Optimal arrangement method of grounding grid measurement points based on information quantity

Abstract

The present invention discloses a method for optimizing the arrangement of measuring points of a grounding grid based on information quantity, and relates to a grounding grid detection technology. The method for optimizing the arrangement of measuring points randomly generates a measuring point arrangement sequence according to known accessible nodes of the grounding grid, calculates the information quantity of the measuring point arrangement sequence to calculate the fitness value, and then calculates the distance between each measuring point arrangement sequence and the alternative arrangement sequence, obtains the final iterative sequence after multiple iterations, and finally extracts the measuring points in the iterative sequence, and predicts the corrosion condition of the grounding grid according to the actual value of the branch resistance. The present invention adopts an optimization algorithm to select representative accessible nodes from accessible nodes as measuring points, thereby reducing the number of nodes that need to be measured and reducing the detection cost. The present invention selects accessible nodes with large amounts of information as measuring points, so that the collected measuring point data is more concentrated.

Description

Technical Field

[0001] The present invention relates to the technical field of grounding grid detection, and in particular to a method for optimizing the arrangement of grounding grid measurement points based on information quantity. Background Art

[0002] As an important part of the substation, the grounding grid plays a key role in the stable operation of the power system. However, due to the material and laying location of the grounding grid, the conductor branch of the grounding grid is prone to corrosion, which seriously reduces the performance of the grounding grid and even threatens the personal safety of the electrical equipment connected to it and the staff in the station. Considering the difficulty of excavation diagnosis of the grounding grid, it is necessary to accurately diagnose the corrosion area without power outage and large-scale excavation of the grounding grid. Patent application publication number CN104931853A discloses a substation grounding grid corrosion diagnosis method based on surface potential distribution. The method calculates the surface potential of non-measurement points according to the surface potential of the measurement point, and then obtains the surface potential distribution. If the surface potential of a conductor segment decreases by more than a threshold, and the surface potentials of other conductor segments in the area all rise slightly, then the conductor segment is judged to be a corroded segment. However, in the actual substation environment, the number of available measurement point grounding leads is limited, which affects the range of diagnosable branches. In addition, the actual grounding grid area is large, and the grounding down conductors are not concentrated. If a grounding down conductor is added at each node of the underground grounding grid for comprehensive measurement, the implementation process will be complicated and the construction cost will be high. Therefore, the present invention proposes a method for optimizing the arrangement of grounding grid measurement points based on information quantity. Summary of the invention

[0003] In order to solve the defects of the above-mentioned prior art, the present invention proposes a method for optimizing the layout of grounding grid measurement points based on the amount of information. Under the condition of limited accessible nodes in the grounding grid, fewer key nodes are selected to reflect the corrosion condition of the grounding grid, thereby reducing the redundancy of unnecessary measurement points.

[0004] The technical solution of the present invention is achieved in this way:

[0005] A method for optimizing the arrangement of grounding grid measuring points based on information quantity comprises the following steps:

[0006] Step 1: The grounding grid has N nodes, and adjacent nodes form D branches. M nodes connected to grounding down conductors are taken as accessible nodes to generate an accessible node set, M≤N;

[0007] Step 2: Randomly select each accessible node as a measurement point or a non-measurement point, generate a measurement point arrangement sequence, generate an initial solution set consisting of K groups of measurement point arrangement sequences, and initialize the current number of iterations q=1;

[0008] Step 3: Arrange a sequence X for each measurement point in the initial solution setk , k=1,2,...,K, select any measuring point in the measuring point arrangement sequence as the reference node, generate the correlation matrix A according to the topological structure of the grounding grid, and then obtain the branch conductance matrix G according to the nominal value of the branch resistance between every two adjacent nodes;

[0009] Step 4: Obtain the node conductance matrix Y through the association matrix A and the branch conductance matrix G, and define the injection current matrix I of the grounding grid J×N , J is the measurement point arrangement sequence X k The number of measurement points;

[0010] Step 5: Through the node conductance matrix Y and the injection current matrix I J×N Calculate the node voltage matrix U N×J , and then calculate the relative voltage matrix U N×J ', according to the branch conductance matrix G, the correlation matrix A, the relative voltage matrix U N×J 'Get the branch current matrix I D×J ;

[0011] Step 6: Calculate the branch current matrix I D×J The branch information amount S of the dth row in d , and then calculate the grounding grid information S, establish the fitness function based on the grounding grid information and find the measurement point arrangement sequence X k The fitness value F k ;

[0012] Step 7: Select the three groups of measuring point arrangement sequences with the smallest fitness values ​​from the K groups of measuring point arrangement sequences as the alternative arrangement sequences, calculate the distance between each measuring point arrangement sequence and the three groups of alternative arrangement sequences, and then calculate the difference sequence between each measuring point arrangement sequence and the three groups of alternative arrangement sequences. k The three difference sequences of the measuring point arrangement sequence X are calculated k The iterative sequence X k ';

[0013] Step 8: If the current number of iterations q is greater than or equal to the maximum number of iterations Q, proceed to step 9, otherwise q=q+1, and use the iteration sequence X k 'Covering measurement point arrangement sequence X k , return to step 3;

[0014] Step 9: Select the measuring point arrangement sequence with the smallest fitness value among the K groups of measuring point arrangement sequences, output the iterative sequence of the measuring point arrangement sequence, and obtain the measuring points for this grounding grid corrosion diagnosis.

[0015] The present invention further includes step 10, extracting measurement points in the iterative sequence, injecting detection current between the measurement points, calculating the actual value of the branch resistance, and predicting the corrosion condition of the grounding grid according to the actual value of the branch resistance.

[0016] In the present invention, in step 2, the measurement point arrangement sequence X k ={x1,...,x m ,...,x M}, if the accessible node m is a measurement point, x m =1; if the accessible node m is a non-measurement point, x m =0.

[0017] In the present invention, in step 4, the node conductance matrix Y=AGA T , connect the current source between the reference node and the measurement point in turn, and determine the injection current matrix I J×N .

[0018] In the present invention, in step 5, the node voltage matrix U N×J =I J×N T Y -1 , the node voltage matrix U N×J The potential of the reference node is considered to be 0, and the relative voltage matrix U is calculated based on the voltages of other nodes relative to the reference node. N×J '.

[0019] In the present invention, in step 6, the branch information amount of the dth row is , b dj is the branch current matrix I D×J The value of the element in the dth row and jth column in the , sets the information threshold λ, if the information amount of any branch S d are greater than or equal to λ, then F k =J+1 / S, if at least one branch has information S d is less than λ, then F k =∞.

[0020] In the present invention, in step 7, three sets of difference sequences X are calculated kα , X kβ , X kξ The mean sequence of pseudo measurement point arrangement sequence X is obtained. k "=(X kα +X kβ +X kξ ) / 3, where α, β, ξ≤K, and the pseudo-measurement point arrangement sequence X k "Each element value in is binarized to generate an iterative sequence X k '.

[0021] The implementation of the method for optimizing the arrangement of grounding grid measurement points based on information quantity of the present invention has the following beneficial effects: the present invention adopts an optimization algorithm to select representative accessible nodes from accessible nodes as measurement points, thereby reducing the number of nodes that need to be measured. The present invention reduces the number of measurement points and the complexity of distribution by optimizing the selection of measurement points, making on-site operations more concise and efficient. The present invention selects accessible nodes with large amounts of information as measurement points, and the accessible nodes with large amounts of information can obtain more current response information, making the collected measurement point data more concentrated. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a schematic diagram of a grounding grid;

[0023] Figure 2 A flow chart of the method for optimizing the arrangement of grounding grid measuring points based on information quantity of the present invention;

[0024] Figure 3 It is a schematic diagram of an equivalent model of a grounding grid according to the present invention;

[0025] Figure 4 This is a topological diagram of the grounding grid of the present invention. DETAILED DESCRIPTION

[0026] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0027] The grounding grid is made of galvanized flat steel or round steel as a metal grounding body, forming a horizontal and vertical mesh structure, which can be used as the grounding line of the substation. Figure 1 As shown, the grounding down conductor is connected to the grounding grid, which plays the role of equalizing pressure and discharging current for the power equipment. The nodes connected to the grounding down conductor are called accessible nodes, and the relevant detection equipment can be connected to these accessible nodes to realize the corrosion diagnosis of the grounding grid. In view of the high requirements for the grounding down conductor in the grounding grid corrosion diagnosis process in the prior art, the grounding grid measurement point arrangement scheme obtained by the present invention can reduce the number of underground leads required for subsequent corrosion diagnosis, reduce the excavation of ground measurement points, and reduce construction costs.

[0028] Embodiment 1

[0029] The present invention provides a method for optimizing the arrangement of measuring points of a grounding grid based on information quantity, which is used for optimizing the arrangement of measuring points of a grounding grid in areas such as substations. Figure 2 As shown, the method comprises the following steps:

[0030] Step 1: The grounding grid has N nodes, and the adjacent nodes form D branches. M nodes connected to the grounding down conductors are taken as accessible nodes to generate an accessible node set, M≤N. Since the soil resistivity is much greater than the resistivity of the grounding grid, only the pure resistance effect can be considered when diagnosing the corrosion of the grounding grid. Figure 1 The substation grounding network in the system is equivalent to a pure resistance network. The equivalent model of the grounding network is as follows: Figure 3 As shown in the figure, define the network topology and node and branch numbers of the grounding grid. Circles represent grounding grid nodes, rectangles represent grounding grid branches, black solid circles represent accessible nodes of the grounding grid, and black solid rectangles represent branches with corrosion. Figure 4 As shown, the accessible nodes are the points numbered 1, 2, 4, 5, 6, 8, 9, 11, 13, 15, 20, 21, 24, and 25. Corrosion occurs in branches 1 and 36, and the resistance after corrosion is 10 ohms. The other branches are in good condition, and the nominal value of branch resistance R is 1 ohm. In this embodiment, the total number of points N of the grounding grid is 25, the total number of branches D is 40, and the number of accessible nodes M is 14. These 14 accessible nodes are sorted in ascending order of number to form an accessible node set {1, 2, 4, 5, 6, 8, 9, 11, 13, 15, 20, 21, 24, 25}.

[0031] Step 2: Randomly select each accessible node as a measurement point or a non-measurement point, generate a measurement point arrangement sequence, generate an initial solution set consisting of K groups of measurement point arrangement sequences, and initialize the current iteration number q = 1. Measurement point arrangement sequence X k ={x1,...,x m ,...,x M}, if the accessible node m is a measurement point, x m =1; if the accessible node m is a non-measurement point, x m = 0. In this embodiment, K = 20.

[0032] Step 3: Arrange a sequence X for each measurement point in the initial solution set k , k=1,2,...,K, select any measurement point in the measurement point arrangement sequence as the reference node, generate the association matrix A according to the topological structure of the grounding grid, and then derive the branch conductance matrix G from the nominal value of the branch resistance between every two adjacent nodes.

[0033] In generating the association matrix A=[c nd ] N×D In the process, n is the node number, d is the branch number, and the reference direction is from left to right and from top to bottom. The values ​​of each element in A are obtained according to the following value selection method: If branch d flows out of node n according to the reference direction, c nd Take 1. If branch d flows into node n according to the reference direction, c ndTake -1. If branch d is not connected to node n, c nd Take 0. For specific values, please refer to the description in Example 3. Therefore, each branch d of the present invention corresponds to a column of data. Since the nominal value R of the branch resistance in the grounding grid is 1 ohm, the branch conductance matrix G=diag(1 / R)=diag(1) is obtained. The obtained branch conductance matrix G is a D-order square matrix with all main diagonal elements being 1 and the remaining elements being 0.

[0034] Step 4: Obtain the node conductance matrix Y through the association matrix A and the branch conductance matrix G, and define the injection current matrix I of the grounding grid J×N , J is the measurement point arrangement sequence X k The number of measurement points in the middle. Through the branch conductance matrix G and the association matrix A, the node conductance matrix Y=AGA T , and then connect the current source I0 of 1 ampere between the measurement point and the reference node to generate the injection current matrix I J×N . Connect a current source between the reference node and node n, injecting the current matrix I J×N =[a jn ] J×N The value of each element in is as follows: If node n is the measurement point where the current source is connected, a jn =I0. If node n is the reference node, a jn =-I0. If node n is neither a current measurement point nor a reference node, a jn = 0. Wherein j represents the measurement point number, and the measurement point numbers are sorted from small to large. The specific value can refer to the description of the third embodiment.

[0035] Step 5: Through the node conductance matrix Y and the injection current matrix I J×N Calculate the node voltage matrix U N×J , and then calculate the relative voltage matrix U N×J ', according to the branch conductance matrix G, the correlation matrix A, the relative voltage matrix U N×J 'Get the branch current matrix I D×J Node voltage matrix U N×J =I J×N T Y -1 , and then regard the potential of the reference node in the grounding grid as 0, obtain the relative potential of other nodes, and then recalculate the node voltage matrix to obtain the relative voltage matrix U N×J '. Then the grounding grid branch current matrix I is obtained D×J =GA T U N×J ', where I D×J =[b dj ] D×J , b djRepresents the element value in the dth row and jth column of the branch current matrix.

[0036] Step 6: Calculate the branch current matrix I D×J The branch information amount S of the dth row in d , and then calculate the grounding grid information S, establish the fitness function based on the grounding grid information and find the measurement point arrangement sequence X k The fitness value F k . Branch current matrix I D×J =[b dj ] D×J , each branch corresponds to a row of elements in the branch current matrix. The branch information in the dth row is , b dj is the branch current matrix I D×J The value of the element in the dth row and jth column in . Set the information threshold λ and establish the fitness function to find the arrangement sequence X of each measuring point. k The fitness value F k The fitness function is: when the amount of information S of any branch d ≥λ, then F k =J+1 / S; when there is at least one branch with information S d <λ, then F k =∞. Where J is the measurement point arrangement sequence X k is the number of measurement points, λ is the information threshold, λ can take different values ​​according to actual conditions, and in this embodiment, λ is taken as 0.5.

[0037] The fitness values ​​of the initial 20 measurement point arrangement sequence calculated by this method are as follows:

[0038] Measuring point arrangement sequence Fitness value <![CDATA[X1={1,1,1,1,1,1,0,1,1,1,0,1,1,0}]]> ∞ <![CDATA[X2={1,1,1,1,1,0,0,1,0,0,0,1,1,1}]]> ∞ <![CDATA[X3={1,1,0,0,1,0,1,1,1,1,1,0,1,1}]]> ∞ <![CDATA[X4={0,0,0,1,0,1,1,0,0,0,0,0,0,1}]]> ∞ <![CDATA[X5={1,0,0,0,1,0,1,1,0,0,1,1,1,1}]]> ∞ <![CDATA[X6={1,0,1,1,1,1,1,1,1,1,1,1,1,1}]]> 13.01939425 <![CDATA[X7={1,0,1,1,0,0,1,0,1,0,0,0,1,0}]]> ∞ <![CDATA[X8={0,1,1,1,1,0,1,1,0,1,1,1,1,1}]]> 11.02074289 <![CDATA[X9={1,1,0,1,0,0,0,0,1,0,1,0,0,1}]]> ∞ <![CDATA[X 10 ={0,1,0,1,0,0,1,1,0,1,1,0,1,1}]]> ∞ <![CDATA[X 11 ={1,0,1,1,1,1,1,1,1,1,1,0,1,1}]]> ∞ <![CDATA[X 12 ={1,1,1,1,1,0,1,1,1,0,1,1,1,1}]]> 12.02031895 <![CDATA[X 13 ={1,1,0,1,1,1,0,0,0,1,0,0,1,0}]]> ∞ <![CDATA[X 14 ={1,1,1,1,1,1,1,1,1,1,1,1,1,1}]]> 14.01787611 <![CDATA[X 15 ={0,1,1,1,1,1,0,1,1,1,1,1,1,1}]]> 12.02105612 <![CDATA[X 16 ={0,1,1,0,0,1,0,0,1,1,0,0,1,1}]]> ∞ <![CDATA[X 17 ={1,0,1,1,1,0,0,0,1,0,0,1,1,0}]]> ∞ <![CDATA[X 18 ={1,1,1,1,1,1,1,1,1,0,1,1,1,1}]]> 13.01923929 <![CDATA[X 19 ={0,1,1,1,1,1,1,1,1,1,0,1,1,1}]]> 12.02146286 <![CDATA[X 20 ={0,0,0,0,0,1,0,0,0,0,0,0,0,1}]]> ∞

[0039] Branch current matrix I D×J =[b dj ] D×J In the equation, d is the branch number, and j is the measurement point number. The measurement point number can also represent the number of times the current source is connected between the measurement point and the reference node in the grounding grid. Therefore, I D×J The element in row d of [b d1 ,...,b dj ,...,b dJ ] represents the theoretical value of the current response of the dth branch when the current source is connected J times. So by summing the absolute values ​​of the values ​​in this group of elements, we can get , the current response intensity of the dth branch under the external current source can be calculated, which can reflect the importance of the branch for the diagnosis of grounding grid corrosion. , as the branch information S of branch dd .

[0040] Step 7: Select the three groups of measuring point arrangement sequences with the smallest fitness values ​​from the K groups of measuring point arrangement sequences as the alternative arrangement sequences, calculate the distance between each measuring point arrangement sequence and the three groups of alternative arrangement sequences, and then calculate the difference sequence between each measuring point arrangement sequence and the three groups of alternative arrangement sequences. k The three difference sequences of the measuring point arrangement sequence X are calculated k The iterative sequence X k '. Specifically, arrange each measuring point in sequence X k With three sets of alternative arrangement sequences X α , X β , X ξ The moving difference sequence X kα , X kβ , X kξ Find the average value and get the pseudo measurement point arrangement sequence X k "=(X kα +X kβ +X kξ ) / 3, and then arrange the pseudo measurement points into a sequence X k Each element in "is binarized to obtain the iterative measurement point arrangement sequence X k '. The binarization process is: determine the pseudo measurement point arrangement sequence X k "Every element x in m " and 0.5, if x m ">0.5, then the corresponding measurement point arrangement sequence X after iteration k The x in ' m '=1; if x m "≤0.5, then the corresponding measurement point arrangement sequence X after iteration k The x in ' m '=0.

[0041] For example, the three sets of candidate arrangement sequences are sorted from small to large in terms of fitness value, namely X8, X 12 , X 15 , these three sets of alternative arrangement sequences are defined as the first alternative arrangement sequence X α , the second alternative arrangement sequence X β , the third alternative arrangement sequence X ξ . Calculate each measuring point arrangement sequence X in turn k With three alternative arrangement sequences X α , X β , X ξ Distance Z kα , Z kβ , Z kξ , where Z kα =|Ckα X α -X k |、Z kβ =|C kβ X β -X k |、Z kξ =|C kξ X ξ -X k |, C is the swing factor, which affects the update amplitude of the measurement point arrangement sequence in the solution space, C=2p1, p1 is a random number between [0,1]. Then find each measurement point arrangement sequence X k With three groups of alternative measuring point arrangement sequences X α , X β , X ξ The difference sequence X kα , X kβ , X kξ , where X kα =X α -H kα Z kα , X kβ =X β -H kβ Z kβ , X kξ =X ξ -H kξ Z kξ , H is the convergence factor used to control the range of variation of the measurement point arrangement sequence, H=4(p2-1)(1-q / Q), p2 is a random number between [0,1], q is the current iteration number, and Q is the maximum iteration number.

[0042] Step 8: If the current number of iterations q is greater than or equal to the maximum number of iterations Q, proceed to step 9, otherwise q=q+1, and the iterative sequence X k 'Covering measurement point arrangement sequence X k , return to step 3. The maximum number of iterations Q = 100. After the data is covered, X k 'It becomes a new measurement point arrangement sequence, preparing for the next iterative calculation.

[0043] Step 9: Select the measurement point arrangement sequence with the smallest fitness value among the K groups of measurement point arrangement sequences, output the iterative sequence of the measurement point arrangement sequence, and obtain the measurement points for this grounding grid corrosion diagnosis. k 'Select the measurement point arrangement sequence with the smallest fitness value as {1,1,0,1,1,0,0,0,0,0,1,1,1,1}, and compare this measurement point arrangement sequence with the accessible node set {1,2,4,5,6,8,9,11,13,15,20,21,24,25}. If x m=1, then the accessible node m is used as the measurement point; if x m =0, the accessible node m is not used as the measurement point. In this grounding grid corrosion diagnosis, nodes numbered 1, 2, 5, 6, 20, 21, 24, and 25 are selected as measurement points.

[0044] Step 10: According to the iterative sequence of the measurement point arrangement sequence, multiple accessible nodes are selected as measurement points, detection current is injected between the measurement points, the actual value of the branch resistance is calculated, and the corrosion condition of the grounding grid is predicted according to the actual value of the branch resistance. The main purpose of the present invention is to provide a measurement point arrangement optimization method, which does not limit the subsequent technical solution for predicting the corrosion condition of the grounding grid, and can adopt resistance measurement methods such as the three-pole method, the four-pole method, and the clamp grounding resistance tester.

[0045] Embodiment 2

[0046] This embodiment further describes step 2 and step 7 of embodiment 1 in detail.

[0047] In step 2, the initial K groups of measuring point arrangement sequences are randomly generated, and the fitness values ​​of the K groups are obtained through the operations of steps 3, 4, 5, and 6. If the fitness values ​​of less than 3 groups of measuring point arrangement sequences are finite values, the task is terminated and the process is returned to step 2 to generate the measuring point arrangement sequences. This process is continued until the fitness values ​​of at least 3 groups of measuring point arrangement sequences among the initial K groups of measuring point arrangement sequences are finite values.

[0048] If after the qth iteration, the K groups of iterative sequences covering the measurement point arrangement sequence are obtained, and the K groups of fitness values ​​obtained through steps 3, 4, 5, and 6 are less than three groups, return to step 7 of the qth iteration and recalculate the iterative sequence. That is, adjust the convergence factor H and the swing factor C, and recalculate the qth iteration difference sequence X kα , X kβ , X kξ , and then obtain the iterative sequence. After the reorganized iterative sequence returns to the next cycle, there are at least three sets of measurement point arrangement sequences with finite fitness values.

[0049] Embodiment 3

[0050] This embodiment further illustrates the correlation matrix A in step 3 of the first embodiment and the injection current matrix I in step 4 of the first embodiment. J×N The method for getting the value of the element in.

[0051] Reference Figure 4 , the association matrix A=[c nd ] N×D , N=25, D=40, so the incidence matrix consists of 25×40 elements, n is the node number, d is the branch number. 11 The value of observes the relationship between node 1 and branch 1. Figure 4 In the example, since the reference direction is from left to right and from top to bottom, branch 1 flows out of node 1 according to the reference direction. Therefore, according to the value-taking method of step 3 of embodiment 1, c 11 =1; for example, take c 21 , then observe the relationship between node 2 and branch 1. Figure 4 In the example, since the reference direction is from left to right and from top to bottom, branch 1 flows into node 2 according to the reference direction. Therefore, according to the value-taking method of step 3 in embodiment 1, c 21 = -1. If node n is not directly connected to branch d, then c nd = 0. Thus, the values ​​of all elements in the association matrix A can be calculated.

[0052] In the injection current matrix I J×N In the embodiment 1, the first group of measurement point arrangement sequence X1 in the K groups of measurement point arrangement sequences initially randomly generated is taken as an example. X1={1,1,1,1,1,1,0,1,1,1,0,1,1,0}, and the accessible node set is {1,2,4,5,6,8,9,11,13,15,20,21,24,25}, so the measurement points of the measurement point arrangement sequence X1 are selected as nodes numbered 1, 2, 4, 5, 6, 8, 11, 13, 15, 21, 24, so in the injection current matrix I J×N =[a jn ] J×N Where J=11, N=25. The injection current matrix is ​​generated by sequentially connecting a current source with a current magnitude of I0 between the measurement point and the reference node. For example, node 24 is selected as the reference node. In the first row, j=1, j is a serial number sorted in ascending order according to the measurement point number. The data information in the first row indicates that a current source I0 is connected between the first measurement point, i.e., node 1, and the reference node, i.e., node 24. In this row, except for node 1 as the current-connected measurement point and node 24 as the reference node, the remaining nodes are regarded as neither circuit-connected measurement points nor reference nodes. Therefore, according to the value-taking method in step 4 of embodiment 1, the injection current matrix I can be obtained. J×N =[a jn ] J×NThe first row of elements is [I0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, -I0,0], and so on, the elements of each row can be obtained, and then the entire injection current matrix can be obtained. However, in the row corresponding to the reference node, that is, j=J=11, since the 11th measurement point is the reference node at this time, and no current source is injected between the reference nodes, all elements in the 11th row are 0, so in the injection current matrix, the element values ​​of the row corresponding to the reference node are all 0.

[0053] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing the arrangement of grounding grid measurement points based on information quantity, characterized in that: The following steps are involved: Step 1: The grounding grid has N nodes, and adjacent nodes form D branches. M nodes connected to grounding down conductors are taken as accessible nodes to generate an accessible node set, M≤N; Step 2: Randomly select each accessible node as a measurement point or a non-measurement point, generate a measurement point arrangement sequence, generate an initial solution set consisting of K groups of measurement point arrangement sequences, and initialize the current number of iterations q=1; Step 3: Arrange a sequence X for each measurement point in the initial solution set k , k=1,2,...,K, select any measuring point in the measuring point arrangement sequence as the reference node, generate the correlation matrix A according to the topological structure of the grounding grid, and then obtain the branch conductance matrix G according to the nominal value of the branch resistance between every two adjacent nodes; Step 4: Obtain the node conductance matrix Y through the association matrix A and the branch conductance matrix G, and define the injection current matrix I of the grounding grid J×N , J is the measurement point arrangement sequence X k The number of measurement points in the middle, where the node conductance matrix Y = AGA T , connect the current source between the reference node and the measurement point in turn, and determine the injection current matrix I J×N ; Step 5: Through the node conductance matrix Y and the injection current matrix I J×N Calculate the node voltage matrix U N×J , and then calculate the relative voltage matrix U N×J ', according to the branch conductance matrix G, the correlation matrix A, the relative voltage matrix U N×J 'Get the branch current matrix I D×J , where the node voltage matrix U N×J =I J×N T Y -1 , the node voltage matrix U N×J The potential of the reference node is considered to be 0, and the relative voltage matrix U is calculated based on the voltages of other nodes relative to the reference node. N×J '; Step 6: Calculate the branch current matrix I D×J The branch information amount S of the dth row in d , and then calculate the grounding grid information S, establish the fitness function based on the grounding grid information and find the measurement point arrangement sequence X k The fitness value F k , where the branch information amount of the dth row is , b dj is the branch current matrix I D×J The value of the element in the dth row and jth column in the , sets the information threshold λ, if the information amount of any branch S d are greater than or equal to λ, then F k =J+1 / S, if at least one branch has information S d is less than λ, then F k =∞; Step 7: Select the three groups of measuring point arrangement sequences with the smallest fitness values ​​from the K groups of measuring point arrangement sequences as the alternative arrangement sequences, calculate the distance between each measuring point arrangement sequence and the three groups of alternative arrangement sequences, and then calculate the difference sequence between each measuring point arrangement sequence and the three groups of alternative arrangement sequences. k The three difference sequences of the measuring point arrangement sequence X are calculated k The iterative sequence X k '; Step 8: If the current number of iterations q is greater than or equal to the maximum number of iterations Q, proceed to step 9, otherwise q=q+1, and use the iteration sequence X k 'Covering measurement point arrangement sequence X k , return to step 3; Step 9: Select the measuring point arrangement sequence with the smallest fitness value among the K groups of measuring point arrangement sequences, output the iterative sequence of the measuring point arrangement sequence, and obtain the measuring points for this grounding grid corrosion diagnosis.

2. The method for optimizing the arrangement of grounding grid measuring points based on information quantity according to claim 1 is characterized in that: The method further includes step 10 of extracting measurement points in the iterative sequence, injecting detection current between the measurement points, calculating the actual value of the branch resistance, and predicting the corrosion condition of the grounding grid according to the actual value of the branch resistance.

3. The method for optimizing the arrangement of grounding grid measuring points based on information quantity according to claim 1 is characterized in that: In step 2, the measurement point arrangement sequence X k ={x1,...,x m ,...,x M }, if the accessible node m is a measurement point, x m =1; if the accessible node m is a non-measurement point, x m =0.

4. The method for optimizing the arrangement of grounding grid measuring points based on information quantity according to claim 1 is characterized in that: In step 7, three sets of difference sequences X are calculated kα , X kβ , X kξ The mean sequence of pseudo measurement point arrangement sequence X is obtained. k "=(X kα +X kβ +X kξ ) / 3, where α, β, ξ≤K, and the pseudo-measurement point arrangement sequence X k "Each element value in is binarized to generate an iterative sequence X k '.

Images