An optimization method for offshore wind power transmission system with multi-layer facility interconnection

Abstract

The present invention discloses an optimization method for an offshore wind power transmission system with multi-layer facility interconnection, including: determining the location and number of offshore booster stations according to a greedy algorithm to update an offshore wind farm set; determining an initial connection relationship variable set according to the offshore wind farm set, and sorting the connection relationship variables in the connection relationship variable set; establishing an objective function and corresponding constraints of the connection relationship variables based on an E-MFLP model, and scaling the objective function by an adaptive kernel search method; iteratively solving the objective function based on mixed integer linear programming according to the scaled model, and obtaining an optimal vector solution formed by the connection relationship variables. The problem model is constructed as a mixed integer linear programming problem, solving the optimization problem involving integer variables and continuous variables, and realizing the site selection of the energy island; realizing the optimal connection between the energy island and the onshore converter station, and ensuring the lowest overall cost.

Description

Technical Field

[0001] The present invention relates to the technical field of offshore wind power transmission system optimization, and in particular to an offshore wind power transmission system optimization method with multi-layer facility interconnection. Background Art

[0002] With the development trend of offshore wind farms in large scale and offshore, building offshore wind turbines and developing "energy islands" will help increase the development of wind energy in deepwater areas. Offshore wind farms are connected to adjacent energy islands. For wind farms far away from the islands, it is necessary to build offshore booster stations to transmit the electricity generated by adjacent wind farms to the energy islands for storage and use. The electricity stored in the energy islands can also be transmitted to onshore converter stations for use. The existing similar problem models include hierarchical facility site selection problems. The multi-layer facility site selection optimization based on this problem model is mainly divided into three categories: 1) heuristic-based algorithms; 2) intelligent algorithm-based solutions. The optimal solution obtained by this algorithm is an approximate optimal solution and is more likely to fall into a local optimal solution. The gap between the obtained approximate optimal solution and the actual optimal solution cannot be measured and estimated; 3) mathematical programming-based methods, branch and bound methods, can guarantee the optimal solution but require a high solver price and a long calculation time for large wind farms. The current problem is that directly using either mathematical programming methods or existing heuristic methods for existing problem models requires high computational complexity and inflexible calculation methods.

[0003] For example, a "method and storage medium for optimizing the structure of an offshore wind power cluster transmission system" disclosed in a Chinese patent document, with publication number: CN107609694B, discloses the following: obtaining an expression model of the structure of an offshore wind power cluster transmission system based on input information of the offshore wind power cluster; constructing an objective function for optimizing the structure of an offshore wind power cluster transmission system based on the expression model; obtaining an optimal solution set for the objective function based on an ant colony algorithm to obtain a wind power cluster transmission cable structure matrix; constructing an offshore wind power cluster transmission system structure based on the wind power cluster transmission cable structure matrix. This solution also has the problem of an inflexible optimization process. Summary of the invention

[0004] In order to solve the problem that the existing technology cannot flexibly optimize the offshore wind power transmission system for energy islands, the present invention provides an optimization method for an offshore wind power transmission system with multi-layer facility interconnection, which constructs the problem model into a mixed integer linear programming problem, and then solves the optimization problem involving integer variables and continuous variables to realize the site selection of the energy island.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] An optimization method for an offshore wind power transmission system with multi-layer facility interconnection includes: determining the location and number of offshore booster stations according to a greedy algorithm to update an offshore wind farm set;

[0007] Determine an initial connection relationship variable set according to the offshore wind farm set, and sort the connection relationship variables in the connection relationship variable set;

[0008] Establishing the objective function and corresponding constraint conditions of the E-MFLP model based on the connection relationship variables, and scaling the objective function based on the characteristics of the adaptive kernel search method;

[0009] According to the objective function after model scaling, the adaptive kernel search method is used to iteratively solve the problem based on mixed integer linear programming to obtain the optimal vector solution formed by the connection relationship variables. The problem model is constructed as a mixed integer linear programming problem to solve the optimization problem involving integer variables and continuous variables to achieve the site selection of the energy island. The optimal connection between the energy island and the onshore converter station is achieved to ensure the lowest overall cost.

[0010] Preferably, the connection relationship variables include variables for connecting the offshore wind farm with the energy island, variables for connecting the energy island with the onshore converter station, and variables for connecting the offshore wind farm with the onshore converter station. The variables in the entire energy transmission system under different settings of the offshore wind farm can be flexibly determined.

[0011] Preferably, the corresponding constraints include the capacity constraints of the energy island and the capacity constraints of the onshore converter station, and also the connection quantity constraints between the offshore wind farm and the energy island. The capacity constraints enable resources to be effectively allocated between the energy island and the onshore converter station to avoid resource waste or overload. The connection quantity constraints promote a more balanced distribution of the output of the offshore wind farm and guide the optimal connection method between the offshore wind farm and the energy island.

[0012] As a preferred method, it includes generating several offshore booster stations through a heuristic algorithm; determining an initial set of offshore wind farms, and updating the set of offshore wind farms in combination with the offshore booster stations. The heuristic algorithm can generate multiple potential locations of offshore booster stations, which helps to find the optimal or near-optimal layout plan in a vast sea area. By updating the set of offshore wind farms, the effective connection between the wind farms and the booster stations can be ensured, thereby optimizing resource allocation and improving energy transmission efficiency.

[0013] As a preferred method, offshore wind farms without energy islands within the range threshold are established as offshore wind farms, and the location and number of offshore booster stations are determined according to the location of the offshore wind farms, so as to facilitate the determination of system layout in special cases.

[0014] Preferably, it includes determining a cable cost objective function, and screening the activated offshore wind farms in the updated offshore wind farm set according to the cable cost objective function, so as to achieve further screening of the offshore wind farms.

[0015] Preferably, an adaptive core search model after determining the model scaling is included, in which the connection relationship variable is used as a binary variable, and the core is updated according to the connection between the offshore booster station and the energy island. The connection status between the offshore wind farm, the energy island and the onshore converter station can be described more accurately. The simplicity of the binary variable helps to reduce the amount of model calculation and improve the search efficiency.

[0016] As a preferred method, the connection relationship between the offshore booster station and the energy island is projected into the updated offshore wind farm set; the connection relationship set between the offshore wind farm and the onshore converter station is determined, and the corresponding constraint conditions are determined. The layout of the entire offshore wind power system is optimized, and flexible optimization is achieved through diversified connection relationships.

[0017] Preferably, in each iteration, a new set of indexes is added to the core of the adaptive core search model, so as to realize dynamic update in the optimization process, gradually expand the search range, and realize flexible optimization for each offshore wind farm.

[0018] As a preferred method, the method includes determining the most potential connection relationship variable among the connection relationship variables, and performing an adaptive core search on the most potential connection relationship variable, thereby reducing useless searches and improving optimization efficiency.

[0019] The present invention has the following advantages:

[0020] (1) The problem model is constructed as a mixed integer linear programming (MILP) problem, which can then solve the optimization problem involving integer variables and continuous variables, that is, the location of the energy island can be realized; (2) The variables in the entire energy transmission system under different settings of the offshore wind farm can be flexibly determined; (3) The connection status between the offshore wind farm, the energy island and the onshore converter station can be described more accurately. The simplicity of binary variables helps to reduce the amount of model calculation and improve search efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described below are merely exemplary. A person skilled in the art can derive other implementation drawings based on the provided drawings without inventive effort.

[0022] Figure 1 Schematic diagram of the method steps in the embodiment.

[0023] Figure 2 Schematic diagram of transmission connection of deep sea wind farm in the embodiment. DETAILED DESCRIPTION

[0024] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0025] like Figure 1-Figure 2 As shown, in a preferred embodiment, the present invention discloses a method for optimizing an offshore wind power transmission system with multi-layer facility interconnection, including: determining the location and number of offshore booster stations according to a greedy algorithm to update the offshore wind farm set; using a greedy algorithm to preliminarily determine the location and number of offshore booster stations. This method can provide a relatively reasonable initial solution in a relatively short time, laying a good foundation for the subsequent optimization process. In practical applications, it can often obtain a solution close to the optimal one with high computational efficiency.

[0026] In this step, a greedy algorithm is used to determine the location and number of offshore booster stations and update the set , determine the initial set of binary x-variables. Note that the variables with the greatest potential to take the value 1 in the optimal solution are considered first. This step is usually initialized according to the specific location of the wind farm in the specific scenario. The order of selection has a huge impact on the solution obtained by the heuristic algorithm. Only the adaptive core search is applied to the x-variables.

[0027] The initial connection relationship variable set is determined according to the offshore wind farm set, and the connection relationship variables in the connection relationship variable set are sorted; sorting the connection relationship variable set helps to prioritize key connections in the subsequent optimization process, reduce unnecessary search space, and improve optimization efficiency. At the same time, sorting may also make the problem solving process more orderly, easier to manage and understand.

[0028] Establishing the objective function and corresponding constraints of the E-MFLP model based on the connection relationship variables can comprehensively consider multiple factors such as the site selection of facilities (such as offshore booster stations), the connection relationship of the transmission network, and the transmission cost, thereby constructing a more comprehensive and reasonable objective function. The objective function is model-scaled through the adaptive kernel search method; the search range and step size can be automatically adjusted to improve the search efficiency and accuracy.

[0029] According to the objective function after model scaling, the optimal vector solution formed by the connection relationship variables is obtained through iterative solution based on mixed integer linear programming. It can ensure that an optimal solution or approximate optimal solution that meets the constraints is obtained in each iteration. At the same time, through iterative solution, the global optimal solution can be gradually approached, improving the accuracy and reliability of the optimization results.

[0030] In other embodiments, the connection relationship variables in this scheme include variables for connecting the offshore wind farm and the energy island. At this time, the cost coefficient of the connection between the offshore wind farm and the energy island is determined, and the connection characteristics when the offshore wind farm and the energy island are directly connected are determined. Including variables for connecting the energy island and the onshore converter station. At this time, the cost coefficient of the connection between the energy island and the onshore converter station is determined, and the state and characteristics when the energy island and the onshore converter station are directly connected are determined. Including variables for connecting the offshore wind farm and the onshore converter station. At this time, the cost coefficient and connection characteristics when the offshore wind farm and the onshore converter station are directly connected are determined. By providing connection variables between the offshore wind farm and the energy island, the energy island and the onshore converter station, and the offshore wind farm directly and the onshore converter station, the overall plan can be determined more flexibly, which helps to achieve diversified transmission and efficient utilization of energy, and facilitates accurate and detailed comprehensive evaluation of the overall plan.

[0031] In other embodiments, the corresponding constraints included in this solution include the capacity constraints of the energy island and the capacity constraints of the onshore converter station. By setting a reasonable upper limit on the energy island capacity, it can be ensured that the energy island will not be overloaded when receiving, storing and distributing electricity, thereby improving the stability and reliability of the entire system. This helps to reduce system failures and downtime caused by overload. It also includes constraints on the number of connections between offshore wind farms and energy islands. As a key node in power transmission, the capacity limit of the converter station helps to optimize the allocation of power resources. By rationally planning the capacity of the converter station, it can be ensured that electricity is efficiently used during the transmission process and energy waste is reduced. On the premise of meeting various constraints, the solution can ensure the efficiency of power transmission and distribution, thereby reducing costs. It is convenient to flexibly optimize resource allocation.

[0032] In other embodiments, several offshore booster stations are generated by a heuristic algorithm; the heuristic algorithm can provide a feasible solution to the problem within a reasonable time. Although it is not necessarily a global optimal solution, its flexibility enables the algorithm to adapt to offshore wind farm planning problems of different scales and complexities. Determine the initial set of offshore wind farms, and update the set of offshore wind farms in combination with the offshore booster stations. By generating multiple offshore booster station schemes, the layout of the offshore wind farm can be flexibly adjusted to adapt to different terrain, wind speed, water depth and other conditions, thereby improving the efficiency and reliability of the overall system.

[0033] Offshore wind farms usually involve a large amount of resources and equipment investment, including wind turbines, submarine cables, and offshore booster stations. By generating multiple offshore booster station solutions through heuristic algorithms, the allocation of these resources can be optimized, unnecessary waste can be reduced, and overall economic efficiency can be improved.

[0034] This includes optimizing according to factors such as wind resource distribution, submarine cable laying costs, and construction difficulty, and selecting the most suitable location for the offshore booster station. As the power collection center of the offshore wind farm, the stability and reliability of the offshore booster station are crucial to the operation of the entire wind farm. By generating multiple offshore booster station solutions and optimizing them, the redundancy and fault tolerance of the system can be ensured, and the reliability of the entire wind farm can be improved. In an offshore wind farm, there will be certain losses in the process of transmitting electricity from wind turbines to the land power grid. By optimizing the location and number of offshore booster stations, the length of the collection line can be shortened, the transmission loss can be reduced, and the efficiency of the entire wind farm can be improved. At the same time, the offshore booster station can convert low-voltage electricity into high voltage before transmission, further reducing losses during transmission.

[0035] The heuristic algorithm is customized according to the needs of the specific project. For example, based on the scale, geographical location, environmental conditions and other factors of the wind farm, an offshore booster station solution that meets the needs is generated.

[0036] In other embodiments, this solution includes establishing offshore wind farms without energy islands within the range threshold as offshore wind farms. Offshore wind farms are usually located in sea areas that are farther offshore, have higher wind speeds and are more stable, and these areas have richer wind energy resources. By establishing such areas as wind farms, the use of wind energy resources can be maximized and the power generation efficiency can be improved. Thereby corresponding optimization is achieved. The location and number of offshore substations are determined according to the location of the offshore wind farm. Since offshore wind farms are far away from the land, there will be certain losses during the power transmission process. By reasonably determining the location and number of offshore substations, the power transmission distance can be shortened, the energy loss during the power transmission process can be reduced, thereby improving the overall power generation efficiency, and achieving optimization of energy loss and overall power generation efficiency from different angles during the optimization process.

[0037] Determining the location and number of offshore booster stations based on the specific location of offshore wind farms can avoid unnecessary waste in infrastructure construction.

[0038] Through scientific planning and reasonable layout, the length of submarine cable laying can be reduced, the construction difficulty and cost can be reduced, and the subsequent operation and maintenance costs can be reduced. At the same time, as a key facility for power collection and transmission, the reasonable layout of offshore substations can also help improve the reliability of the overall system.

[0039] By optimizing the settings of offshore wind farms, the optimal layout is determined in consideration of the situation of large losses.

[0040] In other embodiments, the scheme includes determining a cable cost objective function, determining an offshore wind farm layout optimization objective through the cable cost objective function, and screening the activated offshore wind farms in the updated offshore wind farm set according to the cable cost objective function, thereby achieving further screening of the offshore wind farms.

[0041] In other embodiments, the present solution includes determining an adaptive core search model after the model is scaled, improving the adaptability of the model, and being able to dynamically adjust its internal parameters and structure according to the actual connection between the offshore booster station and the energy island. This dynamic adjustment mechanism enables the model to better adapt to the complex and changing marine environment and improve the accuracy and timeliness of decision-making. During the construction of offshore wind farms and energy islands, the connection relationship may change due to various factors such as wind direction changes, ocean current influences, and technological advances. The adaptive core search model can flexibly respond to these changes and ensure the effectiveness and reliability of the model.

[0042] The adaptive core search model uses the connection relationship variable as a binary variable. By incorporating the connection relationship variable as a binary variable into the model, the power transmission capacity and efficiency between the offshore booster station and the energy island can be predicted more accurately. This helps to optimize resource allocation in the planning stage and avoid unnecessary waste and loss. Based on the accurate prediction results, different connection schemes can be analyzed for cost-effectiveness and the optimal connection scheme can be selected for implementation.

[0043] The core is updated according to the connection status of the offshore booster station to the energy island. The adaptive core search model can reflect the connection status between the offshore booster station and the energy island in real time, discover potential risks and problems in different connection combinations, and achieve continuous optimization and improvement.

[0044] In other embodiments, the connection relationship between the offshore booster station and the energy island is projected into the updated offshore wind farm set; by clarifying the connection relationship between the offshore booster station and the energy island and between the offshore wind farm and the onshore converter station, the layout of the entire offshore wind power system can be optimized, and the optimization with the minimum loss can be achieved. Determine the set of connection relationships between the offshore wind farm and the onshore converter station, and determine the corresponding constraints. Determine the constraints under the synergy of the offshore booster station, the energy island, the offshore wind farm, and the onshore converter station. Achieve diversified connection relationships, disperse the risks of the system, and ensure the reliability of the optimization results. It can facilitate the precise optimization of offshore wind farms.

[0045] In other embodiments, in each iteration, a new set of indexes is added to the core of the adaptive core search model. By continuously iterating and adding new index sets, the search model can gradually adapt to the changing layout of offshore wind farms, and determine the optimization intention of offshore wind farms through dynamic optimization. In addition, the new index set can enrich the knowledge base of the model and improve the accuracy and relevance of search results.

[0046] As the index collection continues to increase, the search range that the search model can cover will gradually expand. This means that the model can handle more diverse query requests and meet the search needs of different users. And through the continuous increase in the index collection, the model in the optimization process can continuously follow the changes in the optimization target and optimize itself.

[0047] In other embodiments, the scheme includes determining the most potential connection relationship variable among the connection relationship variables, determining a most potential variable for any set of offshore booster stations, and assigning its variable value to 1. The variable is used as the core for searching, that is, an adaptive core search is performed on the most potential connection relationship variable. At this time, the remaining variables are disassembled, and the variable factors of the same characteristics are sorted separately to obtain multiple groups of different sorting sequences, and the multiple groups of sorting sequences are subjected to mixed integer linear programming, and the search core is updated according to the parameters of the selected offshore booster station connected to the energy island through the adaptive core search method.

[0048] The connection relationship variable between the offshore booster station and the energy island is projected into the original wind farm set, the wind farms connected to the offshore booster station in the original wind farm set are removed, and the offshore booster station is added to the wind farm set.

[0049] The wind farm set is updated through the offshore booster station to achieve unified optimization of different types of elements; E-MFLP modeling is performed based on the updated wind farm set, energy island and onshore converter station, and the variables are relaxed linearly. Constraints are gradually obtained through the adaptive kernel search method, and the output is the connection relationship between the wind farm, energy island and onshore converter station.

[0050] In another embodiment, the present solution discloses a hierarchical facility site selection model, which includes three levels. The first level includes deep-sea wind farms producing wind energy, the second level includes energy islands, and the third level includes onshore converter stations. For wind farms in different locations, it is determined whether to directly access the energy island, or to select a suitable location to build an offshore booster station to access the energy island, or to directly connect the offshore wind farm to the onshore converter station. The location and number of offshore booster stations in the above process, and the connection methods of offshore wind farms, offshore booster stations, energy islands, and onshore converter stations are optimized.

[0051] The method includes the following steps: first, defining a multi-layer facility site selection problem. Under the E-MFLP framework, assumptions and definitions are made. An initial connection logic is defined, wherein the initial connection logic includes that an offshore wind farm is directly connected to a nearby energy island, an offshore wind farm close to an onshore converter station is directly connected to a nearby onshore converter station, and an offshore wind farm far from an onshore converter station is connected to the energy island through an offshore booster station.

[0052] The method includes treating offshore wind farms without energy islands within the threshold distance n as offshore wind farms, and determining the location and number of offshore booster stations according to the location of offshore wind farms. First, a set of wind farms without energy islands within the coverage radius is determined, and the offshore booster stations are connected to the energy islands. A set of locations of offshore booster stations is determined, and each offshore booster station at each location is connected to at least one energy island.

[0053] Through the heuristic algorithm, a number of offshore booster stations are generated to form a set L. A new set of offshore wind farms is established based on the set of offshore wind farms and offshore booster stations.

[0054] Then the new offshore wind farms will replace the original offshore wind farms, and the connection between the offshore wind farms and the energy islands and onshore converter stations will be planned.

[0055] Determine the first correlation coefficient of the offshore wind farm island energy island, determine the second correlation coefficient of the energy island onshore power station, and determine the third correlation coefficient from the offshore wind farm to the onshore converter station. Based on the location and number of offshore booster stations determined by E-MFLP, select the activated energy islands and optimize the connection between the wind farm and the energy island and onshore converter station.

[0056] The method includes simulating E-MFLP through binary variables, wherein the binary variables include a first variable, a second variable and a third variable. The first variable includes whether the offshore wind farm is connected to an energy island, the second variable includes whether the energy island is connected to an onshore converter station, and the third variable includes whether the offshore wind farm is connected to an onshore converter station.

[0057] A constraint relationship is established, wherein the constraint relationship includes that each offshore wind farm must be connected to at least one energy island or onshore converter station.

[0058] Each energy island must be connected to at least one onshore converter station.

[0059] Set a threshold for the number of energy islands.

[0060] A threshold is set for the number of onshore power plants.

[0061] Including establishing an adaptive core search model. The core search method allows the variables to be screened and adjusted. More refined variable processing is achieved through the adaptive core search model. Among them, it includes scaling the variables and scaling the range of the variables. The core of the E-MFLP is updated according to the result of selecting the energy island connected to the offshore booster station. The relationship of the energy island connected by the offshore booster station is projected into the set of new offshore wind farms. Determine the set of offshore wind farms connected to the onshore converter station.

[0062] Extending constraints based on adaptive kernel search method.

[0063] After the constraints are extended, iterations are performed during which new index sets are added to the core.

[0064] Based on the adaptive kernel search method, constraints are added to the E-MFLP, and the added constraints include that the first variable includes at least one variable in a new index set.

[0065] The output of the E-MFLP includes a vector based on the first variable, the second variable and the third variable.

[0066] The mathematical model of E-MFLP is as follows: ; The objective function represents the optimization goal, including the cable cost from the wind farm to the energy island , the cable cost between the wind farm and the onshore grid connection point , the cost between the energy island and the onshore grid connection point , and the cost of the booster station that needs to be built a is a parameter related to the cost of building a power station. Here, at the energy island and onshore converter station, it means that the construction cost is determined based on the capacity of all connections.

[0067] Among the constraints, It indicates the connection relationship between the wind farm and the energy island. It is a variable that is either 0 or 1. 0 means no connection and 1 means connection. It shows the connection relationship between the energy island and the onshore grid connection point. It shows the connection relationship between the wind farm and the grid connection point on the road. It means that the connection between the wind farm and the energy island is greater than or equal to the connection between the offshore substation and the energy island. This constraint is naturally established according to the designed model. The original problem model E-MFLP is adjusted according to the characteristics of the adaptive core search method. The core search method is a technique widely recognized in the existing literature, which allows variables to be screened and adjusted. Based on this method, the adaptive core search is further introduced to achieve more refined variable processing, which can scale the variables. (14); Then we can use the adaptive kernel search method. That is, the scaling of this step meets the conditions for using the kernel search method. Next, in order to speed up the calculation, we design an adaptive kernel search method.

[0068] definition , , the restricted mixed integer linear program (MILP) derived from the original problem will be called , where if ,variable is treated as a binary variable, otherwise it is fixed to 0. For each j in (15);

[0069] That is, according to the selected offshore booster station connected to energy island j, the core In addition, for The variables are Treated as a binary variable, all other u variables are fixed to zero, representing the energy island j connected by the offshore booster station, and projecting the relationship to the set That is, in the original wind farm set, the wind farm connected to the offshore booster station is removed, and the offshore booster station is added to the wind farm set. In addition, define For offshore wind farms A collection of which is connected to the onshore converter station i and there is satisfy and .

[0070] Based on the adaptive kernel search method, in order to meet the restrictions of the adaptive kernel search method, the constraints are extended:

[0071] (16) (17); This constraint is naturally established because the wind farm connected to energy island j includes the offshore booster station. This constraint can be solved by using an adaptive kernel search method.

[0072] At this stage, an iterative process begins. In each iteration, a new set of indexes is added to the core, named .

[0073] Then based on the adaptive kernel search method, The model adds the following constraints to ensure that the wind farm can be connected to the energy island: (18).

[0074] In addition, in order to enforce Select at least the collection to ensure that different solutions are found. If the previous iteration found a feasible solution but could not prove the optimality of that solution due to time constraints, then constraint (18) is replaced by the following constraint: (19); The purpose of constraints (18) and (19) is to ensure that all wind farms can find energy islands for connection during the optimization process.

[0075] Final question: Output of E-MFLP Is The vectors formed represent the connection between wind farms, energy islands, and on-road grid connection points.

[0076] The problem model is constructed as a mixed integer linear programming (MILP) problem, which can then solve the optimization problem involving integer variables and continuous variables, that is, the site selection of the energy island can be realized and the construction cost can be optimized.

[0077] The first stage of the algorithm: Use the greedy algorithm to determine the location and number of offshore booster stations and update the set , determine the initial set of binary x-variables; that is, the initial core. In this step, the remaining x-variables are sorted. Note that the variables with the greatest potential to take the value 1 in the optimal solution are considered first. This step is usually initialized according to the specific location of the wind farm in the specific scenario, which can be an empirical choice. The order of selection has a huge impact on the solution obtained by the heuristic algorithm. Only the adaptive core search is applied to the x-variables.

[0078] Phase II: Solve a series of constrained mixed integer linear programs (MILPs) derived from the original problem, including considering the location of offshore booster stations, to optimize the grid connection scheme of offshore wind farms that are greater than 30 km from both the energy island and the onshore converter station. The offshore booster station is integrated into the offshore wind farm set to optimize the overall energy island and converter station location and the layout of the entire wind power cluster. During the scheme optimization process, the adaptive kernel search method can be used to accelerate the solution of the E-MFLP problem according to the characteristics of the wind farm. The kernel search method is used for the wind farm set to ensure that better solution bounds are gradually obtained. The solution of the previous constrained MILP and the computational work required to solve it guide the construction of the subsequent core. In each iteration, the core will be updated to test other promising variables and remove useless variables. Finally, the location of the energy island, the onshore converter station, and the connection layout between the wind farm, the energy island, and the onshore converter station, including the offshore boost, can be obtained.

[0079] In another embodiment, the present solution discloses a three-layer facility site selection method for an energy island, comprising the following steps. 1. Input: number of wind farms , number of energy islands , Number of onshore converter stations , Energy Island Capacity Limitation , onshore converter station capacity limitation , connection cost .

[0080] 2. Use heuristic algorithms to determine the construction demand and location of offshore substations and generate new wind farm sets , to achieve the connection of wind farms that are farther away from the energy island and the onshore converter station than the specified distance through the offshore booster station to the energy island or the onshore converter station. The specific access plan optimizes the offshore booster station together with the rest of the wind farms.

[0081] 3. Based on , J, I modeling, and linearizing and relaxing the problem model; by observing the relationship between variables, it is found that the variables Setting Range After that, the conditions of the adaptive kernel search method can be met to optimize and solve the model E-MFLP.

[0082] 4. Sort the x variables using the classic method in the core search method and generate an initial kernel , to select an initial set of wind farms.

[0083] 5. Collect Initialize to , initialized with the highest ranked x variable.

[0084] 6. Solution problem, and let Become the solution to the problem and initialize the iterator t to 0.

[0085] 7. Use sorting criteria to determine , to the collection Add new variables in .

[0086] 8. If t=0 is true, go to step 10; if t=0 is false, determine whether step t=0 is optimal. If so, go to step 9; if false, go to step 10.

[0087] 9. Solution And satisfy (18), go to step 11; if the previous solution is optimal, the feasible solution to this problem is .

[0088] 10. Solution And satisfy (19), go to step 11; the feasible solution of this problem is if the previous solution is feasible, then Can improve the speed of optimization.

[0089] 11. Selected Collection The variables in are added to the core set and are denoted as , and the core set variables that have not been selected in the past p iterations are removed from the core set, denoted as .

[0090] 12. Update , update t=t+1; in each iteration, a new set of wind farms is added to the core.

[0091] 13. Determine whether the stopping criteria are met, when the quality of the solution is no longer significantly improved or when the iteration time or number of iterations is limited; this problem may not be feasible, so a time limit is imposed. In addition, a time limit is added to find a feasible solution, that is, if a feasible solution is not found within the time limit, the subproblem will be stopped. In addition, if the current solution does not improve after a fixed time limit, the subproblem will also be stopped.

[0092] 14. Output , that is, the offshore wind farm includes the connection structure of the offshore booster station, energy island and onshore converter station.

[0093] These modifications or improvements made on the basis of the present invention without departing from the spirit of the present invention all fall within the scope of protection required by the present invention.

Claims

1. A method for optimizing an offshore wind power transmission system with multi-layer facility interconnection, characterized in that: include: Update the offshore wind farm set by determining the location and number of offshore booster stations based on a greedy algorithm; Determine an initial connection relationship variable set according to the offshore wind farm set, and sort the connection relationship variables in the connection relationship variable set; Establishing the connection relationship variables based on the objective function and corresponding constraints of the E-MFLP model, and scaling the objective function by an adaptive core search method; adding constraints to the E-MFLP model based on the adaptive core search method, wherein the added constraints include at least one variable in a new index set in the first variable; The corresponding constraints include the capacity constraints of the energy island and the capacity constraints of the onshore converter station and the connection quantity constraints between the offshore wind farm and the energy island; in the adaptive core search model, the connection relationship variable is used as a binary variable, and the core is updated according to the connection between the offshore booster station and the energy island; the most potential connection relationship variable among the connection relationship variables is determined, and the most potential connection relationship variable is adaptively searched; According to the objective function after model scaling, the optimal vector solution formed by the connection relationship variables is obtained by iterative solution based on mixed integer linear programming.

2. The method for optimizing an offshore wind power transmission system with multi-layer facility interconnection according to claim 1, characterized in that: The connection relationship variables include variables for connecting the offshore wind farm and the energy island, variables for connecting the energy island and the onshore converter station, and variables for connecting the offshore wind farm and the onshore converter station.

3. The method for optimizing an offshore wind power transmission system with multi-layer facility interconnection according to claim 1, characterized in that: The capacity constraint enables resources to be effectively allocated between the energy island and the onshore converter station; the connection quantity constraint evenly distributes the output of the offshore wind farm.

4. A method for optimizing an offshore wind power transmission system with multi-layer facility interconnection according to claim 1 or 2, characterized in that: The method includes generating a number of offshore booster stations through a heuristic algorithm; determining an initial set of offshore wind farms, and updating the set of offshore wind farms in combination with the offshore booster stations.

5. The method for optimizing an offshore wind power transmission system with multi-layer facility interconnection according to claim 4, characterized in that: This includes establishing offshore wind farms without energy islands within the range threshold as offshore wind farms, and determining the location and number of offshore substations based on the location of the offshore wind farms.

6. A method for optimizing an offshore wind power transmission system with multi-layer facility interconnection according to claim 1 or 2, characterized in that: The method comprises determining a cable cost objective function, and screening activated offshore wind farms in an updated offshore wind farm set according to the cable cost objective function.

7. The method for optimizing an offshore wind power transmission system with multi-layer facility interconnection according to claim 1, characterized in that: This includes projecting the connection relationship between the offshore booster station and the energy island into the updated offshore wind farm set; determining the connection relationship set between the offshore wind farm and the onshore converter station, and determining the corresponding constraints.

8. The method for optimizing an offshore wind power transmission system with multi-layer facility interconnection according to claim 1, characterized in that: This involves adding a new set of indexes to the core of the adaptive core search model in each iteration.

Images