Automatic topology identification method for complex regional power grid structure

Abstract

The invention discloses an automatic topology identification method for a grid structure of a complex regional power grid, and the method comprises the steps: carrying out the traversal search in a regional power grid, finding out all circuit breakers and disconnecting switches, and carrying out the numbering; the adjacent circuit breakers are searched and assigned to be 1, the non-adjacent circuit breakers are assigned to be 0, and a group of vectors are formed; constructing a circuit breaker association adjacency matrix for the regional power grid, that is, when the assignment of the corresponding number of the adjacency vector matrix is 1, associating every two circuit breakers with the assignment as 1; selecting a plurality of flow vectors, selecting a next target node according to a roulette algorithm, calculating path pheromones, and updating the roulette probability of the target node; after certain iteration and updating, when the minimum path is smaller than a threshold value, the path is closed; and finally, automatic topology identification of the regional power grid structure is obtained. According to the method, automatic topology identification of the grid structure of a complex regional power grid or an isolated power grid can be realized, and a more accurate judgment basis is provided for relay protection and a stability control strategy.

Description

[Technical field]

[0001] The invention belongs to the technical field of relay protection of electric power systems and relates to an automatic topology identification method for a grid structure of a complex regional power grid. [Background technology]

[0002] The present invention aims at the complex grid structure and diverse operation modes of regional power grids. The traditional identification method cannot fully traverse all operation modes, which increases the risk of misjudgment of operation modes and brings hidden dangers to the safe and stable operation of regional power grids. In some regional power grids in China, there are many cases of malfunction of on-site power systems due to the failure to timely identify the grid structure topology. Therefore, it is urgent to realize automatic topological identification of the grid structure of complex regional power grids.

[0003] In the prior art, in the invention patent CN115730410A, the sag line segment function is determined based on the distance between each adjacent tower of multiple tower nodes, and the topology of the distribution network is obtained based on the curve of the overhead line. This invention patent does not fully consider the physical relationship between different towers at the same distance under a complex power grid, and it is difficult to achieve full coverage of the physical relationship of the entire grid. In the invention patent CN111666456A, the ledger data and relationship data of each distribution network system are selected, the data model and data relationship of each system are analyzed, and a unified meta-model is constructed to realize the topological identification of the power grid architecture. It is still based on a relational database to store power grid data, perform association analysis, and parse multiple lists at the same time to perform matching processing, which has a large workload and low efficiency, and the risk of mismatch in the identification of the grid structure is relatively high.

[0004] At present, there is still a lack of a more reasonable, effective and rapid identification method for automatic topology identification of power grid structures in complex areas. [Summary of the invention]

[0005] The purpose of the present invention is to provide a method for automatic topology recognition of complex regional power grid structures, so as to solve the risks of complex identification process of regional power grid topology structure and misjudgment of operation mode, realize automatic topology of regional power grid structure, and provide a simple and fast judgment basis for relay protection.

[0006] The purpose of the present invention is achieved through the following technical solutions:

[0007] A method for automatic topology identification of a complex regional power grid structure, characterized by the following steps:

[0008] 1) Draw a primary electrical component wiring diagram of the regional power grid, perform a traversal search on the drawn primary electrical component wiring diagram of the regional power grid, find out all the circuit breaker closing and isolating switches in the regional power grid, and number each circuit breaker closing and isolating switch;

[0009] 2) For each numbered circuit breaker, find its adjacent circuit breaker and assign it a value of 1, and assign a value of 0 to non-adjacent circuit breakers to form a set of vectors. For example, there are 8 circuit breakers in the regional power grid, numbered 1 to 8. The adjacent vector of circuit breaker number 1 is expressed as:

[0010] Table 1:

[0011] breaker No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 No. 8 No. 1 0 1 1 0 0 0 0 0

[0012] 3) Construct an adjacency matrix based on the adjacency vectors of all circuit breakers. That is, when the value of the corresponding number in the adjacency vector matrix is ​​1, the association between each pair of circuit breakers is 1. For example, an 8*8 adjacency matrix is ​​constructed.

[0013] 4) Construct several flow vectors to find the path from the starting circuit breaker to the target circuit breaker. In the initial stage, since there is no pheromone, the walking path of any flow vector is random. During the walking process, it will continuously release pheromones to mark the path it has passed.

[0014] According to the adjacency matrix, any flow vector selects the next breaker node from the current breaker node among the selectable next nodes according to the roulette algorithm. At the same time, pheromones related to the path length are released.

[0015] Since the behavior trajectory of the flow vector is randomly distributed, the number of flow vectors on the short path is greater than the density of flow vectors on the long path per unit time, and the pheromone concentration left by the short path is also higher. Update the pheromone on each path and update the probability value of each roulette area, such as Figure 2 shown.

[0016] Assuming that the total number of flow vectors is M, each flow vector moves in the regional power grid adjacency distance matrix and selects the next circuit breaker according to the state transition rule. Assuming that at time t, flow vector k is located at circuit breaker i, then the probability that flow vector k selects the next circuit breaker j is:

[0017]

[0018] Where: V represents the set of next circuit breakers that flow vector k can select; α is the pheromone concentration heuristic factor. The larger the α, the more the flow vector K tends to choose the path that most flow vectors have traveled; β represents the expected heuristic factor, which reflects the role of visibility information in the flow vector's selection of the next position. The larger the β value, the more the flow vector k tends to choose the circuit breaker close to the target point.

[0019] 5) Set the flow vector and the number of loop iterations, compare the shortest path from the starting circuit breaker to the target circuit breaker through the path collection of all flow vectors, and update the shortest path sequence chain.

[0020] The shortest path between the starting circuit breaker and the target circuit breaker is compared with the set threshold value. If it is less than the threshold value, it means that the path between the two circuit breakers is closed. Otherwise, the circuit breakers are open-circuited, and finally the automatic topology identification of the regional power grid structure is obtained.

[0021] The beneficial effects of the present invention are:

[0022] The present invention avoids the relationship association between global circuit breakers by traversing and searching the edited electrical wiring diagram, constructs a state adjacency matrix only through the operating status between adjacent circuit breakers, and finds the minimum path from the starting circuit breaker to the target circuit breaker through a reasonable roulette algorithm. In this process, pheromones and the probability of roulette on each path are updated in time to achieve the minimum optimization principle, realize the automatic topology of the regional power grid structure, and provide a simple and fast judgment basis for relay protection.

Brief Description of the Drawings

[0023] Figure 1 Design a block diagram for the logic of the overall algorithm;

[0024] Figure 2 A schematic diagram of the roulette algorithm for the direction of the tide;

[0025] Figure 3 It is a schematic diagram of the regional power grid structure system;

[0026] [Specific implementation method]

[0027] The specific implementation of the present invention will be further described below in conjunction with the accompanying drawings.

[0028] A method for automatic topology recognition of complex regional power grid structures, identifying each circuit breaker in the regional power grid and its relationship with other circuit breakers, Figure 1 The logic design block diagram of the overall algorithm is as follows:

[0029] (1) Use a dedicated graphic editing tool to draw a wiring diagram of the primary electrical components of the regional power grid, such as Figure 3 As shown, the electrical components that need to be drawn include circuit breakers, transformers, disconnectors, busbars, grid-connected lines, etc.

[0030] (2) Traverse the state of each circuit breaker in the system wiring diagram and number them to ensure the uniqueness of the numbers. When the power grid capacity is increased, draw the new components in the wiring diagram and search again, but ensure that the numbers of the existing circuit breakers and isolating switches remain unchanged. Figure 3 , the circuit breakers are numbered 1-8.

[0031] (3) Based on Figure 2The state constructs the adjacency matrix, when two adjacent circuit breakers are set to 1, otherwise it is 0. As shown in Table 1.

[0032] Table 1: Construction of regional power grid adjacency matrix

[0033] breaker No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 No. 8 No. 1 - 1 0 0 0 0 1 0 No. 2 1 - 1 0 0 0 1 1 No. 3 0 1 - 0 0 0 0 1 No. 4 0 0 0 - 1 0 1 0 No. 5 0 0 0 1 - 0 1 1 No. 6 0 0 0 0 0 - 0 0 No. 7 1 1 0 1 1 0 - 0 No. 8 0 1 1 0 1 0 0 -

[0034] (4) Set the number of flow vector iterations to 100 and the number of flow vectors to 50, and initialize the roulette probability of each node as shown in Table 2.

[0035] Table 2: Roulette initialization probability of each node in the regional power grid

[0036] breaker No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 No. 8 No. 1 0 50％ 0 0 0 0 50％ 0 No. 2 25％ 0 25％ 0 0 0 25％ 25％ No. 3 0 50％ 0 0 0 0 0 50％ No. 4 0 0 0 0 50％ 0 50％ 0 No. 5 0 0 0 33.30％ 0 0 33.30％ 33.30％ No. 6 0 0 0 0 0 0 0 0 No. 7 25％ 25％ 0 25％ 25％ 0 0 0 No. 8 0 33.30％ 33.30％ 0 33.30％ 0 0 0

[0037] Set the starting circuit breaker node to 1 and the target circuit breaker node to 3; construct the algorithm function:

[0038] OnAcoNet(intNoBktol,int**MatrixAco,intFirAdd,int LastAdd)

[0039] Where: NoBktol represents the number of circuit breaker nodes in the regional power grid

[0040] MatrixAco represents the regional power grid adjacency matrix

[0041] FirAdd indicates the starting circuit breaker number address

[0042] LastAdd indicates the target circuit breaker number address.

[0043] Assuming that the total number of flow vectors is M, each flow vector moves in the regional power grid adjacency distance matrix and selects the next circuit breaker according to the state transition rule. Assuming that at time t, flow vector k is located at circuit breaker i, then the probability that flow vector k selects the next circuit breaker j is:

[0044]

[0045] Where: V represents the set of next circuit breakers that flow vector k can select; α is the pheromone concentration heuristic factor. The larger the α, the more the flow vector K tends to choose the path that most flow vectors have traveled; β represents the expected heuristic factor, which reflects the role of visibility information in the flow vector's selection of the next position. The larger the β value, the more the flow vector k tends to choose the circuit breaker close to the target point.

[0046] 41) The starting address of the circuit breaker is 1. After the roulette algorithm, the next node circuit breaker number can be selected as 2 or 7.

[0047] 42) The variable S records the path distance of the current flow vector, and at this time S = 1; when the circuit breaker number is selected as 2, it is necessary to select the next node. Through the roulette algorithm, the next node can be selected as circuit breaker number 3 or 7, 8.

[0048] 43) If the circuit breaker number is 3, the node is consistent with the target circuit breaker node, indicating that the current flow vector has reached the target circuit breaker, and the path variable is updated to 2.

[0049] 44) Update the path pheromones and volatiles according to formula (1), and update the roulette probability of passing nodes.

[0050] PP[l]=pow(Tau[W-1][LJD[l]-1],Alpha)*pow((Eta[LJD[l]-1]),Beta);

[0051] pp_sum+=PP[l];

[0052] Where PP[l] represents the roulette probability calculation of node l.

[0053] Repeat steps 41) and 44) ​​to complete all flow vectors and iterations, and then compare and obtain the minimum path S_min between the start circuit breaker address and the target circuit breaker address, and record the circuit breaker nodes passed under the minimum path Route.

[0054] S_min=(S <S_min)?S:S_min;;

[0055] Route_SH = Route[];

[0056] (5) Compare the minimum path S_min with the threshold. Once the minimum path S_min is smaller than the threshold, it can be determined that the path from the starting circuit breaker node to the target circuit breaker node is in a closed-loop operation.

[0057] The final result is that the shortest distance from the starting circuit breaker 1 to the target short-circuit node is 2, and the circuit breaker nodes passed include number 1, number 2, and number 3.

[0058] The association relationship between the circuit breakers in the regional power grid is obtained to achieve automatic topology recognition.

[0059] The present invention searches and traverses all circuit breakers in the grid structure of a complex regional power grid, constructs a state adjacency matrix, optimizes and decomposes the grid structure through an intelligent roulette algorithm, determines the association between each pair of circuit breakers, and finally constructs the grid topology structure of the regional power grid. The present invention realizes automatic topological recognition of the grid structure of a complex regional power grid or an isolated power grid, providing a more accurate judgment basis for relay protection and stabilization control strategies.

Claims

1. A method for automatic topology identification of a complex regional power grid structure, characterized in that: Here are the steps: 1) Draw a primary electrical component wiring diagram of the regional power grid, perform a traversal search on the drawn primary electrical component wiring diagram of the regional power grid, find out all the circuit breaker closing and isolating switches in the regional power grid, and number each circuit breaker closing and isolating switch; 2) For each numbered circuit breaker, find its adjacent circuit breaker and assign it a value of 1, and assign a value of 0 to non-adjacent circuit breakers to form a set of vectors; 3) Construct a circuit breaker association adjacency matrix for the regional power grid, that is, when the value of the corresponding number in the adjacency vector matrix is ​​1, the association between two circuit breakers is assigned to 1; 4) Select several flow vectors, select the next target node according to the roulette algorithm, calculate the path pheromone, and update the roulette probability of the target node. The number of flow vectors on the short path is greater than the density of flow vectors on the long path. The higher the pheromone concentration left by the short path, the higher the roulette probability value of the node path; 5) After a certain number of iterations and updates, the minimum path from the starting circuit breaker node to the target circuit breaker node is counted. When the minimum path is less than the threshold, it means that the path is closed; finally, the automatic topology recognition of the regional power grid structure is obtained.

2. The method for automatic topology identification of a complex regional power grid structure according to claim 1 is characterized by: In step 4), in the initial stage, there is no pheromone, and the walking path of any flow vector is random. During the walking process, pheromones are continuously released to mark the path it has passed; According to the adjacency matrix, any flow vector selects the next breaker node from the current breaker node among the selectable next nodes according to the roulette algorithm, and at the same time, releases pheromones related to the path length; The behavior trajectory of the flow vector is randomly distributed. In a unit of time, the number of flow vectors on a short path is greater than that on a long path, and the concentration of pheromones left by the short path is also higher. The pheromones on each path are updated, and the probability values ​​of each roulette area are updated. Assuming that the total number of flow vectors is M, each flow vector moves in the regional power grid adjacency distance matrix and selects the next circuit breaker according to the state transition rule. Assuming that at time t, flow vector k is located at circuit breaker i, then the probability that flow vector k selects the next circuit breaker j is: Where: V represents the set of next circuit breakers that flow vector k can select; α is the pheromone concentration heuristic factor. The larger the α, the more the flow vector K tends to choose the path that most flow vectors have traveled; β represents the expected heuristic factor, which reflects the role of visibility information in the flow vector's selection of the next position. The larger the β value, the more the flow vector k tends to choose the circuit breaker close to the target point.

3. The method for automatic topology identification of complex regional power grid structure according to claim 1 is characterized by: In the step 1), the electrical components drawn include circuit breakers, isolating switches, busbars and grid-connected lines, and are numbered.

4. The method for automatic topology identification of a complex regional power grid structure according to claim 3 is characterized in that: When the capacity of the regional power grid is increased, the newly added components are drawn in the electrical component wiring diagram and a new search and traversal is performed again to ensure that the numbers of the existing circuit breakers and isolating switches remain unchanged.

Images