Optimization method for power topological structure of offshore wind plant

Abstract

The invention discloses a method for optimizing an electric power topological structure of an offshore wind power plant, which overcomes the problems that the optimization of the electric power topological structure of the offshore wind power plant in the prior art cannot ensure optimality and the result is unstable, and comprises the following steps of: acquiring wind resource data of the wind power plant, and acquiring fan parameters and cable parameters of the wind power plant; setting a cable topology optimization constraint by taking the lowest cable connection cost as an objective function, and establishing an optimization model; second-order cone constraints are set, so that the optimization model meets the solving form of a mixed integer second-order cone model, and the mixed integer second-order cone model is obtained; and solving the mixed integer second-order cone model to obtain an optimal cable topology connection form and position coordinates of the offshore substation. A detailed cable load flow calculation method is comprehensively considered, integrated modeling of topological connection of a transformer substation and a cable in a continuous domain is achieved, a mixed integer second-order cone model is formed, and the optimality of the obtained final solution can be guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of offshore wind farm design, and in particular to a method for optimizing the power topology structure of an offshore wind farm. Background Art

[0002] The two most important tasks in optimizing offshore wind farm electrical systems are substation site selection and cable topology generation. These considerations include cable costs, the economic impact of cable losses over the wind farm's lifecycle, and the cable footprint and installation costs.

[0003] Existing technologies for calculating cable connection topology and substation locations primarily rely on intelligent algorithms. These algorithms can consider substation locations within a continuous domain and simultaneously optimize cable paths and cross-sectional areas. However, because they are based on non-gradient algorithms, they cannot guarantee theoretical optimality. Furthermore, intelligent algorithms are more suitable for solving continuous function optimization problems and are prone to local solutions for mixed-integer optimization problems. This requires continuous manual adjustment of optimization parameters for different wind farms, and multiple simulations to select the best solution as the final solution.

[0004] For example, the Chinese Patent Office published a patent on December 20, 2024: CN119167554A, a method for optimizing cable topology connections in offshore wind farms that considers cable trench sharing. The method includes calculating the cost of each cable topology model in the same trench; optimizing each model with the goal of minimizing the cost, and obtaining the cable topology corresponding to the optimal solution as the optimal cable topology. Although both consider the cost issue and need to optimize the cable connection topology and the location of the substation, the particle swarm algorithm used is also an intelligent algorithm that can only solve physical models. In some wind farms with special layouts, the final result may be inferior to the manually designed solution, and the optimality of the solution cannot be guaranteed. Moreover, the results are unstable, and it is necessary to run the program multiple times and continuously adjust the parameters based on experienced engineers to obtain a better solution. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem in the prior art that the optimization of the power topology structure of offshore wind farms cannot guarantee optimality and the results are unstable. A method for optimizing the power topology structure of offshore wind farms is provided, which comprehensively considers the detailed cable flow calculation method and the integrated modeling of substation and cable topology connections in the continuous domain to form a mixed integer second-order cone model. The final solution obtained can guarantee optimality.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A method for optimizing an offshore wind farm power topology structure comprises the following steps: Collect wind resource data of wind farms, wind turbine parameters and cable parameters of wind farms; Taking the minimum cable connection cost as the objective function, the cable topology optimization constraints are set and the optimization model is established; Setting a second-order cone constraint so that the optimization model satisfies the solution form of the mixed-integer second-order cone model, thereby obtaining the mixed-integer second-order cone model; The mixed integer second-order cone model is solved to obtain the optimal cable topology connection form and the location coordinates of the offshore substation.

[0007] The present invention comprehensively considers the detailed cable flow calculation method and the integrated modeling of substation and cable topology connections in a continuous domain, and proposes a mixed integer second-order cone programming model for solving the cable connection topology and substation site selection of offshore wind farms. It can ensure that the location of the offshore substation and the cable connection topology are solved simultaneously in the continuous domain.

[0008] Preferably, the second-order cone constraint includes: the distance from the j-th wind turbine to the offshore substation is greater than or equal to || (X s , Y s )-(Xt j , Yt j )||2-(1-B 1，j，k )*M and the distance from the j-th wind turbine to the offshore substation is greater than or equal to 0.

[0009] As a preferred method, the branch-and-bound method is used to solve the mixed integer second-order cone model: if there is no connection between the offshore substation and wind turbine j, then the distance from the jth wind turbine to the offshore substation is greater than or equal to 0; if the offshore substation and wind turbine j are connected by the kth type of cable, then the distance from the jth wind turbine to the offshore substation is greater than or equal to || (X s , Y s )-(Xt j , Yt j )||2.

[0010] Preferably, the cable topology optimization constraints include: the number of cables flowing into each wind turbine is less than or equal to a first threshold, and the number of cables flowing out of each wind turbine is equal to 1; and the voltage drop along the cables is less than or equal to a second threshold.

[0011] Preferably, the cable topology optimization constraints include: wind turbine energy transmission constraints, on-road substation energy balance constraints and cable carrying capacity constraints.

[0012] Preferably, the second threshold is one-mth of the maximum voltage difference, and m is preferably square root of three.

[0013] Preferably, in the cable topology connection, if there is a connection between wind turbine i and wind turbine j, then one and only one cable cross-sectional area is selected on the path between wind turbine i and wind turbine j.

[0014] Preferably, the collected wind farm turbine parameters include location information and capacity information of all wind turbines in the current wind farm; the collected cable parameters include cable prices and electrical parameters in the current candidate library, and the electrical parameters include cable current carrying capacity.

[0015] Preferably, the objective function of minimizing the cable connection cost includes: taking the sum of the cable laying cost and the sea occupation cost, the cable body cost and the cable loss cost as the cable connection cost, and establishing the cable optimal selection matrix.

[0016] Preferably, the mixed-integer second-order cone model is solved using a solver, which includes a CPLEX solver, a GUROBI solver and a COPT solver.

[0017] Therefore, the present invention has the following beneficial effects: it comprehensively considers the detailed cable flow calculation method, the integrated modeling of the substation and cable topology connection in the continuous domain, establishes an optimization model, sets the second-order cone constraint, transforms the established optimization model so that it can meet the solution form of the second-order cone optimization model, and constructs a mixed integer second-order cone programming model for solving the cable connection topology and substation site selection of offshore wind farms, which can ensure that the location of the offshore substation and the cable connection topology are solved simultaneously in the continuous domain, and the location of the substation at sea can be defined more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flowchart of the overall steps of the method for optimizing the power topology structure of an offshore wind farm in Example 1.

[0019] Figure 2 This is the existing generated cable connection topology diagram for offshore wind farms.

[0020] Figure 3 This is the cable connection topology diagram of the offshore wind farm generated in Example 2. DETAILED DESCRIPTION

[0021] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1: This embodiment provides a method for optimizing the power topology structure of an offshore wind farm. Figure 1 As shown in the figure, the operation process is as follows: Step 1, collect wind resource data of the wind farm, collect wind farm wind turbine parameters and cable parameters; Step 2, take the lowest cable connection cost as the objective function, set the cable topology optimization constraints, and establish an optimization model; Step 3, according to the recognition results, based on the preset source code template, generate a calibration data source code file; Step 4, integrate the obtained calibration data source code file in the controller to obtain the calibration quantity data value change point.

[0022] The offshore wind farm power topology optimization method provided in this embodiment is primarily focused on the joint optimization of the offshore wind farm transmission and collection system topology. This method comprehensively considers detailed cable power flow calculation methods and the integrated modeling of substation and cable topology connections within a continuous domain. By establishing an optimization model, setting second-order cone constraints, and transforming the established optimization model to meet the solution requirements of the second-order cone optimization model, a mixed-integer second-order cone programming model is constructed to solve the offshore wind farm cable connection topology and substation site selection. This ensures that the location of the offshore substation and the cable connection topology are simultaneously solved within the continuous domain, allowing for a more precise definition of the substation's offshore location.

[0023] The following further illustrates the technical solutions and technical effects of the present invention through specific examples and specific application scenarios. The following examples are intended to explain the present invention, but the present invention is not limited to the following examples.

[0024] Specifically, such as Figure 1 As shown, a method for optimizing the power topology structure of an offshore wind farm specifically includes the following steps: Step 1: Collect wind resource data of the wind farm, and collect wind farm wind turbine parameters and cable parameters.

[0025] The collected wind farm wind turbine parameters include parameters such as the location information and capacity information of all wind turbines in the current wind farm, where the location information includes the coordinate position of the wind turbine and the height of the wind turbine; the collected cable parameters include the cable prices and electrical parameters in the current alternative library, and the electrical parameters include information such as the cable current carrying capacity and the cable unit distance impedance.

[0026] Step 2: Taking the lowest cable connection cost as the objective function, set the cable topology optimization constraints and establish the optimization model.

[0027] Specifically include: Step (1): Define the optimization variable X i,j , optimize the intermediate variable As i,j 、B i,j,k 、Dn i,j wait.

[0028] Among them, X i,j Is a binary variable, indicating whether the path between wind turbine i and wind turbine j is selected. i,j represents the cable length from fan i to fan j, B i,j,k represents a binary optimization variable, indicating whether wind turbine i and wind turbine j are connected by the kth cable, Dn i,j Represents the distance matrix, that is, the geometric distance between wind turbine i and wind turbine j.

[0029] Step (2): Establish an optimization model.

[0030] Among them, establishing the optimization model specifically includes the following steps: Step (2-1): Calculate the cable connection cost.

[0031] In this embodiment, the cable connection cost includes the cable laying cost and sea occupation cost, the investment cost of the cable body, and the loss cost along the cable line.

[0032] Specifically, the cable laying cost and sea occupation cost for: In the above formula, represents the sea area occupied and installation cost of the cable between wind turbines i and j, which is linearly related to the cable length. i and i represent the wind turbine numbers, and k represents the cross-sectional area of the selected cable, representing a cable type. Even with the same connection path, cable costs and losses will vary when different cross-sectional areas are selected. It should be noted that the optimization model of this embodiment uses a matrix to represent the cable connection method. Therefore, when there is no cable connection between wind turbines i and j, the value of this location is assigned to 0.

[0033] Among them, the cable laying cost and sea occupation cost are linearly related to the length of the cable, so c unit，TC It refers to the cost per unit length. The latter part is to calculate the length of the entire cable, and this length is related to the location of the offshore substation and the topology of the cable based on the location of this substation.

[0034] As i，j，k is an intermediate variable. If there is a cable connection between wind turbines (WT) i and WTj, and they bear the power generation of k wind turbines, then B i，j，k =1 and As i，j，k Equal to the cable length (distance) between them, otherwise As i，j，k If i is 1, the electricity generated by wind turbine j will be directly transmitted to the offshore substation (OS), that is, i=1 represents the location number of the OS, and other numbers represent wind turbines.

[0035] L j，k represents the distance from the j-th wind turbine to the offshore substation, k represents the cable type, and it is recorded that the cable from the j-th wind turbine to the offshore substation is of the k-th type. The optimization variable is defined to facilitate the subsequent loss calculation.

[0036] B i，j，k represents a binary optimization variable. When it is equal to 1, it means that wind turbine i and wind turbine j are connected by the kth cable. When it is equal to 0, it means that there is no cable connection between wind turbine i and wind turbine j.

[0037] N n Indicates the number of fans, N t Indicates the maximum number of wind turbines that the existing cable can support.

[0038] The investment cost of the cable itself is: In the above formula, s* represents the optimal cable type when a cable exists between wind turbines i and j, and the cable carries the power of k wind turbines. The cost of the cable is related to both its length and type (i.e., its cross-sectional area). Therefore, the cable type selection for each segment can be determined in advance by proof, eliminating the need for optimization. This significantly reduces the complexity of the problem. In other words, once the cable connection layout is determined, the optimal cable type can be selected accordingly.

[0039] It represents the cost of the cable per unit length when the cable type is s*, and Type_best(k) represents the optimal cable type k.

[0040] The loss cost along the cable is: Where, It represents the economic efficiency of the overall cable loss over the entire life cycle (assuming the wind farm has a 20-year operating cycle), and represents the net present value of the power loss cost over the life of the wind farm. In this embodiment, it is calculated based on the current price of 0.405 yuan / kWh.

[0041] R s* Represents the resistance per unit length of the cable when the cross section k is selected, Irate d Represents the rated current of the wind turbine, NY represents the wind farm operation period, which is generally 20 years, and C e is the on-grid electricity price, which is 0.405 yuan / kWh in this embodiment. f represents the capital discount rate, τ ΔPmax It stands for the equivalent full power hours, which is the number of hours of equivalent rated power operation of the wind farm throughout the year calculated based on statistical data. is the power factor, Y K Indicates the energy loss flowing through the cable, P rated Indicates the rated power of the fan, U rated Indicates the rated operating voltage of the fan.

[0042] Since the rated power of each unit is the same, the energy loss flowing through the cable can be expressed by the number of wind turbines currently carried by the cable, which is represented by Yk. That is:

[0043] Step (2-2): Set the objective function and constraints and establish the optimization model.

[0044] In this embodiment, the objective function is set as follows: the sum of the cable connection cost including the cable laying cost and sea occupation cost, the investment cost of the cable body and the loss cost along the cable is minimized.

[0045] The constraints set include: (a) Wind turbine energy delivery constraints.

[0046] Among them, P i，j represents the energy transmitted between fans i and j, X i，j Indicates whether there is a cable connection between wind turbines i and j. 0 represents no connection, and 1 represents yes connection. The maximum value cannot exceed the maximum current carrying capacity of the cable. Nt represents the maximum number of wind turbines that the current cable can carry.

[0047] (b) Cable routing constraints.

[0048] P i，j represents the energy transmitted between fan i and fan j, P j，i = represents the energy delivered from wind turbine j to wind turbine i. Cable route constraints are designed to prevent duplicate routes. For example, if a path P12 is already formed, P12 and P21 are considered equivalent. Cable route constraints ensure that the same route is not added twice.

[0049] (c) Energy balance constraints of the on-road substation.

[0050] The energy balance constraint of the substation on the road indicates the energy flowing into the substation on the road. Equal to the energy of the booster station flowing out

[0051] (d) Cable cross-sectional area selection constraints.

[0052] Indicates that when X i，j When this path is assigned 1, only one cable cross-section is selected on this path.

[0053] (e) Cable carrying capacity constraints.

[0054] Where Cy k It is a discrete integer representing the capacity that different cable cross-sectional areas can carry, calculated as an integer multiple of the wind turbine capacity.

[0055] (f) Energy transfer constraints.

[0056] Where, X 1，j N represents the path from the substation to wind turbine j, f_max The energy transmission constraint is a constraint based on some engineering constraints, which means that there must be at least one cable to transmit the energy generated by the wind turbine (WT) to the offshore substation (OS). The number of cables that can be installed on the offshore substation platform is limited to N. cable .

[0057] (g) Cable quantity constraints.

[0058] The cable quantity constraint indicates the number of cables that can flow into each wind turbine. is finite, with a maximum value of N in_max , There can be only one cable outflow from each fan, i.e. is equal to 1, and each wind turbine and substation cannot be connected to itself, that is, X i，i Equal to 0, i∈[1, N n ].

[0059] (h) Cable voltage drop constraint.

[0060] The cable voltage drop constraint means that in order to ensure the normal operation of the wind farm, the voltage drop along the cable needs to be specified within a certain range. That is: Where, L i，j，k Indicates that fan i to fan j are connected by the kth cable, Rc i，j，k The unit resistance of the cable connected from fan i to fan j by the kth cable, XL i，j，k Indicates the reactance value of the unit cable, ΔU max Indicates the maximum voltage fluctuation allowed for cable operation.

[0061] According to the above objective function and constraints, an optimization model is established.

[0062] Step 3: Set the second-order cone constraint so that the optimization model satisfies the solution form of the mixed-integer second-order cone model, and obtain the mixed-integer second-order cone model.

[0063] In this embodiment, the second-order cone constraints set include: Where, X s Indicates the x-coordinate of the offshore transformer station, Y s Indicates the y coordinate of the offshore transformer station, Xt j Indicates the x-coordinate of the j-th wind turbine, Yt jrepresents the y coordinate of the j-th fan, M represents a relatively large value, and is a parameter used in the linearization process using the large M method, ||(X s , Y s )-(Xt j , Yt j )||2 represents the geometric distance between the offshore substation and wind turbine j.

[0064] The core idea of the Big M method is to introduce artificial variables into the constraints and add corresponding M or -M terms to the objective function. 3 M is an arbitrarily large (but not infinite) positive number, and its selection requires care to ensure it does not cause numerical problems. This method can transform originally nonlinear constraints into linear constraints, making them easier to solve using a linear programming solver.

[0065] The above second-order cone constraints are to form a transformation of the existing optimization model by the mixed integer second-order cone model. By adding the second-order cone constraints, the existing optimization model can meet the solution form of the mixed integer second-order cone model for easy solution.

[0066] Step 4: Solve the mixed integer second-order cone model to obtain the optimal cable topology connection form and the location coordinates of the offshore substation.

[0067] By establishing an equivalent second-order cone model, the original non-convex nonlinear optimization problem is rewritten into its convex form. The second-order cone model constructed in this form can be solved directly using the branch-and-bound method using a commercial solver, that is, when B 1，j，k =0 when 0≤L j，k , whereas B 1，j，k =1 then ||(X s , Y s )-(Xt j , Yt j )||2≤L j，k .

[0068] According to the solution results, the optimal cable topology connection form and the location coordinates of the offshore substation are obtained.

[0069] The present embodiment provides a method for optimizing the power topology structure of an offshore wind farm, which has the following beneficial effects: 1. A mixed integer second-order cone programming model is proposed for solving the cable connection topology and substation site selection of an offshore wind farm, which can ensure that the location of the offshore substation and the cable connection topology are solved simultaneously in a continuous domain.

[0070] 2. The substation's offshore location can be more precisely defined. This can further reduce investment costs compared to the current approach of selecting an optimal substation location from a pool of multiple alternatives (typically three or four in current literature). This is because substation location has a greater impact on the overall optimization result than optimizing the cables alone. The model proposed in this example ensures that the theoretically optimal substation location is found within a continuous domain.

[0071] Example 2: This embodiment provides a method for optimizing the power topology of an offshore wind farm, which optimizes the power topology of an offshore wind farm by taking into account specific application scenarios.

[0072] like Figure 2 As shown, Figure 2 It shows an example of an optimized cable topology and substation site selection for a designed offshore wind farm, where s1 represents the location of the offshore substation, s2-s36 represent the location of the wind turbines, and the dotted line segments represent different types of cables (generally speaking, different types of cables or cable selection actually refers to the selection of cables with different cable cross-sectional areas). Cables with different cross-sectional areas have different rated flow rates, which means that thicker cables can carry more energy transmitted by wind turbines. For example, cables s2-s9 only carry the electrical energy generated by wind turbine s2, and cables s18-s17 carry the energy generated by wind turbines s2, s9, s16, and s17. The key task of optimizing the overall offshore electrical system is to determine the location of the substation, that is, Figure 2 The example shown shows the location of s1 and the cable topology connecting all wind turbines. Figure 2 In the middle dotted line segment, it is also necessary to ensure that the cross-sectional area of each cable can bear the energy generated when the wind turbine is at rated power.

[0073] The heuristic algorithm in the existing technical solution is based on the minimum spanning tree problem or the traveler problem in graph theory, and uses a deterministic algorithm to solve it. This solution method is fast and the results are stable. However, it can only decouple the selection of cable cross-sectional area, cable connection topology and substation site optimization with the shortest cable length as the goal. The overall investment cost of offshore wind farm cables is closely related to these three aspects. The generation of the optimal solution cannot be guaranteed, and sometimes it may produce a solution that is more expensive than manual design. The current mathematical programming-based method can only consider several alternative offshore substation locations for the optimization of offshore substation locations, and the final solution is an integer programming model. It is impossible to solve the location of the substation in a continuous domain. In addition, the work based on this part generally uses the most simplified calculation when considering the current of the cable throughout its life cycle, and cannot solve in detail the impact of cable loss on the final electrical optimization design solution.

[0074] Therefore, the present embodiment provides a method for optimizing the power topology structure of an offshore wind farm, which adopts the concept of mathematical programming, comprehensively considers the detailed cable power flow calculation method, and the integrated modeling of substation and cable topology connections in the continuous domain, forming a mixed integer second-order cone programming model. The final solution obtained can ensure optimality.

[0075] Specifically, the embodiment provides a method for optimizing the power topology structure of an offshore wind farm. The overall algorithm flow of the model is as follows: Step 1: Collect one year's wind resource data for the wind farm, the coordinates of all wind turbines in the current wind farm, the capacity, height, and other parameters of the wind turbines, and obtain the cable prices and electrical parameters (such as unit distance impedance) in the current candidate library; Step 2: Build an optimization model.

[0076] Step 2-1: Define the optimization variable X i,j , optimize the intermediate variable As i,j 、B i,j,k 、Dn i,j wait; Step 2-2: Calculate the cable laying cost and sea occupation cost, the investment cost of the cable body and the loss cost along the cable line, and construct the cable optimal model matrix (that is, the optimization variables in the optimization model are all expressed in matrix form) to simplify the optimization modeling.

[0077] Step 2-3: Establish an optimization objective function. In this embodiment, the objective function is: Where, represents the sea area occupied and installation cost of the cable between wind turbine i and wind turbine j. This cost is linearly related to the length of the cable. Represents an investment in cables, This represents the economic efficiency of overall cable losses over the entire lifecycle (assuming a 20-year wind farm operation cycle), calculated at the current price of 0.405 yuan / kWh. I, j represents the wind turbine number, and k represents the selected cable cross-sectional area. Even with the same connection path, cable costs and losses vary when different cross-sectional areas are selected. Note that this model uses a matrix to represent cable connections, so when there is no cable connection between turbines i and j, the value in this location is assigned to 0.

[0078] Step 2-5: Establish cable topology optimization constraints. The cable topology optimization constraints in this embodiment are consistent with those in the first embodiment.

[0079] Step 2-6: Build a mixed-integer second-order cone model and select a solver for solving it. You can choose the commercial CPLEX, GUROBI, or COPT solver.

[0080] Step 3: Select the optimized GAP value, the program stops, and the final result is output.

[0081] In a solver, GAP is a key parameter used to assess the gap between the current solution state and the target optimal solution. GAP is generally expressed as a relative value, measuring the gap between the solution found by the current solver and the optimal solution (or the best known solution). When solving large-scale mixed integer programming problems, GAP is an important metric for measuring solver performance. The solver continuously attempts to narrow the GAP to find a solution closer to the optimal solution. Depending on the specific problem, solver parameters such as the number of iterations and time limit can be adjusted to optimize GAP.

[0082] In mixed integer programming, GAP is usually defined as: GAP=|BP-BF| / |BP| Among them, BP represents the optimal solution obtained after integer continuation (that is, the optimal solution to the relaxed problem), and BF represents the integer solution obtained by the current solver. Most solvers (such as Gurobi, CPLEX, etc.) provide a MIPGap parameter to control when the solution process stops. When GAP is less than MIPGap, the solver will assume that a good enough solution has been found and stop solving. The value of MIPGap can be adjusted according to actual needs. If you want the solver to find a better solution faster, you can increase MIPGap appropriately; if you want the solver to find the optimal solution more accurately, you can reduce MIPGap.

[0083] Theoretically, when GAP is 0, it is the theoretical optimum. However, considering the time required to solve complex problems in practical engineering, the GAP can be set to an acceptable range for engineering to stop the program early. This GAP represents the percentage of the distance from the optimal solution. Generally speaking, the time required from 1% to 0% is very long. For large optimization problems, it may take dozens of hours. However, in engineering, the production structure may still need to be fine-tuned, so the optimization GAP can be defined according to the required accuracy.

[0084] Using the method provided in this embodiment, the cable topology connection form and the location of the offshore substation are finally obtained as follows: Figure 3 As shown, Figure 3 middle, 1 represents the location of the offshore substation, 2-37 represent the location of the wind turbine, and the solid line segments represent different types of cables (generally speaking, different types of cables or cable selection actually refers to selecting cables with different cable cross-sectional areas). Cables with different cross-sectional areas have different rated flow rates, which means that thicker cables can carry more energy transmitted by the wind turbine. Figure 3 There are four types of cables, such as 1-37 is the fourth type of cable, 1-12, 1-13, 1-27 and 1-26 are the third type of cables, 1-18 is the second type of cable, and the remaining solid line segments represent the first type of cables.

[0085] The key task of optimizing the overall offshore electrical topology is to determine the location of the substation, i.e. Figure 3 The example shown shows the location of 1 and the cable topology connecting all wind turbines. Figure 3 In the solid line segment, make sure that the cross-sectional area of each cable can carry the energy generated by the wind turbine at rated power.

[0086] This embodiment also provides an offshore wind farm power topology structure optimization system, including: a data acquisition module, a cost calculation module, an optimization model establishment module, a mixed integer second-order cone model establishment module, and a model solving module. The data acquisition module is connected to the cost calculation module, the cost calculation module is connected to the optimization model establishment module, the mixed integer second-order cone model establishment module is connected to the optimization model establishment module, and the model solving module is connected to the mixed integer second-order cone model establishment module.

[0087] Specifically: The data acquisition module is used to collect the wind resource data of the current wind farm for one year, the coordinates of all wind turbines in the current wind farm, the capacity, height, and other parameters of the wind turbines, and obtain the cable prices and electrical parameters (unit distance impedance and other information) in the current candidate library.

[0088] The cost calculation module is used to calculate the cable laying cost and sea occupation fee, and calculate the cable body cost and cable loss cost.

[0089] The optimization model establishment module uses the total cost calculated by the cost calculation module as the cable connection cost, takes the lowest cable connection cost as the objective function, and sets the cable topology optimization constraints to establish the optimization model.

[0090] The mixed integer second-order cone model establishment module sets the second-order cone constraint and transforms the established optimization model. By adding this constraint, the existing optimization model can meet the solution form of the second-order cone optimization model for easy solution.

[0091] The model solving module solves the established mixed integer second-order cone model to obtain the optimal cable topology connection form and the location coordinates of the offshore substation, and outputs the cable topology map.

[0092] The embodiment described above is only a preferred solution of the present invention and does not limit the present invention in any form. Other variations and modifications are possible without exceeding the technical solution described in the claims.

Claims

1. A method for optimizing the power topology of an offshore wind farm, characterized in that: include: Collect wind resource data of wind farms, wind turbine parameters and cable parameters of wind farms; Taking the minimum cable connection cost as the objective function, the cable topology optimization constraints are set and the optimization model is established; Setting a second-order cone constraint, converting the optimization model into a solution form of a mixed-integer second-order cone model, and obtaining a mixed-integer second-order cone model; The mixed integer second-order cone model is solved to obtain the optimal cable topology connection form and the location coordinates of the offshore substation.

2. The method for optimizing the power topology of an offshore wind farm according to claim 1, characterized in that: The second-order cone constraint includes: the distance from the jth wind turbine to the offshore substation is greater than or equal to the difference between the geometric distance between the offshore substation and wind turbine j and the linear constraint amount, and the distance from the jth wind turbine to the offshore substation is greater than or equal to 0.

3. The method for optimizing the power topology of an offshore wind farm according to claim 2, characterized in that: The branch-and-bound method is used to solve the mixed-integer second-order cone model: if there is no connection between the offshore substation and wind turbine j, then the distance from the jth wind turbine to the offshore substation is greater than or equal to 0; if the offshore substation and wind turbine j are connected by the kth type of cable, then the distance from the jth wind turbine to the offshore substation is greater than or equal to the geometric distance between the offshore substation and wind turbine j.

4. The method for optimizing the power topology structure of an offshore wind farm according to claim 1, characterized in that: The cable topology optimization constraints include: the number of cables flowing into each wind turbine is less than or equal to a first threshold, the number of cables flowing out of each wind turbine is equal to 1; and the voltage drop along the cables is less than or equal to a second threshold.

5. The method for optimizing the power topology structure of an offshore wind farm according to claim 1, 2, 3 or 4, characterized in that: The cable topology optimization constraints include: wind turbine energy transmission constraints, substation energy balance constraints and cable carrying capacity constraints.

6. The method for optimizing the power topology structure of an offshore wind farm according to claim 4, characterized in that: The second threshold is one mth of the maximum voltage difference.

7. A method for optimizing the power topology of an offshore wind farm according to claim 4 or 6, characterized in that: In the cable topology connection, if there is a connection between wind turbine i and wind turbine j, there is only one cable cross-sectional area selected on the path between wind turbine i and wind turbine j.

8. The method for optimizing the power topology structure of an offshore wind farm according to claim 1, 2, 3, 4 or 6, characterized in that: The collected wind farm turbine parameters include the location information and capacity information of all wind turbines in the current wind farm; the collected cable parameters include the price and electrical parameters of cables in the current candidate library, and the electrical parameters include the cable current carrying capacity.

9. The method for optimizing the power topology structure of an offshore wind farm according to claim 1, 2, 3, 4 or 6, characterized in that: The objective function of minimizing the cable connection cost includes: taking the sum of the cable laying cost and the sea occupation cost, the cable body cost and the cable loss cost as the cable connection cost, and establishing the cable optimal selection matrix.

10. The method for optimizing the power topology structure of an offshore wind farm according to claim 1, 2, 3, 4 or 6, characterized in that: The mixed-integer second-order cone model is solved by using a solver, wherein the solver includes a CPLEX solver, a GUROBI solver and a COPT solver.

Images