set and the fourth set are combined into an empty set. The second set stores the information of all nodes, including actual nodes and virtual nodes. The generation of the tree structure starts from the substation location until all cable connection processes are completed, the second set becomes an empty set, and the construction of the co-channel cable topology structure model is completed.

[0011] As a preferred solution, determine whether the end node number of the current edge is not greater than N, where N is the total number of fans. If so, add 1 to the number of fans carried by the current edge in the fourth set, obtain all the preceding edges connected to the trunk of the edge in the fourth set, and add 1 to the number of fans carried by these edges; if not, no operation is performed in the fourth set.

[0012] In this scheme, the current edge refers to the edge for which the cumulative analysis of the number of fans it carries is performed. Whenever a new edge is connected and the number of fans it carries increases by 1, the number of fans carried by the edge connected to the trunk of the edge also increases accordingly. Then, all the previous edges connected to the trunk of the edge are obtained, and the number of fans carried by each of them is increased by 1.

[0013] In the present invention, the wind turbine node numbering sets the starting substation position to 0, the actual nodes are numbered in sequence, the virtual node numbering starts from the actual node number and is numbered in sequence. The actual node is the wind turbine node. When the node number is not greater than the number of wind turbine nodes, the node is the actual node. When the node number is greater than the number of wind turbine nodes, the node is the virtual node. The edge connected to the actual node needs to increase the number of wind turbines it carries by 1, while the edge connected to the virtual node does not need to increase the number of wind turbines it carries by 1.

[0014] As a preferred solution, Set multiple levels of cable carrying capacity and the corresponding cross-sectional area of ​​each level of cable; According to the number of wind turbines carried by each edge in the same-channel cable topology model, cables with corresponding levels of cable carrying capacity are configured.

[0015] This solution pre-sets multiple levels of cable carrying capacity, which is the number of wind turbines that the cable can carry. At the same time, the cross-sectional area corresponding to the cable of this level is set, and the optimal cable cross-sectional area is configured based on the number of wind turbines carried by the tree structure edge.

[0016] As a preferred solution, the number of wind turbines carried by each edge in the fourth set is compared with the cable carrying capacity to determine whether the number of wind turbines carried is greater than the cable carrying capacity. If so, a penalty value is added to the corresponding edge.

[0017] After establishing the topological structure model of the same-channel cable, this solution makes a reasonable judgment on the number of wind turbines carried by each side in the model. For a certain type of cable, the maximum number of wind turbines that it can carry is certain. If the number of wind turbines carried by a certain side in the fourth set of the model is greater than the carrying capacity of the cable, it means that the cable topology is unreasonable, and a penalty value is added to the side. The penalty value is a larger cable cost value, that is, a larger cost is set for the cable corresponding to the side, so that the model can be excluded when optimizing each model with the minimum cost as the goal in the future.

[0018] As a preferred solution, the cost of calculating each co-channel cable topology model includes: The cost of the cable topology is calculated based on the sum of the installation costs of all cables between nodes, the cable investment, and the economic efficiency of the overall loss of the cables over their entire life cycle.

[0019] This scheme is expressed by the formula: in, represents the installation cost of the cable between fan i and fan j, which is linearly related to the length of the cable. represents the investment in cables, It represents the economic efficiency of the overall cable loss in the whole life cycle (assuming that the operation cycle of the wind farm is 20 years), calculated according to the current price of 0.405 yuan / kWh, i, j represent the numbers of the wind turbines, k represents the cross-sectional area of ​​the selected cable, when different cross-sectional area cables are selected, even if the connection path is the same, the cost and loss of the cable will change. In addition, the present invention assumes that the installation cost of laying cables in the same trench is It is 70% of the original.

[0020] As a preferred solution, the particle swarm optimization algorithm is used to optimize each model, including: S41. Initialize the values ​​of the first vector and the second vector as the position of the particle, and randomly generate the velocity of the particle; S42. Calculate the fitness of all particles in the population; Obtain the cost of each co-channel cable topology model, which is the fitness; S43. Compare the fitness values ​​of each particle position, and obtain the global optimal particle and the local optimal particle by comparison; S44. Update the velocity and position of the particle; S45. According to the updated speed and position of the particle, the cost of each co-channel cable topology model is recalculated. If the fitness of the new particle is better than the original one, the particle is updated to be the local optimal particle. If the fitness of the new particle is better than the global optimal particle, the particle is updated to be the global optimal particle. S46. Repeat S42-S45 until the maximum number of iterations is reached, and output the final result, which is the optimal cable topology.

[0021] As a preferred solution, a topological tree structure of cables in the same trench is randomly generated according to wind turbine nodes, and the generation process includes: S11. Decision making starts from the substation; S12. Randomly select wind turbine nodes for connection; S13. Randomly set the cable connection situation in the same trench; S14. Make a decision based on the current selectable node, and repeat S12-S13 until all hierarchical nodes are connected.

[0022] Therefore, the advantages of the present invention are: a model construction method for laying cables in the same trench is proposed, and the optimal cable topology connection structure is obtained by optimizing the model, which not only ensures the rationality of laying cables in the same trench and avoids crossing, but also minimizes the cost and saves installation costs. The constructed topological structure model of cables in the same trench can accurately reflect the situation of laying cables in the same trench, and the construction method is simple and easy to implement, which provides a calculation basis for subsequent optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of the steps of constructing a topological structure model of a same-channel cable in the present invention.

[0024] Figure 2 It is a structural schematic diagram of a randomly generated same-channel cable topology tree structure in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The technical solution of the present invention is further specifically described below through embodiments and in conjunction with the accompanying drawings.

[0026] Example: This embodiment provides a method for optimizing cable topology connection in an offshore wind farm considering cable trench sharing, including the following steps: S1. Randomly generate a topological tree structure of cables in the same trench according to wind turbine nodes; S2. Based on the generation process, a set of tree structure edges, weights of each edge and the number of wind turbines carried by each edge is obtained to construct a topological structure model of the same-channel cable; S3. Calculate the cost of each co-channel cable topology model; S4. Taking the minimum cost as the goal, the particle swarm optimization algorithm is used to optimize each model, and the cable topology structure corresponding to the optimal solution is obtained as the optimal cable topology structure.

[0027] The present invention proposes a model construction method for laying cables in the same trench. By optimizing the model, the optimal cable topology connection structure is obtained, which not only ensures the rationality of laying cables in the same trench and avoids crossing, but also minimizes the cost and saves installation costs. The constructed topological structure model of cables in the same trench can accurately reflect the situation of laying cables in the same trench. The construction method is simple and easy to implement, which provides a calculation basis for subsequent optimization.

[0028] As a preferred solution of this embodiment, the process of randomly generating a topological tree structure of cables in the same trench in step S1 includes: S11. Decision making starts from the substation; S12. Randomly select wind turbine nodes for connection; S13. Randomly set the cable connection situation in the same trench; S14. Make a decision based on the current selectable node, and repeat S12-S13 until all hierarchical nodes are connected.

[0029] The wind turbine nodes include actual nodes and virtual nodes. The actual nodes represent the wind turbine positions respectively, and the actual nodes are numbered according to the set order; the virtual nodes represent the nodes at the same position of the same trench cable respectively, and the virtual nodes are numbered according to the set order following the actual node number; the number of actual nodes is the same as that of virtual nodes, and the positions correspond one to one. Each actual node is equipped with a virtual node, and the number of the virtual node is numbered after the actual node number in the same order.

[0030] This embodiment uses a specific example to illustrate step S1. Figure 2 As shown in the figure, the square block is the offshore substation, 0 represents its node number, the solid circle represents the wind turbine node position, 1-6 numbers represent the node number of the wind turbine position, 7-12 represent the virtual node number of the same trench cable at the same position, and the hollow circle represents the connected virtual node, where the solid line, dotted line, and dashed line represent different cables respectively. The generation process is to form a tree structure to connect all wind turbines to the offshore substation. The specific formation process is as follows: 1) Decision making starts from 0, that is, the substation selects a wind turbine to connect to. Assume that wind turbine No. 1 is randomly selected at this time; 2) Whether it is necessary to bury another cable in the same trench along the same path (0-1), assuming the random choice is yes, lay the cable in the same trench (0-7); 3) Nodes 0, 1, and 7 make decisions on which node to connect to which wind turbine. Assume that node 7 is randomly selected to connect to wind turbine 3. 4) Similarly, decide whether the same path (7-3) needs to bury another cable in the same trench. Assuming the random choice is yes, lay the cable in the same trench (1-9); 5) Nodes 0, 1, 3, 7, and 9 make decisions on which node to connect to which wind turbine. Assume that node 1 is randomly selected to connect to wind turbine 2. 6) Similarly, decide whether to bury another cable in the same ditch along the same path (1-2). Assume that the random choice is no; 7) Nodes 0, 2, 3, and 9 make decisions on which node to connect to which wind turbine. Assume that node 3 is randomly selected to connect to wind turbine 5. 8) Similarly, decide whether to bury another cable in the same ditch along the same path (3-5). Assume that the random choice is no; 9) Nodes 0, 2, 3, 5, and 9 make decisions on which node to connect to which wind turbine. Assume that node 9 is randomly selected to connect to wind turbine 6. 10) Similarly, decide whether to bury another cable in the same ditch along the same path (9-6). Assume that the random choice is no; 11) Six nodes 0, 2, 3, 5, 6, and 9 make decisions on which node to connect to which wind turbine. Assume that node 3 is randomly selected to connect to wind turbine 4. At this point, all levels are connected and the generation process is complete.

[0031] In the process of generating the topological tree structure of the same-channel cable, a first optimization variable and a second optimization variable representing the selection of each cable connection step in the cable connection step are simultaneously established. The first optimization variable represents the current cable number and the cable connection node number in each step of the cable connection process, and the second optimization variable represents whether the cables are installed in the same trench in the same path when the same node number is selected in each step of the cable connection process. Specifically, the first optimization variable and the second optimization variable are x n and n To represent, where n represents the nth step of the cable connection process, the first optimization variable x n is an integer variable, including the current node number and the selected node number, the second optimization variable y n is a binary variable, where 0 indicates that there is no assumed cable in the same trench, and 1 indicates that the cables are installed in the same trench. All the first optimization variables and the second optimization variables describe the selection of each step in the cable topology tree structure generation process.

[0032] As a preferred solution of this embodiment, step S2 is based on the generation process, and obtains the tree structure edge, the weight of each edge and the number of wind turbines carried by each edge to construct a co-channel cable topology model, such as Figure 1 As shown, the specific process includes: S21. Setting a first set, a second set, a third set and a fourth set, and a distance neighbor matrix; The first set contains the nodes that have been visited so far. The second set contains nodes that have not yet been visited. The third set contains the weights of the edges of the currently formed tree structure. The fourth set includes the number of wind turbines carried by all edges of the tree structure. Distance proximity matrix, a two-dimensional matrix containing the mutual distance information between all nodes; S22. Get the first variable x of the current co-channel cable topology tree structure n and the second variable y n ; S23. At the beginning of the process, the first set, the third set and the fourth set are set to empty sets, and the second set stores the information of all actual nodes. The tree structure is generated starting from the substation location, and the current node is deleted from the second set and added to the first set; S24. According to the first variable x of the current connection process n , obtain the node number selected by the current node, delete the selected node from the second set and add it to the first set, find the weight of the corresponding edge from the distance proximity matrix according to the actual node number of the current node and the selected node, the weight is the distance between the two nodes, add the weight of the edge to the third set, and perform cumulative analysis on the number of wind turbines carried by the edge; S25. According to the second variable y of the current connection process n , obtain information on whether this section of connection needs to lay cables in the same trench. If cables need to be laid in the same trench, obtain the node numbers at both ends of the edge connected in the same trench according to the edge of this section of connection, delete the nodes at both ends from the second set and add them to the first set, and at the same time obtain the weight of the edge connected in the same trench, add the weight of the edge to the third set, and perform cumulative analysis on the number of wind turbines carried by the edge; If it is not necessary to lay cables in the same trench, do not perform the operation; S26. Enter the next connection process, repeat S24-S25 until all connection processes are completed, and obtain the first set, the third set and the fourth set as the current co-channel cable topology model.

[0033] In the above steps S24 and S25, a cumulative analysis of the number of wind turbines carried by the edge is performed, which specifically includes the following process: determine whether the node number at the end of the current edge is not greater than N, where N is the total number of wind turbines; if so, add 1 to the number of wind turbines carried by the current edge in the fourth set, obtain all the preceding edges connected to the trunk of the edge in the fourth set, and add 1 to the number of wind turbines carried by these edges; if not, do not perform any operation.

[0034] In the process of constructing the same-channel cable topology structure model, when the number of wind turbines carried by the corresponding edge in the fourth set is increased by 1, the number of wind turbines carried by other edges connected to this edge must be superimposed. Specifically, when a new edge is added and the number of wind turbines carried by this edge is increased by 1, all the previous edges connected to the trunk of this edge in the fourth set are obtained, and the number of wind turbines carried by these edges is increased by 1.

[0035] In this embodiment, the wind turbine node numbering sets the starting substation position to 0, the actual nodes are numbered in sequence, the virtual node numbering starts from the actual node number and is numbered in sequence. The actual node is the wind turbine node. When the node number is not greater than the number of wind turbine nodes, the node is the actual node. When the node number is greater than the number of wind turbine nodes, the node is the virtual node. The edge connected to the actual node needs to increase the number of wind turbines it carries by 1, while the edge connected to the virtual node does not need to increase the number of wind turbines it carries by 1.

[0036] Set multiple levels of cable carrying capacity and the corresponding cross-sectional area of ​​each level of cable; the cable carrying capacity refers to the number of fans that the cable can carry, and the corresponding cross-sectional area of ​​the cable at this level is set at the same time.

[0037] After the same-ditch cable topology structure model is constructed, the fourth set of data is analyzed, and cables with corresponding levels of cable carrying capacity are configured according to the number of wind turbines carried by each edge in the same-ditch cable topology structure model, that is, cables with optimal cable cross-sectional area are configured.

[0038] During the configuration process, the rationality of the number of wind turbines carried by each side is judged at the same time. Specifically, the number of wind turbines carried by each side in the fourth set of the same-channel cable topology model is compared with the cable carrying capacity to determine whether the number of wind turbines carried is greater than the cable carrying capacity. If so, a penalty value is added to the corresponding side. For a certain type of cable, the maximum number of wind turbines it can carry is certain. If the number of wind turbines carried by an edge in the fourth set of the model is greater than the carrying capacity of all levels of cables, it means that the cable topology is unreasonable, and a penalty value is added to the edge. The penalty value is a larger cable cost value, that is, a larger cost is set for the cable corresponding to the edge, so that the model can be excluded when the models are optimized with the minimum cost as the goal in the subsequent optimization.

[0039] Specifically, this embodiment establishes each set and distance neighbor matrix, determines the accessed nodes, the edges between nodes and the edges erected in the same trench, and the weights and wind turbine loads corresponding to the edges according to each step of the cable connection process in the process of generating the same trench cable topology tree structure, and stores these data characterizing the cable topology structure in the corresponding set, and finally obtains the first set, the third set and the fourth set to construct the current same trench cable topology structure model. According to the above steps, the same trench cable topology structure model is constructed for each same trench cable topology tree structure. At the beginning of the construction, the first set, the third set and the fourth set are combined into an empty set, and the second set stores the information of all actual nodes. The generation of the tree structure starts with the substation location as the starting point until all cable connection processes are completed, the second set becomes an empty set, and the construction of the same trench cable topology structure model is completed.

[0040] As a preferred solution of this embodiment, the calculation of the cost of each co-channel cable topology structure model in step S3 specifically includes: The cost of the cable topology is calculated based on the sum of the installation costs of all cables between nodes, the cable investment, and the economic efficiency of the overall loss of the cables over their entire life cycle.

[0041] The formula is as follows: in, represents the installation cost of the cable between fan i and fan j, which is linearly related to the length of the cable. represents the investment in cables, It represents the economic efficiency of the overall cable loss in the whole life cycle (assuming that the operation cycle of the wind farm is 20 years), calculated according to the current price of 0.405 yuan / kWh, i, j represent the numbers of the wind turbines, k represents the cross-sectional area of ​​the selected cable, when different cross-sectional area cables are selected, even if the connection path is the same, the cost and loss of the cable will change. In addition, the present invention assumes that the installation cost of laying cables in the same trench is It is 70% of the original. The optimal cable cross-sectional area can be obtained for each cable connection segment according to the cable level it is configured for, and the length of each cable connection segment can be obtained according to the model, and the cable investment can be calculated. In addition, the cable loss is calculated using Ohm's law.

[0042] As a preferred solution of this embodiment, step S4 specifically includes the following steps: S41. Initialize the values ​​of the first vector and the second vector as the position of the particle, and randomly generate the velocity of the particle; S42. Calculate the fitness of all particles in population j; Obtain the cost of each co-channel cable topology model, which is the fitness; S43. Compare the fitness values ​​of each particle position, and obtain the global optimal particle Qg and the local optimal particle Qi by comparison; S44. Update the velocity and position of the particle; it can be expressed as: in is the velocity of the ith particle at time t, is the local optimal position of the current particle, is the global optimal position of all particles, that is, the global optimal value, r1 and r2 are random numbers between [0,1], w, l1, l2 are the total inertia, local learning factor, and global learning factor respectively.

[0043] S45. According to the updated speed and position of the particle, the cost of each co-channel cable topology model is recalculated. If the fitness of the new particle is better than the original one, the particle is updated to be the local optimal particle. If the fitness of the new particle is better than the global optimal particle, the particle is updated to be the global optimal particle. S46. Repeat S42-S45 until the maximum number of iterations is reached, and output the final result, which is the optimal cable topology.

[0044] This embodiment also includes a cable topology connection optimization system for an offshore wind farm that takes into account cable trench sharing, which specifically includes a topology structure random generation module, a model construction module, a cost calculation module, and an optimization module. The topology structure random generation module is connected to the model construction module, the model construction module is connected to the calculation module, and the calculation module is connected to the optimization module.

[0045] Random topology generation module: Randomly generate the topology tree structure of the same-channel cable according to the wind turbine nodes. Among them, the wind turbine nodes include actual nodes and virtual nodes. The actual nodes represent the positions of the wind turbines respectively, and the actual nodes are numbered according to the set order; the virtual nodes represent the nodes at the same position of the same-channel cable respectively, and the virtual nodes are numbered according to the set order according to the actual node numbering; the number of actual nodes is the same as that of virtual nodes, and the positions correspond one to one.

[0046] Randomly generate the topological tree structure of the same-channel cable. The generation process includes: (1) Decision-making starts at the substation; (2) Randomly select wind turbine nodes for connection; (3) Randomly set the cable connection situation in the same trench; (4) Make a decision based on the current selectable node and repeat (2)-(3) until all hierarchical nodes are connected.

[0047] The first optimization variable and the second optimization variable are established according to the generation process of the topological tree structure of the cables in the same trench; the first optimization variable represents the cable connection node number selected in each step of the cable connection process; the second optimization variable represents whether the cables are installed in the same trench in the same path when the same node number is selected in each step of the cable connection process.

[0048] Model building module: Based on the generation process, the tree structure edges, the weights of each edge and the number of wind turbines carried by each edge are obtained to build a co-channel cable topology model. The construction process is to establish each set and distance neighbor matrix, determine the accessed nodes, the edges between nodes and the edges erected in the same channel, as well as the weights and the number of wind turbines carried by the edges according to each step of the cable connection process in the co-channel cable topology tree structure generation process, store these data representing the cable topology structure in the corresponding set, and finally obtain the first set, the third set and the fourth set to construct the current co-channel cable topology model.

[0049] After the model is constructed, the optimal cable cross-sectional area is configured for each edge of the model, including pre-setting multiple levels of cable carrying capacity. The cable carrying capacity is the number of wind turbines that the cable can carry, and the corresponding cross-sectional area of ​​the cable at this level is set. According to the number of wind turbines carried by each edge in the same-trench cable topology model, cables with corresponding levels of cable carrying capacity are configured.

[0050] Cost calculation module: Calculate the cost of each cable topology model in the same trench. The total cost of the cable topology is the sum of the installation cost of all cables between nodes, the investment in cables, and the economic efficiency of the overall loss of cables during their entire life cycle. The specific expression is: Optimization module: With the goal of minimizing cost, the particle swarm optimization algorithm is used to optimize each model, and the cable topology corresponding to the optimal solution is obtained as the optimal cable topology.

[0051] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, but they will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

[0052] Although the terms such as the same-channel cable topology tree structure, the same-channel cable topology structure model, the first optimization variable, the second optimization variable, the actual node, and the virtual node are used more frequently in this article, the possibility of using other terms is not excluded. The use of these terms is only for the purpose of more conveniently describing and explaining the essence of the present invention; interpreting them as any additional limitation is contrary to the spirit of the present invention.

Claims

1. A cable topology connection optimization method for offshore wind farms considering cable trench sharing, characterized in that: The following steps are involved: Randomly generate the topological tree structure of the same-channel cables according to the wind turbine nodes; Based on the generation process, the tree structure edges, the weight of each edge and the number of wind turbines carried by each edge are obtained to build a topological structure model of the same-channel cable; Calculate the cost of each co-channel cable topology model; With the goal of minimizing cost, the particle swarm optimization algorithm is used to optimize each model, and the cable topology corresponding to the optimal solution is obtained as the optimal cable topology.

2. According to claim 1, a cable topology connection optimization method for an offshore wind farm considering cable trench sharing is characterized by: Establishing a first optimization variable and a second optimization variable according to a generation process of a topological tree structure of cables in the same trench; The first optimization variable represents the current node number and the node number selected for cable connection in each step of the cable connection process; The second optimization variable represents whether the cables are laid in the same trench along the same path when the same node number is selected in each step of the cable connection process.

3. The cable topology connection optimization method for offshore wind farms considering cable trench sharing according to claim 2 is characterized by: The wind turbine nodes include actual nodes and virtual nodes. The actual nodes represent wind turbine locations and are numbered in a set order. The virtual nodes represent nodes at the same location of the same trench cable and are numbered in a set order based on the actual node numbers. The number of actual nodes is consistent with that of virtual nodes, and their positions correspond one to one.

4. The cable topology connection optimization method for offshore wind farms considering cable trench sharing according to claim 3 is characterized by: The steps to construct the same-channel cable topology model include: S21. Set a first set, a second set, a third set and a fourth set, and a distance neighbor matrix; The first set contains the nodes that have been visited so far. The second set contains nodes that have not yet been visited. The third set contains the weights of the edges of the currently formed tree structure. The fourth set includes the number of wind turbines carried by all edges of the tree structure. Distance proximity matrix, a two-dimensional matrix containing the mutual distance information between all nodes; S22. Obtain the first variable and the second variable of the current co-channel cable topology tree structure; S23. Generate a tree structure starting from the substation location, delete the current node from the second set and add it to the first set; S24. According to the first variable of the current connection process, the selection node number is obtained, the selection node is deleted from the second set and added to the first set, the weight of the corresponding edge is found from the distance neighbor matrix according to the current node and the selection node number, the weight of the edge is added to the third set, and the number of wind turbines carried by the edge is cumulatively analyzed; S25. According to the second variable of the current connection process, obtain information on whether this connection requires the cable to be laid in the same trench, If it is necessary to lay cables in the same trench, obtain the node numbers at both ends of the edge connected in the same trench according to the edge connected in this section, delete the nodes at both ends from the second set and add them to the first set, and at the same time obtain the weight of the edge connected in the same trench, add the weight of the edge to the third set, and perform cumulative analysis on the number of wind turbines carried by the edge; If it is not necessary to lay cables in the same trench, do not perform the operation; S26. Enter the next connection process, repeat S24-S25 until all connection processes are completed, and obtain the first set, the third set and the fourth set as the current co-channel cable topology model.

5. The cable topology connection optimization method for offshore wind farms considering cable trench sharing according to claim 4 is characterized by: The cumulative analysis of the number of wind turbines carried by the side includes the following process: Determine whether the node number at the end of the current edge is not greater than N, where N is the total number of fans. If so, add 1 to the number of fans carried by the current edge in the fourth set, obtain all the previous edges connected to the trunk of the edge in the fourth set, and add 1 to the number of fans carried by these edges; if not, do not perform any operation.

6. A cable topology connection optimization method for offshore wind farms considering cable trench sharing according to claim 4 or 5, characterized in that: Set multiple levels of cable carrying capacity and the corresponding cross-sectional area of ​​each level of cable; According to the number of wind turbines carried by each edge in the same-channel cable topology model, cables with corresponding levels of cable carrying capacity are configured.

7. The method for optimizing cable topology connection of offshore wind farms considering cable trench sharing according to claim 6 is characterized by: The number of wind turbines carried by each edge in the fourth set is compared with the cable carrying capacity to determine whether the number of wind turbines carried is greater than the cable carrying capacity. If so, a penalty value is added to the corresponding edge.

8. The cable topology connection optimization method for offshore wind farms considering cable trench sharing according to claim 7 is characterized by: The cost of calculating each co-channel cable topology model includes: The cost of the cable topology is calculated based on the sum of the installation costs of all cables between nodes, the cable investment, and the economic efficiency of the overall loss of the cables over their entire life cycle.

9. The method for optimizing cable topology connection in offshore wind farms considering cable trench sharing according to claim 8, characterized in that The particle swarm optimization algorithm is used to optimize each model, including: S41. Initialize the values ​​of the first vector and the second vector as the position of the particle, and randomly generate the velocity of the particle; S42. Calculate the fitness of all particles in the population; Obtain the cost of each co-channel cable topology model, which is the fitness; S43. Compare the fitness values ​​of each particle position, and obtain the global optimal particle and the local optimal particle by comparison; S44. Update the velocity and position of the particle; S45. According to the updated speed and position of the particle, the cost of each co-channel cable topology model is recalculated. If the fitness of the new particle is better than the original one, the particle is updated to be the local optimal particle. If the fitness of the new particle is better than the global optimal particle, the particle is updated to be the global optimal particle. S46. Repeat S42-S45 until the maximum number of iterations is reached, and output the final result, which is the optimal cable topology.

10. A method for optimizing cable topology connection in offshore wind farms considering cable trench sharing according to any one of claims 1 to 5, characterized in that: The topological tree structure of the same-channel cable is randomly generated according to the wind turbine nodes. The generation process includes: S11. Decision making starts from the substation; S12. Randomly select wind turbine nodes for connection; S13. Randomly set the cable connection situation in the same trench; S14. Make a decision based on the current selectable node, and repeat S12-S13 until all hierarchical nodes are connected.

Images