Wind power plant current collection system topology optimization method based on adaptive minimum spanning tree

Abstract

The invention discloses a wind power plant current collection system topology optimization method based on an adaptive minimum spanning tree, and belongs to the technical field of calculation, reckoning or counting. The method comprises the following steps: determining coordinate positions of a transformer substation and each fan in a wind power plant; executing a first round of circulation of the algorithm, initializing the radiation type topology, and selecting a proper branch to be added into a tree network by taking the shortest distance as an optimization target; executing the mth cycle of the algorithm, and on the basis of completing branch cross validation, obtaining the downstream condition of each node of the current tree network by using the adaptive algorithm, calculating the current corresponding to each branch, matching the cable, and calculating the cable cost and the long-term network loss cost corresponding to each branch; selecting branches to be added in the mth cycle by taking the minimum investment cost of the main cable of the tree network and the minimum long-term network loss cost as a target; and repeatedly executing the loop until the set of unconnected nodes becomes an empty set. The method provides an innovative strategy and theoretical support for topological optimization of the wind power plant current collection system.

Description

TECHNICAL FIELD

[0001] The present application relates to the topology optimization technology of wind farm power collection system, and particularly discloses a wind farm power collection system topology optimization method based on adaptive minimum spanning tree, and belongs to the technical field of calculation, estimation or counting. BACKGROUND

[0003] At present, the topology structure design of wind farm power collection system generally adopts the traditional minimum spanning tree algorithm based on graph theory, and the core optimization target is usually set as the shortest total length of cable path. However, this method has many inherent defects. First, the traditional minimum spanning tree algorithm is difficult to effectively adapt to the selection problem of multiple cable specifications, i.e. cross sections, in the optimization process. In actual engineering, in order to balance the initial investment and transmission loss, different cross section cables are often selected in different sections according to the current load, and the traditional minimum spanning tree algorithm with a single optimization target cannot optimize the combination of such multi-specification cables simultaneously in topology construction, resulting in that the obtained optimization result may be only a local optimum or cannot meet the actual economic requirements. Second, the traditional minimum spanning tree algorithm lacks reasonable constraints on the spatial layout of cable paths in the iteration process, such as avoiding intersection constraints, resulting in that the generated topology structure may have cable intersection in the plane or three-dimensional space. Cable intersection not only increases the risk of electromagnetic interference and potential physical damage, but also brings great difficulty and safety risk to the cable detection, positioning, maintenance and replacement work of the wind farm in the later stage, significantly increasing the operation and maintenance complexity and cost in the whole life cycle. Although some researches use the improved dynamic minimum spanning tree (DMST) method to optimize the topology of power collection system, both DMST and the traditional minimum spanning tree method have two defects: first, DMST or the traditional minimum spanning tree method only focuses on minimizing the cable investment cost, without considering the long-term network loss after the operation of the wind farm. The present value of the network loss cost during the operation of the power collection system as a long-term power network is huge, especially the cumulative amount is considerable in the life cycle of the wind farm. Ignoring this factor makes the traditional method seriously insufficient in considering the overall economy of the whole life cycle of the system, and it is difficult to achieve real cost optimization; second, each round of loop of DMST or the traditional minimum spanning tree method needs to calculate the total cost of all remaining edges, i.e. tree branches, first, and then sort them, and then start from the minimum cost optional edge, and cross-check all remaining edges with each edge already joined in the tree. However, this method has low calculation efficiency, especially in the case of a large number of cables or considering cable network loss in each round of loop, which will greatly increase the calculation burden. SUMMARY

[0004] The application aims at solving the problems of the prior art and provides a wind farm power collection system topology optimization method based on adaptive minimum spanning tree.

[0005] The application achieves the above-mentioned application aims by adopting the following technical scheme.

[0006] The wind farm power collection system topology optimization method based on adaptive minimum spanning tree comprises the following steps.

[0007] Step one: determining the coordinate positions of the transformer station node and each wind turbine node in the wind farm, and calculating the Euclidean distances between the coordinate positions of each node.

[0008] Step two: performing the first round of cycle of the adaptive minimum spanning tree algorithm, initializing the radial tree network topology, and selecting the tree branch connected with the transformer station node to join the radial tree network topology with the shortest distance as the target.

[0009] Step three: performing the mth round of cycle of the adaptive minimum spanning tree algorithm, performing tree branch cross verification on the tree branch to be newly added, updating the radial tree network topology according to the newly added tree branch passing the tree branch cross verification, obtaining the downstream condition of each node in the current radial tree network topology by using the adaptive algorithm, matching the cable for each tree branch, calculating the cable cost and long-term network loss cost of each tree branch, and m is an integer greater than or equal to 2.

[0010] Step four: for the newly added tree branch passing the tree branch cross verification in the mth round of cycle of the adaptive minimum spanning tree algorithm, selecting the newly added tree branch that minimizes the sum of the cable cost and long-term network loss cost of the wind farm power collection system to join the radial tree network topology.

[0011] Step five: m = m + 1, returning to step three until all wind turbine nodes are connected to the radial tree network topology.

[0012] As a further optimization scheme of the wind farm power collection system topology optimization method based on adaptive minimum spanning tree, in step one, the calculation of the Euclidean distances between the coordinate positions of each node is specifically as follows: wherein, , are the coordinates of the nodes in the Cartesian coordinate system, respectively. are the coordinates of the nodes in the Cartesian coordinate system, respectively.​​​

[0013] As a further optimization scheme of the wind farm power collection system topology optimization method based on adaptive minimum spanning tree, in step two, the radiation tree network topology is initialized, specifically: T={v sub}, U=V wt , E=empty set, T is the set of connected nodes, U is the set of unconnected nodes, E is the set of tree branches, v sub is the substation node, V wt is the set of wind turbine nodes, V wt ={v1, v2,..., v n}, and n is the number of wind turbines in the wind farm.

[0014] As a further optimization scheme of the wind farm power collection system topology optimization method based on adaptive minimum spanning tree, in step two, the tree branch connected with the substation node is selected to join the radiation tree network topology with the shortest distance as the target, specifically: a wind turbine node v u is selected from the set of unconnected nodes U, the wind turbine node v u satisfies , wherein is the distance between the substation node v sub and the wind turbine node v u .

[0015] As a further optimization scheme of the wind farm power collection system topology optimization method based on adaptive minimum spanning tree, in step three, the tree branch cross verification is performed on the tree branch to be added, specifically: when any prepared tree branch (v a , v b ) and any existing tree branch (v c , v d ) in the set of tree branches E satisfy the line segment crossing condition: , the prepared tree branch (v a , v b ) is an invalid tree branch, and the mth round of loop of the adaptive minimum spanning tree algorithm is ended, wherein the a node v a , the c node v c , and the d node v d are in the set of connected nodes T, and the b node v b is in the set of unconnected nodes U; otherwise, the radiation tree network topology is updated according to the prepared tree branch (v a , v b ).

[0016] As a further optimization scheme of the wind farm power collection system topology optimization method based on adaptive minimum spanning tree, in step three, the cable is matched for each tree branch, specifically: for the downstream connection k ijTyphoon tree branches (v i ,v j ), branches (v i ,v j The rated current of ) is: ,according to Numerical selection of branches (v) i ,v j The corresponding cross-sectional area of ​​the cable, where, For tree branches (v i ,v j The rated current of ), where γ is the safety margin. U is the rated power of the fan. rate The rated voltage of the current collection system, The power factor.

[0017] As a further optimization scheme of the wind farm collector system topology optimization method based on adaptive minimum spanning tree, step three calculates the cable cost and long-term network loss cost of each branch, specifically: , , For tree branches (v i ,v j The cable cost, C ij For I ij The corresponding unit price of the cable. for node , node distance, For tree branches (v i ,v j The long-term network loss cost, For tree branches (v i ,v j The power consumed is τ, where τ is the service life of the wind power.

[0018] As a further optimization scheme of the wind farm collector system topology optimization method based on adaptive minimum spanning tree, in step four, the new tree that minimizes the sum of the cable cost and long-term network loss cost of the wind farm collector system is selected according to the objective function. accomplish.

[0019] An electronic device includes a memory and a processor. The memory stores a computer program that runs on the processor. When the processor runs the computer program, it executes the steps of the above-described wind farm power collection system topology optimization method.

[0020] A computer-readable storage medium having a computer program stored thereon, wherein the computer program executes the steps of the above-described wind farm power collection system topology optimization method when it is run.

[0021] The present application has the following beneficial effects: the present application aims to provide a wind farm power collection system topology optimization method based on adaptive minimum spanning tree, innovatively introduces a line segment crossing condition to determine the cable crossing problem, prevents cable crossing problems in subsequent cycles, avoids difficulties in cable detection, maintenance and replacement during the service process of the wind farm, directly saves the subsequent calculation of cables with crossing conditions, improves the calculation efficiency; at the same time, the present application designs an adaptive algorithm to determine the downstream condition of each node in the tree network in each cycle, provides a basis for economically matching the appropriate cable cross section of each tree branch in the tree network; in addition, the present application takes the minimum total cable investment cost and long-term network loss cost of the tree network as the optimization target under the premise of calculating the investment cost and long-term network loss cost of each cable section, changes the setting of taking the minimum total cable path distance as the optimization target in the traditional minimum spanning tree algorithm, and effectively improves the overall economy of the whole life cycle of the power collection system. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0023] Figure 1 The overall flowchart of the wind farm power collection system topology optimization method based on adaptive minimum spanning tree provided by the present application.

[0024] Fig. 2(a) is a tree branch crossing determination schematic diagram provided in an embodiment of the present application, and Fig. 2(b) is a schematic diagram of matching cables for a tree network with added new tree branches using an adaptive algorithm in an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the present application more clear, the technical solutions of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. The described embodiments are only some embodiments of the present application, not all embodiments. Based on the spirit of the present application, other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0026] As Figure 1 shown, the present application provides a wind farm power collection system topology optimization method based on adaptive minimum spanning tree, which specifically includes the following five steps.

[0027] Step one: determine the coordinate position of substation node and each wind turbine node in wind farm, calculate the Euclidean distance between each node coordinate.

[0028] For wind farm with n wind turbines, let the substation node be v sub , the wind turbine node set be V wt ={v1, v2,..., v n}, and the total node set be .

[0029] Determine the position of each node in set V, calculate the Euclidean distance d(v i , v j ) between any two nodes v i , v j :

[0030] (1)

[0031] In formula (1), (x i , y i ), (x j , y j ) are the coordinates of i node v i , j node v j in Cartesian coordinate system respectively.

[0032] Generate symmetric distance matrix D n×n , where D ij =d(v i , v j ).

[0033] Step two: start to execute the adaptive minimum spanning tree algorithm, initialize the radial topology, namely tree network, hereinafter referred to as tree net, in the first round of loop, select the appropriate tree branch to join the tree net with the shortest distance as the optimization goal.

[0034] When establishing the network topology of wind farm power collection system, the adaptive minimum spanning tree algorithm simulates the process of tree generation, starting from the initial node v sub , uses loop to build radial topology, each round of loop adds a new tree branch to connect a new node to the tree net, until all nodes in set V are included in the tree net.

[0035] First, initialize the tree net:

[0036] The set of connected nodes T={v sub}

[0037] The set of unconnected nodes U=V wt

[0038] The tree net tree branch set E = empty set

[0039] E is represented as a matrix, containing the existing branches in the tree network, and is initially an empty matrix. When the algorithm finishes execution, the branch matrix E is finally an n×2 matrix, where each row represents a branch. For example, a certain row in E (v... i v j () represents a branch in the tree network that extends from node i to node j.

[0040] In the first iteration of the adaptive minimum spanning tree algorithm, the optimization objective is to minimize the distance, starting from the substation node v in the set of connected nodes T. sub Initially, the connection extends to node v in the set of unconnected nodes U. u The first branch (v) is formed sub ,v u ), the node v u The selection must meet the following conditions:

[0041] (2)

[0042] Forming the first branch Then, add v to set T. u Subtract v from set U u Add to E Then the first loop of the adaptive minimum spanning tree algorithm ends.

[0043] Step 3: Execute the m-th iteration of the adaptive minimum spanning tree algorithm, where m ≥ 2. At this point, the tree network branch set E is not empty, and branch crossover verification is required. If the new branch to be added (i.e., a branch not yet added to set E) crosses with any existing branch in set E, then this new branch is invalid, and another new branch that does not cause crossover should be found. The specific method for determining crossover is as follows:

[0044] For any branch to be added (v) a ,v b ) and any branch (v) already exists in E. c ,v d ), node v a , v c ,v d In set T, and node v b In set U, if the condition of line segments spanning each other is satisfied:

[0045] (3)

[0046] The newly added branches (v) a ,v b This will cause an intersection, which is an invalid branch and cannot be added to set E.

[0047] In Cartesian coordinate system, two nodes are subtracted, such as (v a - v b ) is a vector, so the above line segment crossing condition is actually the dot product and cross product calculation of the vector.

[0048] In an embodiment of the present application, the implementation process of cable crossing judgment in step three is illustrated by taking the power collection system topology including a transformer substation and 6 wind turbines as an example. The actual number of wind turbines in a wind farm can be dozens to hundreds. As shown in FIG. 2(a), the red dot represents the transformer substation node v sub , the black dot represents the wind turbine node v1, v2,..., v6, the blue solid line represents the tree branch that has been added to the set E in the previous cycle, the blue dashed line represents the tree branch to be added to the set E in the current cycle, and the thickness of the solid line and the dashed line represents the size of the cable cross-sectional area used by the tree branch. As can be seen, the tree branch (v sub ,v6) to be added to the set E in FIG. 2(a) has a crossing with the existing tree branch (v3,v4) in the set E, so the tree branch (v sub ,v6) is directly excluded, and the subsequent calculation of the cable investment cost and the network loss cost is also omitted, thereby improving the calculation efficiency of the entire method. FIG. 2(b) shows another tree branch (v3,v6) to be added to the set E in the current cycle. Although the addition of the tree branch (v3,v6) will make the tree branch (v sub ,v3) in the set E be equipped with a larger cross-sectional area cable, thereby increasing the cable investment cost, the tree branch (v3,v6) does not form a crossing with any tree branch in the set E, so the tree branch (v3,v6) can be used as a qualified candidate tree branch for the next step of calculating the cable investment cost and the network loss cost.

[0049] The above crossing verification is an improvement of the traditional minimum spanning tree algorithm based on the present patent. Its significance lies in that the power collection system of the actual wind farm cannot have cable crossing. Cable crossing not only increases the risk of electromagnetic interference and potential physical damage, but also brings great difficulty and safety risk to the cable detection, positioning, maintenance and replacement work of the wind farm in the later stage, significantly increasing the operation and maintenance complexity and cost in the whole life cycle.

[0050] After verifying the line segment crossing condition, for the new tree branch that does not cause crossing, an adaptive algorithm is used to match the cable for the tree network adding the new tree branch. The following uses an example to better illustrate, for example: assuming that after adding the new tree branch (v a ,v b ), E becomes the following matrix:

[0051]

[0052] On the basis of traditional minimum spanning tree, the adaptive algorithm is designed, which functions are to determine the downstream condition of each node in the existing tree network, so as to calculate the cable investment cost and long-term network loss cost subsequently, the flow of the adaptive algorithm is as follows:

[0053] ① Determine the search target as the (1,1) element of the matrix E, which is node v in this example sub ;

[0054] ② Search the target in the first column of the matrix E, and obtain the target pointer, that is, in which rows of the matrix E the target exists;

[0055] ③ Record the pointers, and find the second column elements of the corresponding tree branches according to the obtained pointers;

[0056] ④ Take the obtained second column elements as the search target, search the target in the first column of the matrix E, and obtain the target pointer;

[0057] ⑤ Repeat ③-④ until the search result in ④ is empty set, and all the recorded pointers are the downstream condition of the (1,1) element of the matrix E, that is, node v sub ;

[0058] ⑥ Change the search target in ① to the (2,1), (3,1), (4,1),..., (n,1) elements of the matrix E in turn, execute ①-⑤, and obtain the downstream condition of each node in the tree network.

[0059] In the above example of the adaptive algorithm, the downstream pointers of the nodes in the tree network are as follows:

[0060] v sub : 1,2,3,4,5;

[0061] v5: 2,3,4,5;

[0062] v 23 : 3,4,5;

[0063] v7: empty set;

[0064] v a : 5;

[0065] v b : empty set.

[0066] This shows that the downstream of each tree branch in the tree branch matrix E in this example needs to connect 5, 4, 3, 0, 1, and 0 nodes, that is, the fan, so that the optimal cable cross-sectional area required by each tree branch can be further determined.

[0067] Suppose that according to the above adaptive algorithm, a tree branch (v i , v jThe downstream of ) needs to be connected to k ij Typhoon machines, on the other hand, are for tree branches (v i ,v j The corresponding cable has the following rated current:

[0068] (4)

[0069] In equation (4), U is the rated power of the fan. rate U is the rated voltage of the current collector system. rate Typically 33kV, For power factor, Typically, 0.95 is used, and γ is the safety margin, with a value of 1.1.

[0070] Based on the obtained value I ij Based on the cable's size and the cable manufacturer's production data, the corresponding cable cross-sectional area and unit price C per unit length can be obtained. ij Resistance R per unit length of cable ij Then the branches (v i ,v j The investment cost of the cable is W(v). i ,v j )for:

[0071] (5)

[0072] When a wind farm is operating, tree branches (v i ,v j The power consumed P loss (v i ,v j )for:

[0073] (6)

[0074] The service life τ of wind power is generally 20 years. Therefore, during the life cycle of a wind farm, tree branches (v) i ,v j The total network loss, i.e., the long-term network loss cost, is:

[0075] (7)

[0076] Step 4: Based on Step 3, when m≥2 in the m-th iteration of the adaptive minimum spanning tree algorithm, and taking into account both cable investment costs and long-term network loss costs, select a new branch that satisfies the following conditions (v a ,v b Add the new branch (v) to set E, such that the new branch (v) a ,v bAfter the addition, the total cost of the tree corresponding to the set E is the minimum, i.e. in the mth round of circulation, the objective function is:

[0077] (8)

[0078] The constraint condition is:

[0079]

[0080] (9)

[0081] Since the number of wind turbines in the wind farm is limited, usually tens to hundreds, and each round of circulation has a tree cross verification to exclude the infeasible new tree in advance, therefore, in each round of circulation, the optimal new tree can be simply selected by using the enumeration method adopted by the traditional minimum spanning tree algorithm, which will not affect the operation efficiency of the overall algorithm.

[0082] After the optimal tree (v a ,v b ) is selected in the mth round of circulation, v b is added to the set T, v b is subtracted from the set U, (v a ,v b ) is added to the set E, and then the mth round of circulation of the adaptive minimum spanning tree algorithm is ended.

[0083] In an embodiment of the present application, the implementation process of matching the cable for the added new tree by using the adaptive algorithm is described in detail by taking the power collection system including a substation and 6 wind turbines in Fig. 2(b) as an example. After the new tree (v3, v6) is added, E becomes the following matrix:

[0084]

[0085] According to the above adaptive algorithm, the downstream pointers of each node of the tree network are respectively:

[0086] v sub : 1, 2, 3, 4, 5, 6;

[0087] v3: 2, 3, 4, 5, 6;

[0088] v2: 3;

[0089] v1: empty set;

[0090] v4: 5;

[0091] v5: empty set;

[0092] v6: empty set.

[0093] This shows that each branch in the tree matrix E in this example needs to be connected with 6, 5, 1, 0, 1, 0, and 0 nodes downstream, respectively, and thus the optimal cable cross-sectional area required for each branch can be further determined according to formula (4) to formula (9).

[0094] Step five: in the mth round of loop (as a template, m>=2, the loop is repeatedly executed until the set U of unconnected nodes becomes an empty set, at which time the adaptive minimum spanning tree algorithm runs to the mth=n round of loop. The final tree branch set E is the final topology of the wind farm power collection system.

[0095] Generally, the topology optimization of the wind farm power collection system adopts the traditional minimum spanning tree algorithm, which takes the shortest cable laying distance as the target, and gradually expands the tree network from the substation to the entire wind farm through iterative loops. This kind of topology optimization process is only suitable for laying cables of a single type. For a radial network, i.e. a tree network, the current in different sections of the cable is different, therefore, using a single type of cable will affect the investment economy, and the traditional minimum spanning tree algorithm does not consider the long-term network loss cost of the cable. Therefore, the present application makes targeted improvements on the above defects.

[0096] The adaptive minimum spanning tree algorithm changes the setting of the total cable distance as the optimization target from the second round of loop. First, the line segment crossing condition is innovatively introduced for preliminary determination to prevent cable crossing in subsequent loops, avoiding difficulties in cable detection, maintenance and replacement during the service process of the wind farm. Then, an adaptive algorithm is designed to determine the downstream condition of each node in the tree network in each round of loop, providing a basis for economically matching the cross section of the cable for each branch in the tree network. Under the premise of calculating the investment cost and long-term network loss cost of each section of cable, the total cable investment cost and long-term network loss cost are taken as the objective function, and the enumeration method in the traditional minimum spanning tree algorithm is used to find the most suitable new branch for each round of loop. Since the number of wind turbines in the wind farm is controllable, usually tens to hundreds, the use of the enumeration method will not affect the operation efficiency of the overall algorithm.

Claims

1. A topology optimization method for wind farm power collection systems based on adaptive minimum spanning tree, characterized in that, include: Step 1: Determine the coordinates of the substation nodes and each wind turbine node in the wind farm, and calculate the Euclidean distance between each pair of node coordinates. Step 2: Execute the first loop of the adaptive minimum spanning tree algorithm to initialize the radial tree network topology, and select branches connected to the substation nodes to be added to the radial tree network topology with the goal of minimizing distance. Step 3: Execute the m-th iteration of the adaptive minimum spanning tree algorithm, perform branch cross-validation on the branches to be added, update the radial tree network topology based on the new branches that have passed the branch cross-validation, use the adaptive algorithm to obtain the downstream situation of each node in the current radial tree network topology, match cables for each branch, and calculate the cable cost and long-term network loss cost of each branch, where m is an integer greater than or equal to 2. Step 4: For the new branches verified by branch cross-validation in the m-th iteration of the adaptive minimum spanning tree algorithm, select the new branch that minimizes the sum of the cable cost and long-term network loss cost of the wind farm power collection system and add it to the radial tree network topology. Step 5, m = m + 1, return to step 3, until all wind turbine nodes are connected to the radial tree network topology.

2. The wind farm power collection system topology optimization method based on adaptive minimum spanning tree according to claim 1, characterized in that, In step one, the calculation of the pairwise Euclidean distance between each node coordinate is specifically as follows: ,in, , They are respectively node , node Coordinates in the Cartesian coordinate system.

3. The wind farm power collection system topology optimization method based on adaptive minimum spanning tree according to claim 2, characterized in that, In step two, the radial tree network topology is initialized, specifically as follows: T = {v sub }, U=V wt E = empty set, T is the set of connected nodes, U is the set of unconnected nodes, E is the set of tree branches, v sub For substation nodes, V wt V is the set of wind turbine nodes. wt ={v1, v2, ..., v n }, where n is the number of wind turbines in the wind farm.

4. The wind farm power collection system topology optimization method based on adaptive minimum spanning tree according to claim 3, characterized in that, In step two, selecting branches connected to substation nodes and adding them to the radial tree network topology with the shortest distance as the objective is to specifically select wind turbine node v from the set U of unconnected nodes. u The wind turbine node v u satisfy ,in, For substation node v sub With wind turbine node v u The distance.

5. The wind farm power collection system topology optimization method based on adaptive minimum spanning tree according to claim 4, characterized in that, In step three, branch cross-validation is performed on the branches to be added, specifically: any branch to be added (v a ,v b ) and any set of tree branches E containing existing branches (v c ,v d The following conditions must be met for a line segment to cross: At that time, prepare new branches (v a ,v b If node a is an invalid branch, the m-th iteration of the adaptive minimum spanning tree algorithm ends, where node a is an invalid branch. a c node v c d node v d In the set T of connected nodes, node b, v b In the set U of unconnected nodes; otherwise, according to the branch to be added (v) a ,v b Update the radial tree network topology.

6. The wind farm power collection system topology optimization method based on adaptive minimum spanning tree according to claim 5, characterized in that, In step three, matching cables to each tree branch specifically involves: for downstream connection k ij Typhoon tree branches (v i ,v j ), branches (v i ,v j The rated current of ) is: ,according to Numerical selection of branches (v) i ,v j The corresponding cross-sectional area of ​​the cable, where, For tree branches (v i ,v j The rated current of ), where γ is the safety margin. This refers to the rated power of the fan. The rated voltage of the current collection system, The power factor.

7. The wind farm power collection system topology optimization method based on adaptive minimum spanning tree according to claim 6, characterized in that, In step three, the cable cost and long-term network loss cost of each branch are calculated, specifically as follows: , , For tree branches (v i ,v j The cable cost, C ij For I ij The corresponding unit price of the cable. for node , node distance, For tree branches (v i ,v j The long-term network loss cost, For tree branches (v i ,v j The power consumed is τ, where τ is the service life of the wind power.

8. The wind farm power collection system topology optimization method based on adaptive minimum spanning tree according to claim 7, characterized in that, In step four, the new tree branch that minimizes the sum of the cable cost and long-term network loss cost of the wind farm's power collection system is selected based on the objective function. accomplish.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that runs on the processor, characterized in that, When the processor runs a computer program, it executes the steps of the wind farm power collection system topology optimization method according to claim 1.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program runs, it executes the steps of the topology optimization method for the wind farm power collection system as described in claim 1.

Images