Wind power plant parallelogram regular distribution-cable joint optimization method and system

Abstract

The invention discloses a parallelogram regular machine arrangement-cable joint optimization method and system for a wind power plant. The method comprises the steps of obtaining an area range of the wind power plant and the number of wind turbine generators, randomly generating shape variables of a parallelogram rule, establishing a shape variable initial population of a genetic algorithm, and generating a wind power plant generator arrangement scheme; based on a wind power plant unit arrangement scheme, establishing a booster station initial population of a genetic algorithm; the method comprises the following steps: by taking maximized output power and minimized cable cost as optimization objectives, generating a new shape variable of a parallelogram rule and a booster station coordinate, forming an optimization population, performing multi-objective optimization on a wind power plant unit arrangement scheme and a booster station position scheme based on a genetic algorithm to obtain a multi-objective optimized Pareto leading edge, and performing multi-objective optimization on the wind power plant unit arrangement scheme and the booster station position scheme. And selecting an optimal arrangement scheme of the wind power plant according to requirements. While the total output power of the wind power plant is ensured, the cable cost of the wind power plant is reduced, and the comprehensive benefits of the whole life cycle of the wind power plant are improved.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of wind power, and in particular to a wind farm parallelogram regular layout machine-cable joint optimization method and system. Background Art

[0002] In the current design and operation of wind farms, the layout of wind turbines has an important impact on power generation efficiency. Traditional wind farm designs mostly use regular geometric arrangements, such as grid or linear layouts, and manually arrange wind turbines. Although this method is easy to plan, it has many shortcomings in actual applications. First, the directionality and distribution characteristics of wind resources are not fully considered, resulting in low operating efficiency of some wind turbines; second, the reduction effect of the wake effect between wind turbines on power output is not fully considered, especially under specific wind direction conditions.

[0003] Existing related technologies are based on simple regular geometric shapes and use traversal methods to find the best wind turbine layout. Although this can improve the layout effect, it is inefficient and cannot fully optimize, making it difficult to balance dynamic adjustment and cost control under complex wind farm conditions; or try to improve the layout effect and efficiency through optimization methods, but the units generally use random and irregular points, which lack aesthetics and are not convenient for construction and operation and maintenance planning. These limitations provide clear directions and technical requirements for further optimization of wind farm layout methods.

[0004] The micro-site selection of wind turbine locations and cable layout in wind farm planning are usually considered as independent, sequential planning steps, first planning the location of wind turbines and the substation according to the parallelogram layout rule, and then determining the design of the cable layout. However, due to the competition between the power generation of wind turbines and the cost of cables within the wind farm, the traditional step-by-step sequential design may lead to suboptimal solutions in the optimization results. In this case, it is necessary to consider a comprehensive joint planning method that aims to find a balance between wind turbine location and cable layout, which is of great significance to the research and practical application of wind farm planning.

[0005] In summary, in the existing wind farm unit location design process, it is difficult to achieve the optimal wind farm power generation by relying on experience, and the completely random layout lacks aesthetics and is not conducive to the design of operation and maintenance paths. The single pursuit of higher power generation without fully considering the impact of wind farm cable costs cannot achieve the optimal rate of return over the entire life cycle of offshore wind farms. Summary of the invention

[0006] In view of the technical problems existing in the prior art, the present invention provides a wind farm parallelogram regular layout-cable joint optimization method and system, which can ensure the total output power of the wind farm while reducing the cable cost of the wind farm and improving the comprehensive benefits of the wind farm over the entire life cycle.

[0007] In order to solve the above technical problems, the technical solution proposed by the present invention is: A wind farm parallelogram regular layout machine-cable joint optimization method, comprising the steps of: Obtain the regional scope of the wind farm to be optimized, the model and quantity of the units, the wind resource conditions, and the values ​​of various parameters required for the cable layout design; According to the regional scope of the wind farm and the number of wind turbines, shape variables of the parallelogram rule are randomly generated, and an initial population of shape variables of the genetic algorithm is established; wherein the coordinate initial population contains multiple individuals, each of which is a set of shape variables of the parallelogram rule, and a corresponding uniquely determined arrangement scheme of wind farm units of the parallelogram rule can be generated based on the point selection program; Based on the generated wind farm unit arrangement plan, the booster stations in the field are randomly generated within the range that meets the optimization constraints, and the initial population of booster stations of the genetic algorithm is established; wherein the initial population of booster stations includes multiple individuals, each of which is a wind farm booster station location plan; Taking maximizing the target value of the first optimization target and minimizing the target value of the second optimization target as the optimization target, starting from the initial population of shape variables and the initial population of booster stations, new parallelogram-ruled shape variables and booster station coordinates are generated to form an optimization population. Based on the genetic algorithm, multi-objective optimization is performed on the wind farm unit arrangement scheme and the booster station location scheme; the first optimization target is the total output power of the wind farm under the arrangement scheme corresponding to each individual; the second optimization target is the cable cost of the wind farm under the arrangement scheme corresponding to each individual; When both the first optimization objective and the second optimization objective have reached convergence, the optimization process ends and the Pareto frontier of multi-objective optimization is obtained; each individual on the Pareto frontier is a definite wind farm unit arrangement and booster station location scheme, including the coordinates of each unit and booster station; On the Pareto frontier, the optimal layout of wind farms is selected according to demand.

[0008] Preferably, according to the regional scope of the wind farm and the number of wind turbines, the specific steps of randomly generating shape variables of the parallelogram rule are: First, set the constraints of wind turbine coordinates; the constraints include the preset wind farm area range and the distance between any two wind turbines in the wind farm is greater than , that is, the wind turbine rotor diameter of times, of which is a positive integer greater than 1; Secondly, within the closed polygonal area of ​​the wind farm, a point coordinate is randomly selected as the global center point coordinate of the parallelogram regular layout. ; Again, according to the uniform probability distribution, Randomly select the angle between two sides of a parallelogram within the range and the overall rotation angle of the parallelogram , based on unit spacing restrictions A certain multiple range of Randomly pick the length of the parallelogram side and ,exist Randomly extract the gradient coefficient between rows within the range and Two parameters, in Randomly extract the gradient coefficient within the line within the range and Two parameters; Finally, according to the determined number of wind turbines, the shape variables of the generated parallelogram rule are used; according to the point-taking procedure, the corresponding unique wind farm unit layout plan is generated within the wind farm area, and the process is repeated until the total number of units in the unit layout plan meets the number of units.

[0009] Preferably, according to the point selection procedure, the specific steps of generating a corresponding unique wind farm unit arrangement scheme within the wind farm area are: Shape variables based on the parallelogram rule , and the closed polygonal range of the wind farm area, extract the global center point coordinates ;by is a vertex, the horizontal line is an edge, and rotates counterclockwise , draw a straight line that intersects the entire polygon, recorded as the straight line , and then As the vertex, For one edge, rotate counterclockwise Angle, and then draw a straight line to intersect the entire polygon, record it as the straight line ; Taking the straight line l as the reference, from the global center point coordinates Starting from the point, according to the gradient point method, points are taken on both sides, and then several parallel lines of line l are drawn through the points to intersect with the closed area; then, with line m as the reference, according to the gradient point method, points are taken on both sides, and several parallel lines of line m are drawn through the points to intersect with the closed area; the intersection point of these two sets of parallel lines is recorded as , forming a series of parallelograms, let the length of the side of the parallelogram on the straight line l and its parallel lines be a, let the length of the side of the parallelogram on the straight line m and its parallel lines be b, and the distance between the two parallel lines of the parallelogram =a , =b ; The union of all intersection points As the shape variable of the current parallelogram rule The corresponding unique wind farm unit arrangement plan.

[0010] Preferably, take the straight line l as the reference and start from the global center point coordinates Starting from, according to the gradient point method, take points on both sides, and then draw several parallel lines through the points. The specific steps are: The longest hypotenuse distance of the wind farm boundary is , then you need to pass through the global center point , along the perpendicular direction of the straight line l, the number of intervals of gradient points, that is, the number of straight lines, is at most ,in For not more than The largest integer of ; Consider a Gaussian distribution function with a mean of 0 ,exist Uniform extraction within the range Number - , then we can get No less than A collection of gradient distances: ; in and is the parallelogram line spacing gradient coefficient, all gradient points form a set ; Draw a line parallel to line l through the selected gradient point.

[0011] Preferably, taking the straight line m as a reference, and also using the gradient point selection method, points are selected on both sides, and the specific process of drawing several parallel lines of the straight line m through the points is as follows: The longest hypotenuse distance of the wind farm boundary is , then you need to go through the global center point , along the perpendicular direction of the line m, the number of intervals of gradient points, that is, the number of straight lines, is at most ,in For not more than The largest integer of ; Consider a Gaussian distribution function with a mean of 0 ,exist Uniform extraction within the range Number - , then according to the Gaussian distribution function we get No less than A collection of gradient distances: ; in and is the parallelogram line spacing gradient coefficient, all gradient points form a set ; Draw a line parallel to line m through the chosen gradient point.

[0012] Preferably, based on the generated wind farm unit arrangement scheme, the step of randomly generating on-site booster stations within the range that meets the optimization constraints and establishing the initial population of booster stations of the genetic algorithm specifically includes: First, set the selection constraints of the booster station; the constraints include the preset wind farm area range and the setting of the booster station to be greater than the distance of any wind turbine in the wind farm. , which is n times the diameter of the wind turbine rotor, where n is a positive integer greater than 1; Randomly select points within the wind farm area to generate the coordinate position vector of the booster station , repeat the process to meet the point constraints of the booster station.

[0013] Preferably, when both the first optimization objective and the second optimization objective have reached convergence, the optimization process is terminated, and the steps of obtaining the Pareto frontier of the multi-objective optimization specifically include: According to the genetic algorithm, the coordinates of the wind turbine and the booster station are encoded with real numbers, and the coordinate variation processes and probabilities of the wind turbines with large and small scales are set respectively, that is, the large-scale variation process randomly selects new coordinates within the range of 2 / 3 times the diameter of the wind rotor of the original coordinates, and the small-scale variation process randomly selects new coordinates within the range of 1 / 6 times the diameter of the wind rotor of the original coordinates. At the beginning of the optimization, both large-scale and small-scale unit position coordinate variation processes are used. When the first optimization objective and the second optimization objective converge for the first time, the large-scale variation probability is set to 0, and only the small-scale unit position coordinate variation process is used until the first optimization objective and the second optimization objective converge for the second time. The optimization process is terminated to obtain the Pareto frontier of multi-objective optimization.

[0014] Preferably, based on the average wind speed and turbulence intensity of the incoming flow of the wind farm, the wake effect is taken into account, and the wake model is used to calculate the output power and flow turbulence intensity of each wind turbine in the wind farm to obtain the total output power of the wind farm.

[0015] Preferably, the specific steps of obtaining the total output power of the wind farm are: For each wind turbine in the wind farm, the velocity loss of the upstream wind turbine at multiple points on the rotor of the current wind turbine is calculated according to formula (1): (1) in, is the upper point of the wind rotor of the current i-th unit The inflow velocity loss at is the point on the wind rotor of the i-th unit of the j-th unit Average speed loss at is a binary variable; N is the number of wind turbines in the wind farm; For each wind turbine in the wind farm, the additional flow turbulence intensity of the upstream wind turbine at multiple points on the rotor of the current wind turbine is obtained according to formula (2): (2) in, is the additional flow turbulence intensity at the point (x, y, z) on the wind rotor of the current i-th unit; is the additional flow turbulence intensity of the jth unit at the point (x, y, z) on the rotor of the i-th unit; is a binary variable; is the number of wind turbines in the wind farm; According to the inflow velocity loss of multiple points on the wind turbine rotor, the average value is taken, and the wind speed in front of the hub of the current wind turbine under the preset wind condition is obtained according to formula (3): (3) in, is the wind speed in front of the hub of the current i-th wind turbine; is the ambient wind speed; At wind speed , the wind direction angle is Under the wind condition of the wind turbine, the wind speed in front of the hub of the current wind turbine and the power curve of the current wind turbine are obtained, and the output power of the current wind turbine under the wind condition is obtained; wherein the power curve of the wind turbine includes the corresponding relationship between the wind speed in front of the hub and the output power of the wind turbine; According to the corresponding output power under this wind condition and formula (4), the total output power of the wind farm under the current wind farm arrangement scheme is obtained: (4) in, is the total output power of the wind farm under the current wind farm arrangement scheme; The wind speed is , wind direction is wind conditions; For unit i in wind condition Output power under For wind conditions Frequency of occurrence; is the number of wind directions; is the number of wind speed segments taken in a single wind direction; N is the number of units.

[0016] Preferably, the specific steps of calculating the cable cost of the wind farm under each individual arrangement scheme include: Based on the location of the wind farm booster station and the coastline, the coast point closest to the booster station is obtained as the feeder cable node; Based on the unit location and booster station location under the wind farm layout plan, as well as the cable information of various types of farms, and coupling the feeder cable node and feeder cable model information, the total cable cost model of the wind farm is established, and the total cable cost of the wind farm is solved: (5) in, is the cable unit price, Representative Points and The Euclidean distance between is a given safety factor.

[0017] The present invention also discloses a computer program product, comprising a computer program, wherein the computer program executes the steps of the method described above when executed by a processor.

[0018] The present invention further discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are executed.

[0019] The present invention also discloses a wind farm parallelogram regular layout machine-cable joint optimization system, comprising an interconnected memory and a processor, wherein a computer program is stored in the memory, and when the computer program is run by the processor, the steps of the method described above are executed.

[0020] Compared with the prior art, the advantages of the present invention are: The wind farm parallelogram regular layout-cable joint optimization method of the present invention aims to solve the problem that in the process of designing the unit points of the existing wind farm, it is difficult to achieve the mathematical optimality of the wind turbine unit points by relying on experience, and the completely random layout lacks aesthetics and is not conducive to the design of the operation and maintenance path; at the same time, the single pursuit of higher power generation fails to fully consider the impact of the cable cost of the wind farm. To this end, the present invention applies a genetic algorithm to maximize the target value of the first optimization target and minimize the target value of the second optimization target as the optimization target, and performs multi-objective optimization on the unit arrangement scheme of the wind farm. And the unit coordinate variation process and probability of two scales of large and small are used, and the "two-stage" variation probability setting scheme is used to ensure that the optimization target value can converge quickly and fully. Finally, the arrangement optimization scheme of the wind farm can be determined on the multi-objective optimization Pareto frontier solution obtained according to actual needs. Through the above method, the present invention comprehensively considers the total output power and cable cost of the wind farm, while ensuring the total output power of the wind farm, reducing the cable cost of the wind farm and improving the comprehensive benefits of the wind farm throughout its life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The present invention is a flowchart of an optimization method in an embodiment.

[0022] Figure 2 It is a schematic diagram of wind farm feed-out point calculation in the present invention.

[0023] Figure 3 Schematic diagram of the rose diagram of the wind farm wind tower and the power thrust system curve of the unit in the present invention; (a) is the rose diagram of the wind farm wind tower; (b) is the power thrust system curve of the unit.

[0024] Figure 4 This is a connection diagram of the original unit locations, original booster stations and collector lines of the wind farm in the present invention.

[0025] Figure 5 This is a graph showing the evolution of the wind farm output power during the optimization process of the present invention.

[0026] Figure 6 This is a structural diagram of the optimized unit point location, booster station location and cable path layout of the present invention. DETAILED DESCRIPTION

[0027] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0028] like Figure 1 As shown, the wind farm parallelogram regular layout-cable joint optimization method provided by the embodiment of the present invention includes the following steps: Obtain the regional scope of the wind farm to be optimized, the model and quantity of the units, the wind resource conditions, and the values ​​of various parameters required for the cable layout design; According to the preset wind farm area and the determined number of wind turbines, shape variables of the parallelogram rule are randomly generated, and the initial population of shape variables of the genetic algorithm is established; wherein the coordinate initial population contains multiple individuals, each of which is a set of shape variables of the parallelogram rule, and the corresponding uniquely determined parallelogram rule wind farm unit arrangement scheme can be generated based on a specific point selection program; the machine site coordinates generated in each generation during the iteration process are random and can also meet the requirements of regularized machine layout; Based on the above generated wind farm unit arrangement scheme, the booster stations in the field are randomly generated within the range that meets the optimization constraints, and the initial population of booster stations of the genetic algorithm is established; wherein the initial population of booster stations includes multiple individuals, each of which is a wind farm booster station location scheme; Taking maximizing the target value of the first optimization target and minimizing the target value of the second optimization target as the optimization target, starting from the initial population of shape variables and the initial population of booster stations, new parallelogram-ruled shape variables and booster station coordinates are generated to form an optimization population. Based on the genetic algorithm, multi-objective optimization is performed on the wind farm unit arrangement scheme and the booster station location scheme. While ensuring that the power generation remains unchanged, the cable cost can be greatly reduced; the first optimization target is the total output power of the wind farm under the arrangement scheme corresponding to each individual; the second optimization target is the cable cost of the wind farm under the arrangement scheme corresponding to each individual; When both the first optimization objective and the second optimization objective have reached convergence, the optimization process ends and the Pareto frontier of multi-objective optimization is obtained; each individual on the Pareto frontier is a definite wind farm unit arrangement and booster station location scheme, including the coordinates of each unit and booster station; On the Pareto frontier, the optimal layout of wind farms is selected according to demand.

[0029] The wind farm parallelogram regular layout-cable joint optimization method of the present invention reduces the cable investment cost of the wind farm, fully approaches the theoretically optimal completely random unit arrangement, and takes into account the influence of the on-site booster station, collection line and the main cable feeding out of the wind farm, thereby reducing the cable investment cost of the wind farm while increasing the power generation of the wind farm, enhancing the economy of the wind farm throughout the life cycle, and enabling the wind farm to perform better.

[0030] In a specific embodiment, according to a preset wind farm area and a determined number of wind turbines, the step of randomly generating shape variables of a parallelogram rule specifically includes: The shape variables based on the parallelogram rule specifically include , where the parameters are: global center point coordinates , parallelogram angle , overall rotation angle , parallelogram row side length a, parallelogram column side length b, inter-row gradient coefficient And the gradient coefficient within the line ; First, set the constraints of wind turbine coordinates; the constraints include the preset wind farm area range and the distance between any two wind turbines in the wind farm is greater than , that is, the diameter of the wind turbine rotor times, of which is a positive integer greater than 1; Secondly, within the closed polygonal area of ​​the wind farm, a point coordinate is randomly selected as the global center point coordinate of the parallelogram regular layout. ; Again, according to the uniform probability distribution, Randomly select the angle between two sides of a parallelogram within the range and the overall rotation angle of the parallelogram , based on unit spacing restrictions A certain multiple range of Randomly pick the length of the parallelogram side and ,exist Randomly extract the gradient coefficient between rows within the range and Two parameters, in Randomly extract the gradient coefficient within the line within the range and Two parameters; Finally, according to the determined number of wind turbines, the shape variables of the generated parallelogram rule are used , according to the point selection procedure, a corresponding unique wind farm unit arrangement plan is generated within the wind farm area, and the process is repeated until the total number of units in the unit arrangement plan meets the number of units.

[0031] Specifically, according to the point selection procedure, the corresponding unique wind farm unit arrangement scheme is generated based on the shape variables of the parallelogram rule, which specifically includes: Shape variables based on the parallelogram rule , and the closed polygonal range of the wind farm area, extract the global center point coordinates .by is a vertex, the horizontal line is an edge, and rotates counterclockwise , draw a straight line that intersects the entire polygon, recorded as the straight line , and then As the vertex, For one edge, rotate counterclockwise Angle, and then draw a straight line to intersect the entire polygon, record it as the straight line .

[0032] Taking the straight line l as the reference, from the global center point coordinates Starting from line m, according to the gradient point selection method, select points on both sides, and then draw several parallel lines of line l through the points, intersecting with the closed area. Then, taking line m as the reference, according to the gradient point selection method, select points on both sides, and draw several parallel lines of line m through the points, intersecting with the closed area. The intersection of these two sets of parallel lines is recorded as , forming a series of parallelograms, let the length of the side of the parallelogram on the straight line l and its parallel lines be a, let the length of the side of the parallelogram on the straight line m and its parallel lines be b, and the distance between the two parallel lines of the parallelogram =a , =b The collection of all intersection points As the shape variable of the current parallelogram rule The corresponding unique wind farm unit arrangement plan.

[0033] Specifically, taking the straight line l as the reference, from the global center point coordinates Starting from, according to the gradient point selection method, points are selected on both sides, and then several parallel lines of line l are drawn through the points. The specific steps include: The longest hypotenuse distance of the wind farm boundary is , then you need to go through the global center point , along the perpendicular direction of the straight line l, the number of intervals of gradient points, that is, the number of straight lines, is at most ,in For not more than The largest integer of ; Consider a Gaussian distribution function with a mean of 0 ,exist Uniform extraction within the range Number - , then we can get No less than Gradient distance collection ,in and is the parallelogram line spacing gradient coefficient, all gradient points form a set ; Draw a line parallel to line l through the chosen gradient point.

[0034] Specifically, taking the straight line m as a reference, also according to the gradient point selection method, points are selected on both sides, and the steps of drawing several parallel lines through the points m specifically include: The longest hypotenuse distance of the wind farm boundary is , then you need to go through the global center point , along the perpendicular direction of the line m, the number of intervals of gradient points, that is, the number of straight lines, is at most ,in For not more than The largest integer of ; Consider a Gaussian distribution function with a mean of 0 ,exist Uniform extraction within the range Number - , then we can get No less than Gradient distance collection ,in and is the parallelogram line spacing gradient coefficient, all gradient points form a set ; Draw a line parallel to line m through the chosen gradient point.

[0035] In a specific embodiment, based on the above generated wind farm unit arrangement scheme, the steps of randomly generating booster stations within the field within the range that meets the optimization constraints and establishing the initial population of booster stations of the genetic algorithm specifically include: First, set the selection constraints of the booster station; the constraints include the preset wind farm area range and the setting of the booster station to be greater than the distance of any wind turbine in the wind farm. , which is n times the diameter of the wind turbine rotor, where n is a positive integer greater than 1; Randomly select points within the wind farm area to generate the coordinate position vector of the booster station , repeat the process to meet the point constraints of the booster station.

[0036] In a specific embodiment, when the first optimization objective and the second optimization objective have both reached convergence, the optimization process is terminated, and the steps of obtaining the Pareto frontier of the multi-objective optimization specifically include: According to the genetic algorithm, the coordinates of the wind turbine and the booster station are encoded with real numbers, and the coordinate variation processes and probabilities of the wind turbines with large and small scales are set respectively, that is, the large-scale variation process randomly selects new coordinates within the range of 2 / 3 times the diameter of the wind rotor of the original coordinates, and the small-scale variation process randomly selects new coordinates within the range of 1 / 6 times the diameter of the wind rotor of the original coordinates. During the optimization process, a "two-stage" mutation probability setting scheme is adopted, so that the population has both global search and local search capabilities in the early stage of optimization, and only retains a certain local search capability in the later stage of optimization, that is: at the beginning of optimization, both large-scale and small-scale unit position coordinate mutation processes are adopted. When the first optimization objective and the second optimization objective converge for the first time, the large-scale mutation probability is set to 0, and only the small-scale unit position coordinate mutation process is adopted until the first optimization objective and the second optimization objective converge for the second time. The optimization process is terminated to obtain the Pareto frontier of multi-objective optimization.

[0037] In a specific embodiment, the total output power of the wind farm under each individual arrangement scheme is calculated by taking into account the wake effect, based on the average wind speed and turbulence intensity of the incoming flow of the wind farm, using a wake model to calculate the output power and flow turbulence intensity of each wind turbine in the wind farm, and obtaining the total output power of the wind farm. The specific steps are: For each wind turbine in the wind farm, the inflow velocity loss at multiple points on the wind rotor of the current wind turbine is obtained according to the velocity loss of the upstream wind turbine at multiple points on the wind rotor of the current wind turbine, and according to the following formula: (1) in, is the upper point of the wind rotor of the current i-th unit The inflow velocity loss at is the point on the wind rotor of the i-th unit of the j-th unit Average speed loss at is a binary variable, if and only if the current i-th unit is downstream of the j-th unit =1, otherwise =0; N is the number of wind turbines in the wind farm; For each wind turbine in the wind farm, the additional flow turbulence intensity of the inflow at multiple points on the wind rotor of the current wind turbine is obtained according to the additional flow turbulence intensity of the upstream wind turbine at multiple points on the wind rotor of the current wind turbine, and according to the method described in the following formula: (2) in, is the additional flow turbulence intensity at the point (x, y, z) on the wind rotor of the current i-th unit; is the additional flow turbulence intensity of the jth unit at the point (x, y, z) on the rotor of the i-th unit; is a binary variable, if and only if the current i-th unit is downstream of the j-th unit =1, otherwise ; is the number of wind turbines in the wind farm; According to the inflow velocity loss of multiple points on the wind turbine rotor of the current wind turbine, the average value is taken, and the wind speed in front of the hub of the current wind turbine under the preset wind condition is obtained according to the method described in the following formula: (3) in, is the wind speed in front of the hub of the current i-th wind turbine; is the ambient wind speed; At wind speed , the wind direction angle is Under the wind condition of the wind turbine, the wind speed in front of the hub of the current wind turbine and the power curve of the current wind turbine are obtained, and the output power of the current wind turbine under the wind condition is obtained; wherein the power curve of the wind turbine includes the corresponding relationship between the wind speed in front of the hub and the output power of the wind turbine; According to the output power corresponding to this wind condition and the method described in the following formula, the total output power of the wind farm under the current wind farm arrangement scheme is obtained: (4) in, is the total output power of the wind farm under the current wind farm arrangement scheme; The wind speed is , wind direction is wind conditions; For unit i in wind condition Output power under For wind conditions Frequency of occurrence; is the number of wind directions; is the number of wind speed segments taken in a single wind direction; N is the number of units.

[0038] In a specific embodiment, the total cost of the wind farm collection line considering the feeder cable under the arrangement scheme corresponding to each individual is calculated, specifically including: Based on the location of the wind farm booster station and the coastline, the coast point closest to the booster station is obtained as the feeder cable node; Based on the unit location and booster station location under the wind farm layout plan, as well as the cable information of various types of farms, and coupling the feeder cable node and feeder cable model information, the total cable cost model of the wind farm is established, and the total cable cost of the wind farm is solved: (5) in, is the cable unit price, Representative Points and The Euclidean distance between is a given safety factor.

[0039] Specifically, based on the unit location and booster station location under the wind farm layout plan, as well as the cable information of various types of farms, and coupling the feeder cable node and the feeder cable model information, the total cable cost model of the wind farm is established, and the steps of solving the total cable cost of the wind farm specifically include: Binary variables The definition is as follows, where Indicates the cable type: (6) Despite There are 2 types of cables to choose from, but each type of cable is mutually exclusive. Only one type of cable can be selected for laying on the same path. The constraints are as follows: (7) The formula for calculating cable cost is: (8) in, is the cable unit price, Representative Points and The Euclidean distance between is a given safety factor.

[0040] In order to meet the design requirements of the radial wind farm collection system, the cable path layout should meet the following constraints: (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) Formula (9) ensures that each node has only one cable for energy output. Formula (10) ensures that at least one cable is connected to the booster station. Formula (11) prevents the same section of cable from being connected to the same node. Formula (12) ensures that there is no energy output after the feed point. Formula (13) ensures that there is a route from the booster station to the feed point. Formula (14) ensures that there is no route from the unit to the feed point. Formula (15) ensures that cable type 4 is used from the booster station to the feed point. Formula (16) prevents energy loops from occurring between any two nodes. Formula (17) avoids any two cable sections from crossing, and stores crossed cable pairs in a collection. Formula (18) represents the power balance maintained at each node, where is a continuous non-negative variable. Formula (19) ensures that the power flow on each cable is lower than the cable capacity.

[0041] The above cable mathematical model can be solved using the mixed integer linear programming method (MILP) to obtain the total cable cost and cable layout under the wind farm unit arrangement and booster station location scheme.

[0042] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0043] like Figure 1 As shown, a wind farm parallelogram regular layout machine-cable joint optimization method in an embodiment of the present invention mainly includes the following steps (1)-(6): (1) Obtain the regional scope of the wind farm to be optimized, the model and quantity of units, wind resource conditions, and the values ​​of various parameters required for cable layout design.

[0044] In one implementation, a suitable wind turbine model may be selected according to the scope of the wind farm to be constructed and the installed capacity requirements.

[0045] As an example, the region vertices of a wind farm can be , , as well as , the coordinate dimension is meter (m), the unit model is Vestas-V80 unit, and the number of units is , wind rose diagram and unit power and thrust coefficient curves are as follows Figure 3 As shown in Figure 2, the original unit locations of the wind farm are as follows: Figure 4 In the wind farm, the unit prices of various types of cables are , , , , where t1-t3 are on-site cables, t4 is the main feeder cable, and the current carrying capacity (maximum number of connectable units) of each type of cable are , , , .

[0046] (2) Based on the preset wind farm area and the determined number of wind turbines, shape variables of the parallelogram rule are randomly generated, and an initial population of shape variables of the genetic algorithm is established; wherein the coordinate initial population includes multiple individuals, each of which is a set of shape variables of the parallelogram rule, and a corresponding uniquely determined wind farm unit arrangement scheme of the parallelogram rule can be generated based on a specific point selection program.

[0047] In one implementation, real number coding is used to randomly generate shape variables of a parallelogram rule to establish an initialization population. The initialization population includes multiple individuals, each of which is a row vector containing 10 variables (i.e. ), respectively corresponding to a wind farm unit arrangement scheme. Among them, real number encoding means that each gene value of an individual is represented by a floating point number within a certain range.

[0048] The shape variables of this parallelogram rule specifically include The parameters are: The parameters are: The global center point coordinates , parallelogram angle , overall rotation angle , the length of the side of the parallelogram , the length of the side of the parallelogram , inter-row gradient coefficient And the gradient coefficient within the line .

[0049] The initial generation method is as follows: First, the unit coordinates are constrained to not exceed the closed polygon enclosed by the vertices of the wind farm area, and the distance between any two wind turbines is greater than , which is 4 times the diameter of the wind turbine rotor; secondly, within the closed polygon of the wind farm area, a point coordinate is randomly selected as the global center point coordinate of the global center point rule layout. ; Finally, according to the uniform probability distribution, Randomly select the angle between two sides of a parallelogram within the range and the overall rotation angle of the parallelogram , based on unit spacing restrictions A certain multiple range of Randomly pick the length of the parallelogram side and ,exist Randomly extract the gradient coefficient between rows within the range and Two parameters, in Randomly extract the gradient coefficient within the line within the range and Two parameters.

[0050] (3) Based on the wind farm unit arrangement plan generated above, randomly generate booster stations within the field within the range that meets the optimization constraints, and establish an initial population of booster stations for the genetic algorithm; wherein the initial population of booster stations includes multiple individuals, each of which is a wind farm booster station location plan.

[0051] In one implementation, real number coding is used, and the constraints of the wind farm range and the spacing between wind turbines are also considered to generate completely random coordinates of the booster stations within the farm.

[0052] (4) Taking maximizing the target value of the first optimization target and minimizing the target value of the second optimization target as the optimization target, starting from the initial population of shape variables and the initial population of booster stations, new parallelogram-ruled shape variables and booster station coordinates are generated to form an optimization population. Based on the genetic algorithm, multi-objective optimization is performed on the wind farm unit arrangement and booster station location plan.

[0053] In this embodiment, the optimization target may be the total output power of the wind farm under each individual arrangement scheme and the total cost of the collector line considering the feeder cable.

[0054] In one implementation, the genetic algorithm may be a NSGA-II algorithm.

[0055] In one embodiment, the shape variables of the parallelogram rule, the bottom layer random coordinates and the coordinates of the booster station within the field can be optimized based on the constraints in step (2) to obtain new shape variables through crossover mutation.

[0056] (5) When both the first optimization objective and the second optimization objective have reached convergence, the optimization process ends and the Pareto frontier of multi-objective optimization is obtained. Each individual on the Pareto frontier is a definite wind farm unit arrangement and booster station location scheme, including the coordinates of each unit and booster station.

[0057] (6) On the Pareto frontier, the optimal layout of wind farms can be selected according to demand.

[0058] In this embodiment, the specific arrangement of the wind farm can be determined through a specific point selection program according to the shape variables of the parallelogram rule with the best optimization target value. Furthermore, according to the locations of the booster stations in the field corresponding to the shape variables of the optimal parallelogram rule (they have the same sequence numbers in the population) and the information of various types of cables, a total cable cost model of the wind farm is established to calculate the optimal cable arrangement of the wind farm.

[0059] Based on the above steps (1)-(6), the present invention can apply a genetic algorithm according to the initialization population established based on the optimization constraint, with the target value of the first optimization target maximized and the target value of the second optimization target minimized as the optimization target, and at the same time, multi-objective optimization is performed on the shape variable population of the parallelogram rule and the substation population within the field, and the unit coordinate variation process and probability of large and small scales are used. Through the "two-stage" variation probability setting scheme, it is ensured that the optimization target value can converge quickly and fully; finally, the layout of the wind farm and the cable path optimization scheme can be determined on the obtained multi-objective optimization Pareto frontier solution according to actual needs. Through the above method, the present invention comprehensively considers the total output power of the wind farm and the total cable cost including the total feed cable, and meets the regular characteristics of the unit point while ensuring the randomness of the wind farm unit point, thereby reducing the total cost of the wind farm cable while ensuring the total output power of the wind farm, and improving the comprehensive benefits of the wind farm throughout its life cycle.

[0060] In a specific embodiment, in step (4), based on the average wind speed and turbulence intensity of the incoming flow of the wind farm, the wake effect is taken into account, and the output power and flow turbulence intensity of each wind turbine in the wind farm are calculated using a wake model to obtain the total output power of the wind farm. The steps specifically include: (4.1) For each wind turbine in the wind farm, the inflow velocity loss at multiple points on the wind rotor of the current wind turbine is obtained according to the velocity loss of the upstream wind turbine at multiple points on the wind rotor of the current wind turbine, and according to the method described in the following formula: (20) in, is the inflow velocity loss at the point (x, y, z) on the wind rotor of the current i-th unit; is the average speed loss of the jth unit at the point (x, y, z) on the rotor of the ith unit; is a binary variable, if and only if the current i-th unit is downstream of the j-th unit , in other cases =0; N is the number of wind turbines in the wind farm.

[0061] In this embodiment, the average speed loss of the wind turbine can be obtained according to the two-dimensional analytical model of the wind turbine wake in the following formula (21): (twenty one) in, is the thrust coefficient; is the actual expansion rate of the wake boundary; R is the radius of the wind rotor; r is the radial distance between the point (x, y, z) on the wind rotor surface and the center of the wind rotor.

[0062] (4.2) For each wind turbine in the wind farm, the additional flow turbulence intensity of the inflow at multiple points on the wind rotor of the current wind turbine is obtained according to the additional flow turbulence intensity of the upstream wind turbine at multiple points on the wind rotor of the current wind turbine, and according to the method described in the following formula: (twenty two) in, is the additional flow turbulence intensity at the point (x, y, z) on the wind rotor of the current i-th unit; is the additional flow turbulence intensity of the jth unit at the point (x, y, z) on the rotor of the i-th unit; is a binary variable, if and only if the current i-th unit is downstream of the j-th unit =1, otherwise ; is the number of wind turbines in the wind farm.

[0063] In this embodiment, the turbulence intensity of the wind turbine wake can be obtained according to the wind turbine wake additional turbulence intensity analytical model in the following formulas (23)-(27): (twenty three) in, is the standard deviation of the Gaussian curve, which is the same as the velocity loss model; z is the vertical height.

[0064] Flow Direction Function The maximum additional turbulence intensity of the wake section at each flow direction is: (twenty four) in, is the correction value considering the near wake area.

[0065] Formula (25) Spanwise function for: (25) in, and The values ​​are: (26) (27) is the vertical correction function: (28) (4.3) According to the inflow velocity loss of multiple points on the wind turbine rotor, the average value is taken, and the wind speed in front of the hub of the current wind turbine under the preset wind condition is obtained according to the method described in the following formula: (29) in, is the wind speed in front of the hub of the current i-th wind turbine; is the ambient wind speed; At wind speed , the wind direction angle is Under the wind condition, the wind speed in front of the hub of the current wind turbine and the power curve of the current wind turbine are obtained, and the output power of the current wind turbine under this wind condition is obtained; wherein the power curve of the wind turbine includes the corresponding relationship between the wind speed in front of the hub and the output power of the wind turbine.

[0066] According to the output power corresponding to this wind condition and the method described in the following formula, the total output power of the wind farm under the current wind farm arrangement scheme is obtained: (30) in, is the total output power of the wind farm under the current wind farm arrangement scheme; The wind speed is , wind direction is wind conditions; For unit i in wind condition Output power under For wind conditions Frequency of occurrence; is the number of wind directions; is the number of wind speed segments taken in a single wind direction; N is the number of units.

[0067] In a specific embodiment, in step (4), the step of determining the optimal total power generation and total cable cost of the wind farm arrangement corresponding to the optimization individual specifically includes: The vertices at the wind farm area boundary are , , as well as 40 Vestas-V80 wind turbines are installed in the polygonal area (coordinate dimension is meter). The original wind farm unit locations are as follows: Figure 4 The basic parameters of the wind farm are shown in Table 1, and the power and thrust curves of the wind turbines are shown in Figure 3 As shown in (b), the evolution curve of the optimization process is Figure 5 shown.

[0068] Table 1 Wind farm parameters

[0069] Select wind conditions with multiple wind directions and wind speeds. For the probability values ​​of each wind direction and wind speed, refer to Figure 2 . Figure 2 is a wind rose diagram according to an implementation of an embodiment of the present invention, Figure 2 The central angle of the polar coordinate bar graph represents the wind direction angle, and the height of the bar graph represents the wind frequency. The optimization constraints are the wind farm area and the safety distance between any two wind turbines must be greater than 4 times the diameter of the wind rotor. In this example, a flat terrain is selected, and the influence of complex terrain is not considered. The wind farm layout-cable path is jointly optimized by the method of steps (1) to (6) in the above method embodiment, such as Figure 5 The figure shows the maximum output power of each wind farm arrangement scheme in the population at each iteration during the optimization process. It can be seen that the optimized population eventually converges. Figure 6 An optimal unit point on the Pareto frontier of a wind farm in an embodiment of the present invention shown in FIG. Figure 4 The original unit location of the wind farm in the embodiment of the present invention. Figure 6 It can be seen that the optimal unit location of the wind farm has an obvious parallelogram regular arrangement feature. The annual power generation corresponding to this optimal unit location is compared with Figure 4 The annual power generation corresponding to the original unit position of the wind farm in the optimized parallelogram regular arrangement with vacancies has an annual power generation increase of about 3.695%; at the same time, the cable cost of the total feeder cable is reduced by about 6.172% compared with the original cable layout. Therefore, the present invention can significantly improve the power generation of the wind farm throughout its life cycle and reduce the total cable cost of the total feeder cable while ensuring that the unit positions of the wind farm are arranged in a parallelogram regular pattern.

[0070] The wind farm parallelogram regular layout-cable joint optimization method of the present invention aims to solve the problem that in the process of designing the unit points of the existing wind farm, it is difficult to achieve the mathematical optimality of the wind turbine unit points by relying on experience, and the completely random layout lacks aesthetics and is not conducive to the design of the operation and maintenance path; at the same time, the single pursuit of higher power generation fails to fully consider the impact of the cable cost of the wind farm. To this end, the present invention applies a genetic algorithm to maximize the target value of the first optimization target and minimize the target value of the second optimization target as the optimization target, and performs multi-objective optimization on the unit arrangement scheme of the wind farm. And the unit coordinate variation process and probability of two scales of large and small are used, and the "two-stage" variation probability setting scheme is used to ensure that the optimization target value can converge quickly and fully. Finally, the arrangement optimization scheme of the wind farm can be determined on the multi-objective optimization Pareto frontier solution obtained according to actual needs. Through the above method, the present invention comprehensively considers the total output power and cable cost of the wind farm, while ensuring the total output power of the wind farm, reducing the cable cost of the wind farm and improving the comprehensive benefits of the wind farm throughout its life cycle.

[0071] Glossary Micro-site selection for wind farms: the planning and design process within the wind farm planning area, including the location of wind turbines and the layout of collector lines; Pareto front: a series of solutions obtained through multi-objective optimization that can dominate all solutions outside the Pareto front, but do not dominate each other; Speed ​​loss: the attenuation of the wind speed behind the wind turbine rotor relative to the free flow at an infinite distance in front of the wind rotor; Additional flow turbulence intensity: the increase in the flow turbulence intensity behind the wind turbine rotor relative to the ambient turbulence intensity at an infinite distance in front of the wind rotor; Real number coding: all elements in the variable are real numbers; WT: wind turbine; OSS: offshore substation; MILP: mixed integer linear programming method; NSGA-II: non-dominated sorting genetic algorithm with elite strategy; CIC: cable installation cost.

[0072] The present invention also discloses a computer program product, including a computer program, which executes the steps of the above method when executed by a processor. The present invention further discloses a computer-readable storage medium, on which a computer program is stored, which executes the steps of the above method when executed by a processor. The present invention also discloses a wind farm parallelogram regular layout machine-cable joint optimization system, including an interconnected memory and a processor, on which a computer program is stored, which executes the steps of the above method when executed by a processor. The product, medium and system of the present invention correspond to the above method and also have the advantages described in the above method.

[0073] The present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiment can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable storage media include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing computer programs and / or modules stored in the memory, and calling data stored in the memory. The memory may include a high-speed random access memory and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0074] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A wind farm parallelogram regular layout machine-cable joint optimization method, characterized in that: Includes steps: Obtain the regional scope of the wind farm to be optimized, the model and quantity of the units, the wind resource conditions, and the values ​​of various parameters required for the cable layout design; According to the regional scope of the wind farm and the number of wind turbines, shape variables of the parallelogram rule are randomly generated, and an initial population of shape variables of the genetic algorithm is established; wherein the coordinate initial population contains multiple individuals, each of which is a set of shape variables of the parallelogram rule, and a corresponding uniquely determined arrangement scheme of wind farm units of the parallelogram rule can be generated based on the point selection program; Based on the generated wind farm unit arrangement plan, the booster stations in the field are randomly generated within the range that meets the optimization constraints, and the initial population of booster stations of the genetic algorithm is established; wherein the initial population of booster stations includes multiple individuals, each of which is a wind farm booster station location plan; Taking maximizing the target value of the first optimization target and minimizing the target value of the second optimization target as the optimization target, starting from the initial population of shape variables and the initial population of booster stations, new parallelogram-ruled shape variables and booster station coordinates are generated to form an optimization population. Based on the genetic algorithm, multi-objective optimization is performed on the wind farm unit arrangement scheme and the booster station location scheme; the first optimization target is the total output power of the wind farm under the arrangement scheme corresponding to each individual; the second optimization target is the cable cost of the wind farm under the arrangement scheme corresponding to each individual; When both the first optimization objective and the second optimization objective have reached convergence, the optimization process ends and the Pareto frontier of multi-objective optimization is obtained; each individual on the Pareto frontier is a definite wind farm unit arrangement and booster station location scheme, including the coordinates of each unit and booster station; On the Pareto frontier, the optimal layout of wind farms is selected according to demand.

2. The wind farm parallelogram regular layout-cable joint optimization method according to claim 1 is characterized in that: According to the regional scope of the wind farm and the number of wind turbines, the specific steps for randomly generating shape variables of the parallelogram rule are as follows: First, set the constraints of wind turbine coordinates; the constraints include the preset wind farm area range and the distance between any two wind turbines in the wind farm is greater than , that is, the wind turbine rotor diameter of times, of which is a positive integer greater than 1; Secondly, within the closed polygonal area of ​​the wind farm, a point coordinate is randomly selected as the global center point coordinate of the parallelogram regular layout. ; Again, according to the uniform probability distribution, Randomly select the angle between two sides of a parallelogram within the range and the overall rotation angle of the parallelogram , based on unit spacing restrictions A certain multiple range of Randomly pick the length of the parallelogram side and ,exist Randomly extract the gradient coefficient between rows within the range and Two parameters, in Randomly extract the gradient coefficient within the line within the range and Two parameters; Finally, according to the determined number of wind turbines, the shape variables of the generated parallelogram rule are used; according to the point-taking procedure, the corresponding unique wind farm unit layout plan is generated within the wind farm area, and the process is repeated until the total number of units in the unit layout plan meets the number of units.

3. The wind farm parallelogram regular layout-cable joint optimization method according to claim 2 is characterized in that: According to the point selection procedure, the specific steps for generating the corresponding unique wind farm unit arrangement plan within the wind farm area are as follows: Shape variables based on the parallelogram rule , and the closed polygonal range of the wind farm area, extract the global center point coordinates ;by is a vertex, the horizontal line is an edge, and rotates counterclockwise , draw a straight line that intersects the entire polygon, record it as line 1#, and then As the vertex, with line 1# as an edge, rotate counterclockwise Angle, and then draw a straight line to intersect the entire polygon, record it as the straight line ; Taking line 1# as the reference, from the global center point coordinates Starting from the point, according to the gradient point method, points are taken on both sides, and then several parallel lines of line 1# are drawn through the points to intersect with the closed area; then, with line m# as the reference, according to the gradient point method, points are taken on both sides, and several parallel lines of line m# are drawn through the points to intersect with the closed area; The intersection of these two sets of parallel lines is denoted as , forming a series of parallelograms, let the side length of the parallelogram on line 1# and its parallel lines be a, let the side length of the parallelogram on line m# and its parallel lines be b, and the distance between the two parallel lines of the parallelogram =a , =b ; The collection of all intersection points As the shape variable of the current parallelogram rule The corresponding unique wind farm unit arrangement plan.

4. The wind farm parallelogram regular layout-cable joint optimization method according to claim 3 is characterized in that: Taking line 1# as the reference, from the global center point coordinates Starting from here, according to the gradient point selection method, select points on both sides, and then draw several parallel lines of line 1# through the points. The specific steps are: The longest hypotenuse distance of the wind farm boundary is , then you need to pass through the global center point , along the perpendicular direction of line 1#, the number of intervals of gradient points, that is, the number of straight lines, is at most ,in For not more than The largest integer of ; Consider a Gaussian distribution function with a mean of 0 ,exist Uniform extraction within the range Number - , then we can get No less than A collection of gradient distances: ； in and is the parallelogram line spacing gradient coefficient, all gradient points form a set ; Draw a line parallel to straight line l# through the selected gradient point.

5. The wind farm parallelogram regular layout-cable joint optimization method according to claim 3 is characterized in that: Taking the straight line m# as the reference, we also use the gradient point method to select points on both sides and draw several parallel lines through the points to the straight line m#. The specific process is as follows: The longest hypotenuse distance of the wind farm boundary is , then you need to pass through the global center point , along the perpendicular direction of the straight line m#, the number of intervals of gradient points, that is, the number of straight lines, is at most ,in For not more than The largest integer of ; Consider a Gaussian distribution function with a mean of 0 ,exist Uniform extraction within the range Number - , then according to the Gaussian distribution function we get No less than A collection of gradient distances: ； in and is the parallelogram line spacing gradient coefficient, all gradient points form a set ; Draw a line parallel to line m# through the chosen gradient point.

6. The wind farm parallelogram regular layout-cable joint optimization method according to any one of claims 1 to 5, characterized in that: Based on the generated wind farm unit arrangement plan, the steps of randomly generating the on-site booster stations within the range that meets the optimization constraints and establishing the initial population of booster stations of the genetic algorithm specifically include: First, set the selection constraints of the booster station; the constraints include the preset wind farm area range and the setting of the booster station to be greater than the distance of any wind turbine in the wind farm. , which is n times the diameter of the wind turbine rotor, where n is a positive integer greater than 1; Randomly select points within the wind farm area to generate the coordinate position vector of the booster station , repeat the process to meet the point constraints of the booster station.

7. The wind farm parallelogram regular layout-cable joint optimization method according to any one of claims 1 to 5, characterized in that: When the first optimization objective and the second optimization objective have reached convergence, the optimization process ends. The steps to obtain the Pareto frontier of multi-objective optimization include: According to the genetic algorithm, the coordinates of the wind turbine and the booster station are encoded with real numbers, and the coordinate variation processes and probabilities of the wind turbines with large and small scales are set respectively, that is, the large-scale variation process randomly selects new coordinates within the range of 2 / 3 times the diameter of the wind rotor of the original coordinates, and the small-scale variation process randomly selects new coordinates within the range of 1 / 6 times the diameter of the wind rotor of the original coordinates. At the beginning of the optimization, both large-scale and small-scale unit position coordinate variation processes are used. When the first optimization objective and the second optimization objective converge for the first time, the large-scale variation probability is set to 0, and only the small-scale unit position coordinate variation process is used until the first optimization objective and the second optimization objective converge for the second time. The optimization process is terminated to obtain the Pareto frontier of multi-objective optimization.

8. The wind farm parallelogram regular layout-cable joint optimization method according to any one of claims 1 to 5, characterized in that: Based on the average wind speed and turbulence intensity of the incoming flow of the wind farm and considering the wake effect, the wake model is used to calculate the output power of each wind turbine in the wind farm and the flow turbulence intensity, and the total output power of the wind farm is obtained.

9. The wind farm parallelogram regular layout machine-cable joint optimization method according to claim 8, characterized in that: The specific steps to obtain the total output power of the wind farm are: For each wind turbine in the wind farm, the velocity loss of the upstream wind turbine at multiple points on the rotor of the current wind turbine is calculated according to formula (1): （1） in, is the upper point of the wind rotor of the current i-th unit The inflow velocity loss at is the point on the wind rotor of the i-th unit of the j-th unit Average speed loss at is a binary variable; N is the number of wind turbines in the wind farm; For each wind turbine in the wind farm, the additional flow turbulence intensity of the upstream wind turbine at multiple points on the rotor of the current wind turbine is obtained according to formula (2): （2） in, is the additional flow turbulence intensity at the point (x, y, z) on the wind rotor of the current i-th unit; is the additional flow turbulence intensity of the jth unit at the point (x, y, z) on the rotor of the i-th unit; is a binary variable; is the number of wind turbines in the wind farm; According to the inflow velocity loss of multiple points on the wind turbine rotor, the average value is taken, and the wind speed in front of the hub of the current wind turbine under the preset wind condition is obtained according to formula (3): （3） in, is the wind speed in front of the hub of the current i-th wind turbine; is the ambient wind speed; At wind speed , the wind direction angle is Under the wind condition of the wind turbine, the wind speed in front of the hub of the current wind turbine and the power curve of the current wind turbine are obtained, and the output power of the current wind turbine under the wind condition is obtained; wherein the power curve of the wind turbine includes the corresponding relationship between the wind speed in front of the hub and the output power of the wind turbine; According to the corresponding output power under this wind condition and formula (4), the total output power of the wind farm under the current wind farm arrangement scheme is obtained: （4） in, is the total output power of the wind farm under the current wind farm arrangement scheme; The wind speed is , wind direction is wind conditions; For unit i in wind condition Output power under For wind conditions Frequency of occurrence; is the number of wind directions; is the number of wind speed segments taken in a single wind direction; N is the number of units.

10. The wind farm parallelogram regular layout-cable joint optimization method according to any one of claims 1 to 5, characterized in that: The specific steps for calculating the cable cost of the wind farm under each individual arrangement scheme include: Based on the location of the wind farm booster station and the coastline, the coast point closest to the booster station is obtained as the feeder cable node; Based on the unit location and booster station location under the wind farm layout plan, as well as the cable information of various types of farms, and coupling the feeder cable node and feeder cable model information, the total cable cost model of the wind farm is established, and the total cable cost of the wind farm is solved: (5) in, is the cable unit price, Representative Points and The Euclidean distance between is a given safety factor.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are performed.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program performs the steps of the method according to any one of claims 1 to 10.

13. A wind farm parallelogram regular layout machine-cable joint optimization system, comprising a memory and a processor connected to each other, wherein a computer program is stored in the memory, characterized in that: When the computer program is executed by a processor, the computer program performs the steps of the method according to any one of claims 1 to 10.

Images