A Distribution Line Optimization Method and System Based on Multidimensional Index Superposition

Abstract

The present invention relates to a distribution line optimization method and system based on multi-dimensional index superposition, which equivalently analyzes the characteristic quantities of different types of distribution network lines based on a pre-established virtual line case library of the distribution network; determines the structural parameters and technical indicators of the virtual line according to the characteristic quantities; simulates the virtual lines in the case library based on the structural parameters of the virtual line; and calculates the correlation matrix of the technical indicators by using the entropy value method; based on the correlation matrix of the technical indicators, with the goals of optimal comprehensive indicators and minimum investment, establishes a portfolio decision-making optimization model for the distribution network transformation project of the standard virtual line; determines the optimal solution of the distribution line transformation plan by solving the portfolio decision-making optimization model for the distribution network transformation project; thereby maximizing the operation indicators of the distribution network.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution power energy conservation, and particularly relates to a distribution line optimization method and system based on multi-dimensional index superposition. Background Art

[0002] The distribution network is located at the end of the power system and is an important link connecting the transmission and transformation system with users and distributing and supplying electric energy to users. Compared with the transmission network, the distribution network has many points, a wide area, and long lines, with a large number of power equipment. The investment accounts for more than 60% of the total investment in the entire power system. According to statistics, more than 70% of the electric energy loss in the power system occurs in the medium and low voltage distribution network, and 80% of the user power outages are caused by reasons of the distribution network.

[0003] In recent years, although the investment in the construction and transformation of the distribution network has been increased, the huge capital gap in the distribution network is still serious. How to improve the construction effectiveness and investment returns of the distribution network and achieve precise investment in the power grid under limited investment capabilities is the key and difficult problem in the decision-making of the distribution network construction and transformation. Summary of the Invention

[0004] To solve the above problems, the present invention provides a distribution line optimization method and system based on multi-dimensional index superposition, and the purpose thereof is to: under the constraint of limited investment, maximize the operation index of the distribution network. Through the solution of the present invention, the decision-making technology for the implementation of the distribution network project is further improved, and systematic optimization strategy support is provided to achieve "precise investment" in the distribution network.

[0005] The purpose of the present invention is achieved by the following technical solutions:

[0006] A distribution line optimization method based on multi-dimensional index superposition, the method comprising:

[0007] Based on a pre-established virtual line case library of the distribution network, equivalently analyze the characteristic quantities of different types of distribution network lines;

[0008] According to the characteristic quantities, determine the structural parameters and technical indicators of the virtual line;

[0009] Based on the structural parameters of the virtual line, simulate the virtual line in the case library; and use the entropy method to calculate the correlation matrix of the technical indicators;

[0010] Based on the correlation matrix of the technical indicators, with the goal of the best comprehensive index and the minimum investment, establish a combined decision-making optimization model for the distribution network transformation project of the standard virtual line;

[0011] By solving the combined decision-making optimization model for the distribution network transformation project, determine the optimal solution of the distribution line transformation plan.

[0012] Preferably, the characteristic quantities of different types of distribution network lines analyzed equivalently by the virtual line case library of the distribution network established in advance include:

[0013] Randomly extract multiple sample lines from the distribution network lines to be optimized, and perform clustering analysis on all sample lines based on preset indexes to obtain the central values of each index respectively;

[0014] Based on the clustering results, randomly combine any clustering central value of one index with any clustering central value of the other two indexes to obtain the characteristic quantities of the virtual lines in the virtual line case library of the distribution network.

[0015] Further, the preset indexes include the proportion J of high-loss distribution transformers, the average line load rate K, and the average distribution transformer load rate L;

[0016] The central values clustered by the proportion of high-loss distribution transformers are J1, J2... J n , the central values clustered by the average line load rate are K1, K2... K m , and the central values clustered by the average distribution transformer load rate are L1, L2... L w .

[0017] Further, the characteristic quantities of the typical virtual lines in the virtual line case library of the distribution network can be expressed as y = {J n , K m , L w};

[0018] In the formula, J n , K m , L w respectively represent the three typical characteristic quantities of the y-th type of virtual line in the case library.

[0019] Further, determining the structural parameters of the virtual line according to the characteristic quantities includes:

[0020] Classify the selected multiple sample lines approximately according to the characteristic quantities of the phase virtual lines, and store them under different virtual lines according to the sample line types;

[0021] Calculate the structural parameters of the corresponding virtual line according to the structural parameters of the sample lines under each type of virtual line.

[0022] Further, the structural parameters of the virtual line include: the number of lines, the total line length, the total number of distribution transformers, the total capacity of the distribution transformers, and the number of line sectional switches; among them,

[0023] Determine the virtual line length through the following formula:

[0024]

[0025] Determine the number of distribution transformers for the virtual line through the following formula:

[0026]

[0027] Determine the capacity of the distribution transformer for the virtual line through the following formula:

[0028]

[0029] Determine the number of segments of the virtual line through the following formula:

[0030]

[0031] In the formula, y1 represents the length of the y-th type of virtual line, y2 represents the number of distribution transformers for the y-th type of virtual line, y3 represents the capacity of the distribution transformer for the y-th type of virtual line, y4 represents the number of segments of the y-th type of virtual line, y l represents the total length of the 10kV virtual line of the y-th type, y n represents the total number of 10kV virtual lines of the y-th type; y x represents the total number of distribution transformers for the 10kV virtual line of the y-th type; y c represents the total capacity of the distribution transformer for the 10kV virtual line of the y-th type; y z represents the number of sectionalizing switches for the 10kV virtual line of the y-th type.

[0032] Preferably, the determination of the technical indicators of the virtual line according to the characteristic quantity includes:

[0033] Determine the power flow distribution of the distribution line through the forward-backward sweep method;

[0034] Calculate the technical indicators of the virtual line based on the power flow distribution of the distribution line, including the network loss rate, voltage qualification rate, and power factor qualification rate of the entire network.

[0035] Furthermore, the determination of the power flow distribution of the distribution line through the forward-backward sweep method includes:

[0036] a. Calculate the nodal operating power based on the pre-collected initial parameters of the PQ nodes in the entire network;

[0037] b. Determine the power of each branch in the entire network according to the nodal operating power;

[0038] c. Determine the voltage magnitude correction value and phase angle correction value of each node in the entire network according to the power of each branch in the entire network;

[0039] d. Calculate the voltage correction values and reactive power correction values of all PV nodes in the whole network according to the voltage amplitude correction values and phase angle correction values of each node in the whole network, and determine whether the active power and reactive power correction values obtained from the previous and current calculations meet the convergence condition; if not, use the last correction value of each node voltage as the new initial value, and return to step a for iteration until the correction error of the operation power before and after meets the convergence condition; if so, the calculation ends.

[0040] Further, determine the operation power of the node through the following formula:

[0041]

[0042] In the formula, is the operation power of node i, is the load power of node i; is the initial voltage value of all PQ nodes in the whole network; is the initial reactive power injection power of all PV nodes in the whole network; is the admittance of node i to the ground.

[0043] Further, determine the power distribution of each branch in the whole network through the following formula:

[0044]

[0045]

[0046] In the formula, are the branch active power and branch reactive power between node i and node j respectively; are the active component and reactive component of the operation power of node j respectively; are the branch active power loss and branch reactive power loss between node i and node j respectively; R ij , X ij are the branch resistance and branch reactance between node i and node j respectively; C j is the set of all nodes connected to node j except node i; ∑P jk , ∑Q jk are the sum of the active power and the sum of the reactive power of all branches connected to node j except the branch between node i and node j respectively.

[0047] Further, determine the voltage amplitude correction value of each node in the whole network through the following formula:

[0048]

[0049] Determine the electrical phase angle correction value of each node in the whole network through the following formula:

[0050]

[0051] In the formula, is the voltage magnitude correction value of each node in the whole network, is the phase angle correction value of each node in the whole network.

[0052] Furthermore, the voltage correction value of the PV nodes in the whole network is determined by the following formula:

[0053]

[0054] The reactive power correction value of the PV nodes in the whole network is determined by the following formula:

[0055]

[0056] In the formula, is the voltage correction value of the PV nodes in the whole network, is the reactive power correction value of the PV nodes in the whole network.

[0057] Furthermore, the convergence condition is determined by the following formula:

[0058]

[0059]

[0060]

[0061]

[0062] In the formula, ΔP i (1) , is the difference between the previous power correction amount and the next power correction amount of node i; P i (0) , is the correction amount of the active power and reactive power of the previous calculated power of node i; ε1 is the preset convergence error.

[0063] Furthermore, the network loss rate of the whole network is determined by the following formula:

[0064]

[0065] In the formula, ∑ΔP ij is the total active power loss of the whole network, ΔW z is the line power loss, which is equal to the total active power loss ∑ΔP ij of the whole network; W1 is the active power injected at the head end of the line.

[0066] Furthermore, the determination of the voltage qualification rate of the whole network includes:

[0067] Obtain the deviation of each node voltage from the 10 kV standard voltage. If the deviation does not exceed the pre-defined node voltage deviation range, define the current node as a qualified load node; calculate the network voltage qualification rate according to the number of qualified load nodes; where,

[0068] The voltage of each node is determined by the following formula:

[0069]

[0070]

[0071]

[0072] The network voltage qualification rate is determined by the following formula:

[0073]

[0074] In the formula, ΔU ij is the voltage of each node, P ij , Q ij are the active power and reactive power of each branch respectively, R ij , X ij are the branch resistance and branch reactance between node i and node j respectively; U ij is the branch voltage between node i and node j, U i and U j are the voltages of the head and end nodes of the branch respectively.

[0075] Furthermore, the network power factor qualification rate is determined by the following formula:

[0076]

[0077] Preferably, the virtual lines in the structure parameter simulation case library based on virtual lines include:

[0078] According to typical situations, simulate the load distribution on the virtual line to obtain various distribution situations; on the basis of each distribution situation, then simulate the load rate on the virtual line according to the ratio of the wire current-carrying capacity, and establish a technical index value matrix of the standard virtual line;

[0079] Use Digsilent simulation software to simulate the virtual lines in the case library according to the distribution network transformation project connected to the distribution network;

[0080] Among them, the typical situations include end concentration, uniform distribution, gradual increase, gradual decrease, first increase then decrease, and first decrease then increase;

[0081] The distribution network transformation project includes wire capacity increase, transformer substation transformation, transformer substation reactive power compensation, and distributed power sources.

[0082] Furthermore, the technical index value matrix of the standard virtual line is determined by the following formula:

[0083]

[0084] In the formula, represents the technical index value matrix of the y-th type of standard virtual line, (i ∈ M, j ∈ N), where M represents the distribution transformer load rate conditions at different ratios; N represents technical indexes, including the network loss rate of the whole network, the voltage qualification rate, and the power factor qualification rate; (0) represents the situation where the line has not been transformed.

[0085] Furthermore, the entropy value method is adopted to calculate the correlation degree matrix of the technical indexes, including:

[0086] Carry out the distribution network transformation project on the technical index value matrix of the pre-established standard virtual line;

[0087] Carry out data dimensionless processing on the technical index value matrix after transformation;

[0088] Based on the dimensionless processed technical index value matrix, calculate the absolute difference of the technical indexes;

[0089] Based on the absolute difference of the technical indexes, determine the correlation degree coefficient matrix of the technical indexes.

[0090] Furthermore, the dimensionless processed technical index value matrix is determined by the following formula:

[0091]

[0092] In the formula, k represents four types of distribution network transformation projects; represents the technical index value matrix after dimensionless processing; represents the technical index value matrix after the k-th type of distribution network transformation project; represents the technical index value matrix of the line without transformation.

[0093] Furthermore, the absolute difference of the technical indexes is determined by the following formula:

[0094]

[0095] In the formula, is the absolute difference of the technical indexes, indicating the degree of deviation of the technical index value after the distribution network project transformation compared with that before transformation.

[0096] Furthermore, the correlation degree coefficient matrix of the technical indexes is determined by the following formula:

[0097]

[0098]

[0099]

[0100]

[0101] Wherein, V y represents the correlation matrix of multiple projects and multiple indicators on the 10kV standard virtual line of the y-th category, represents the improvement degree of the j-th indicator under the i-th distribution transformer load rate in the k-th distribution network transformation project; represents the information utility value of the improvement degree of the j-th indicator under the k-th category of distribution network transformation project, e j represents the information entropy value of the improvement degree of the j-th indicator under the k-th category of distribution network transformation project; K is a constant,

[0102] Furthermore, based on the correlation matrix of technical indicators, with the goal of optimal comprehensive indicators and minimum investment, the distribution network transformation project portfolio decision optimization model of the standard virtual line under the line case library includes:

[0103] When the cumulative sum of the correlation degrees of the selected project portfolio is maximized, it is defined as the optimal comprehensive indicators;

[0104] With the goal of optimal comprehensive indicators and minimum investment, establish the distribution network transformation project portfolio decision optimization model of the y-th category of standard virtual line under the line case library.

[0105] Furthermore, the distribution network transformation project portfolio decision optimization model is determined by the following formula:

[0106]

[0107]

[0108] Wherein, C is the investment limit; B i is the investment amount of each project; w i is the preset weight of each indicator; z1 is the cumulative sum of the correlation degrees of the project portfolio; z2 is the total investment amount of the project portfolio decision; the correlation degree between the y-th type of typical virtual line and the technical indicator j after the transformation project i; x i is the project decision variable, which only takes 0 or 1. 1 means selecting the i-th project, and 0 means not selecting the i-th project.

[0109] Preferably, the optimal solution of the distribution line transformation scheme is obtained by solving the distribution network transformation project portfolio decision optimization model based on the firefly optimization algorithm.

[0110] A distribution line optimization system based on multi-dimensional index superposition, comprising:

[0111] An analysis module, configured to equivalently analyze characteristic quantities of different types of distribution network lines based on a pre-established virtual line case library of the distribution network;

[0112] A determination module, configured to determine structural parameters and technical indicators of the virtual line according to the characteristic quantities;

[0113] A calculation module, configured to simulate the virtual lines in the case library based on the structural parameters of the virtual line; and calculate the correlation matrix of the technical indicators by using the entropy value method;

[0114] A construction module, configured to establish an optimal combination decision-making optimization model for the distribution network transformation project of the standard virtual line with the goal of the optimal comprehensive index and the minimum investment based on the correlation matrix of the technical indicators;

[0115] An optimization module, configured to determine the optimal solution of the distribution line transformation plan by solving the optimal combination decision-making optimization model of the distribution network transformation project.

[0116] Compared with the closest prior art, the present invention also has the following beneficial effects:

[0117] A distribution line optimization method and system based on multi-dimensional index superposition proposed by the present invention, firstly, equivalently analyze typical characteristic quantities of different types of distribution network lines based on a pre-established typical virtual line case library of the distribution network; determine the structural parameters and technical indicator values of the typical virtual line according to the typical characteristic quantities; through constructing a typical virtual line case library and a calculation method for the line structural parameters of the typical virtual line, conduct a typical analysis on the line, and solve the problems of low operation accuracy and large redundancy caused by diverse line types in practice.

[0118] Secondly, simulate the typical virtual lines in the case library based on the structural parameters of the typical virtual line; and calculate the correlation matrix of the technical indicators by using the entropy value method; by analyzing the influence of typical distribution network projects on the line technical indicators, propose a method for calculating the correlation matrix of various projects and multi-dimensional technical indicators, and solve the problem that it is difficult to quantitatively analyze the complex mutual correlation relationship between different projects and multi-dimensional technical indicators.

[0119] Finally, focusing on the demand for maximizing the investment benefits of distribution network transformation, under a certain investment limit, with the goal of optimal comprehensive technical indicators and minimum investment, by analyzing the correlation between different distribution network projects and multi-dimensional technical indicators, a combined decision-making optimization model for distribution line transformation projects and the final optimization solution are proposed: based on the correlation matrix of technical indicators, with the goal of optimal comprehensive indicators and minimum investment, a combined decision-making optimization model for distribution network transformation projects of typical standard virtual lines under a typical line case library is established; by solving the combined decision-making optimization model for distribution network transformation projects, the optimization results of distribution lines are determined.

[0120] Based on the calculated correlation matrix, a combined decision-making optimization model for line reconstruction projects was established with the goal of achieving the best comprehensive indicators and minimum investment under a certain investment limit. This solved the problem that the current theoretical and technical accuracy of distribution network project implementation decision-making based on investment and construction effectiveness evaluation is low, and there is a lack of systematic optimization strategy support. BRIEF DESCRIPTION OF THE DRAWINGS

[0121] Figure 1 is a flow chart of a distribution line optimization method based on multi-dimensional index superposition provided in an embodiment of the present invention;

[0122] Figure 2 This is a flow chart of a method for making a decision on a power distribution line transformation plan provided in an embodiment of the present invention;

[0123] Figure 3 is a flow chart of a method for extracting typical feature quantities and structural parameters of typical virtual circuits provided in an embodiment of the present invention;

[0124] Figure 4 It is a simulation calculation flow chart of a typical transformation project of a distribution network provided in an embodiment of the present invention;

[0125] Figure 5 It is a flow chart of the firefly algorithm provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0126] The specific implementation modes of the present invention are described in detail below with reference to the accompanying drawings.

[0127] like Figure 1 As shown, in the embodiment of the present invention, a distribution line optimization method based on the superposition of multi-dimensional indicators is applied to maximize the distribution network operation indicators under the constraint of limited investment; the specific steps of the method are as follows:

[0128] S1 equivalently analyzes the characteristic quantities of different types of distribution network lines based on the pre-established distribution network virtual line case library;

[0129] S2 determines the structural parameters and technical indicators of the virtual line according to the characteristic quantity;

[0130] S3 Simulate the virtual lines in the virtual line structure parameter simulation case library based on the virtual lines; and use the entropy method to calculate the correlation matrix of the technical indicators.

[0131] S4 Based on the correlation matrix of the technical indicators, with the goal of the best comprehensive index and the minimum investment, establish an optimization model for the portfolio decision-making of the distribution network transformation project of the standard virtual line.

[0132] S5 By solving the optimization model for the portfolio decision-making of the distribution network transformation project, determine the optimal solution of the distribution line transformation plan.

[0133] In step S1, the typical characteristic quantities for equivalent analysis of different types of distribution network lines based on the pre-established typical virtual line case library of the distribution network include:

[0134] Through the typification analysis of various distribution network line projects, a typical virtual line case library of the distribution network can be established, and the typical virtual lines in the case library are used to equivalently analyze the numerous same-type distribution network lines they represent. Since there are obvious differences in equipment level, load level, and operation level among different distribution network lines, the equipment level, load level, and power grid operation level can be quantitatively evaluated by three index values: the proportion of high-loss distribution transformers, the average load rate of the line, and the average load rate of the distribution transformer. Randomly select several sample lines from the on-site distribution network lines, for example, 100 sample lines; then calculate the three index values of the proportion of high-loss distribution transformers J, the average load rate of the line K, and the average load rate of the distribution transformer L for each sample line, and perform cluster analysis on the three index values of all sample lines to obtain the central values of the three indexes respectively; among them, the central values clustered by the proportion of high-loss distribution transformers are J1, J2... J n , the central values clustered by the average load rate of the line are K1, K2... K m , and the central values clustered by the average load rate of the distribution transformer are L1, L2... L w . Randomly combine any central value of one index with any central value of the other two indexes to obtain the typical characteristic quantities of the typical virtual lines in the typical virtual line case library of the distribution network.

[0135] The typical characteristic quantities of the typical virtual lines in the typical virtual line case library of the distribution network can be expressed as y = {J n , K m , L w};

[0136] In the formula, J n , K m , L w respectively represent the three typical characteristic quantities of the y-th type of typical virtual line in the case library.

[0137] After the typical characteristic quantities of step S1 are determined, the specific structural parameters of the y-th type of typical virtual line need to be obtained. Execute step S2 to approximately classify the previously selected 100 sample lines according to their three index values and the three typical characteristic quantities of the typical virtual lines in the case base, and the 100 sample lines are divided into different typical virtual lines.

[0138] That is, the relevant steps to determine the structural parameters of the typical virtual line according to the typical characteristic quantities in step S2 are as follows:

[0139] Classify the selected several sample lines approximately according to the characteristic quantities of the phase typical virtual line, and store them under different typical virtual lines according to the sample line types;

[0140] Calculate the structural parameters of the corresponding typical virtual line according to the structural parameters of the sample lines under each type of typical virtual line.

[0141] Among them, the structural parameters of the typical virtual line include: the number of lines, the total length of the line, the total number of distribution transformers, the total capacity of the distribution transformers, and the number of line sectional switches; assuming that the structural parameters of each sample line: the number of lines, the total length of the line, the total number of distribution transformers, the total capacity of the distribution transformers, and the number of line sectional switches are known, the structural parameters of the corresponding typical virtual line can be calculated through the structural parameters of the sample lines under each type of typical virtual line. The calculation method of the structural parameters of the typical virtual line is as follows:

[0142] Determine the typical virtual line length through the following formula:

[0143]

[0144] Determine the number of distribution transformers of the typical virtual line through the following formula:

[0145]

[0146] Determine the distribution transformer capacity of the typical virtual line through the following formula:

[0147]

[0148] Determine the number of segments of the typical virtual line through the following formula:

[0149]

[0150] In the formula, y1 represents the length of the y-th type of typical virtual line, y2 represents the number of distribution transformers of the y-th type of typical virtual line, y3 represents the distribution transformer capacity of the y-th type of typical virtual line, y4 represents the number of segments of the y-th type of typical virtual line, y l represents the total length of the 10kV typical virtual line of the y-th type, y n represents the total number of the 10kV typical virtual lines of the y-th type; y xRepresents the total number of distribution transformers of the y - type 10kV typical virtual line; y c Represents the total capacity of distribution transformers of the y - type 10kV typical virtual line; y z Represents the number of sectionalizing switches of the y - type 10kV typical virtual line.

[0151] In addition, step S2 further includes: determining the technical index values of the typical virtual line according to the typical characteristic quantities:

[0152] Determining the power flow distribution of the distribution line by the forward - backward sweep method;

[0153] Calculating the network power loss rate, voltage qualification rate, and power factor qualification rate based on the power flow distribution of the distribution line.

[0154] Among them, determining the power flow distribution of the distribution line by the forward - backward sweep method includes:

[0155] a. Calculating the nodal operating power based on the initial parameters of the whole - network PQ nodes collected in advance;

[0156] b. Determining the power of each branch in the whole network according to the nodal operating power;

[0157] c. Determining the voltage amplitude correction value and phase - angle correction value of each node in the whole network according to the power of each branch in the whole network;

[0158] d. Calculating the voltage correction value and reactive - power correction value of the whole - network PV nodes according to the voltage amplitude correction value and phase - angle correction value of each node in the whole network, and judging whether the active - power and reactive - power correction values obtained from the previous and current calculations meet the convergence condition; if not, taking the last correction value of each node voltage as the new initial value, and returning to step a for iteration until the correction error of the operating power before and after meets the convergence condition; if so, the calculation ends.

[0159] In step a, the nodal operating power is determined by the following formula:

[0160]

[0161] In the formula, P L and Q L respectively represent the active power and reactive power of the whole - network PQ nodes, is the load power of node i; is the initial voltage value of the whole - network PQ nodes; P and U represent the active power and voltage amplitude of the whole - network PV nodes, is the initial reactive - power injection power of the whole - network PV nodes; is the admittance to ground of node i; is the operating power of node i.

[0162] In step b, the power distribution of each branch in the whole network is determined by the following formula:

[0163]

[0164]

[0165] In the formula, are the active power and reactive power of the branch between node i and node j respectively; are the active component and reactive component of the calculated power of node j respectively; are the active power loss and reactive power loss of the branch between node i and node j respectively; R ij , X ij are the branch resistance and branch reactance between node i and node j respectively; C j is the set of all nodes connected to node j except node i; ∑P jk , ∑Q jk are the sum of the active power and the sum of the reactive power of all branches connected to node j except the branch between node i and node j respectively.

[0166] In step c, the voltage magnitude correction values of all nodes in the whole network are determined by the following formula:

[0167]

[0168] The phase angle correction values of all nodes in the whole network are determined by the following formula:

[0169]

[0170] In the formula, is the voltage magnitude correction value of all nodes in the whole network, is the phase angle correction value of all nodes in the whole network.

[0171] In step d, the voltage correction value of the PV nodes in the whole network is determined by the following formula:

[0172]

[0173] The reactive power correction value of the PV nodes in the whole network is determined by the following formula:

[0174]

[0175] In the formula, is the voltage correction value of the PV nodes in the whole network, is the reactive power correction value of the PV nodes in the whole network.

[0176] The convergence condition is determined by the following formula:

[0177]

[0178]

[0179] |ΔP i (1) |<ε1

[0180]

[0181] Wherein, ΔP i (1) , is the difference between the previous power correction amount and the next power correction amount of node i; P i (0) , is the correction amount of the active power and reactive power of the previous calculated power of node i; ε1 is a preset convergence error.

[0182] The total network power loss rate is determined by the following formula:

[0183]

[0184] Wherein, ∑ΔP ij is the total active power loss of the whole network, ΔW z is the line power loss, which is equal in magnitude to the total active power loss ∑ΔP ij of the whole network; W1 is the active power injected at the head end of the line.

[0185] Determining the qualified rate of the whole network voltage includes:

[0186] Obtaining the deviation between the voltage of each node and the 10 kV standard voltage. If the deviation does not exceed the predefined node voltage deviation range, the current node is defined as a qualified load node; calculating the qualified rate of the whole network voltage according to the number of qualified load nodes; wherein, the voltage of each node is determined by the following formula:

[0187]

[0188]

[0189]

[0190] The qualified rate of the whole network voltage is determined by the following formula:

[0191]

[0192] Wherein, P ij , Q ij are the active power and reactive power of each branch respectively, U i and U j are the voltages of the head and end nodes of the branch respectively.

[0193] Determine the qualified rate of the overall network power factor through the following formula:

[0194]

[0195] Step S3 is based on the structural parameters of the typical virtual line. The typical virtual lines in the case library include:

[0196] According to the typical scenarios, simulate the load distribution on the typical virtual line to obtain various distribution scenarios; on the basis of each distribution scenario, then simulate the load rate on the virtual line according to the proportion of the wire current-carrying capacity, and establish a technical index value matrix of the standard virtual line;

[0197] Use Digsilent simulation software to simulate the typical virtual lines in the case library according to the distribution network renovation project accessing the typical distribution network;

[0198] Among them, the typical scenarios include end concentration, uniform distribution, gradual increase, gradual decrease, first increase then decrease, and first decrease then increase;

[0199] The distribution network renovation project includes wire capacity increase, transformer substation transformation, reactive power compensation of the transformer substation, and distributed power sources.

[0200] Determine the technical index value matrix of the standard virtual line through the following formula:

[0201]

[0202] In the formula, represents the technical index value matrix of the y-th type of standard virtual line, (i ∈ M, j ∈ N), where M represents different proportions of transformer load rates; N represents technical indicators, including the overall network power loss rate, voltage qualification rate, and power factor qualification rate; when i is assigned 1, 2, 3, 4, they respectively represent four transformer load rate situations of 20%, 40%, 60%, and 80%; when j is assigned 1, 2, 3, 4, they respectively represent three technical indicators of the power loss rate, voltage qualification rate, and power factor qualification rate; (0) represents the situation where the line has not been renovated.

[0203] In addition, step S3 also includes calculating the correlation matrix of the technical indicators by using the entropy method:

[0204] Carry out the distribution network renovation project transformation on the pre-established technical index value matrix of the standard virtual line;

[0205] Carry out data dimensionless processing on the technical index value matrix after the transformation;

[0206] Based on the dimensionless processed technical index value matrix, calculate the absolute difference of the technical indicators;

[0207] Determine the correlation coefficient matrix of technical indicators based on the absolute difference of the said technical indicators.

[0208] Among them, the matrix of technical indicator values after dimensionless processing is determined by the following formula:

[0209]

[0210] In the formula, k represents four types of distribution network renovation projects; represents the matrix of technical indicator values after dimensionless processing; represents the matrix of technical indicator values after the k-th type of distribution network renovation project; represents the matrix of technical indicator values of the unrenovated line.

[0211] Determine the absolute difference of technical indicators by the following formula:

[0212]

[0213] In the formula, is the absolute difference of technical indicators, indicating the degree of deviation of the technical indicator values after the distribution network project renovation compared with those before renovation.

[0214] Determine the correlation coefficient matrix of technical indicators by the following formula:

[0215]

[0216]

[0217]

[0218]

[0219] In the formula, V y represents the correlation matrix of multiple projects and multiple indicators on the y-th type of 10kV standard virtual line, represents the degree of improvement of the j-th indicator under the i-th type of distribution transformer load rate under the k-th type of distribution network renovation project; represents the information utility value of the degree of improvement of the j-th indicator under the k-th type of distribution network renovation project, e j represents the information entropy value of the degree of improvement of the j-th indicator under the k-th type of distribution network renovation project; K is a constant,

[0220] Step S4 is based on the correlation matrix of technical indicators, aiming at the comprehensive optimization of indicators and the minimum investment, and establishing an optimization model for the combined decision-making of the distribution network renovation project of the typical standard virtual line in the typical line case library, including:

[0221] When the cumulative sum of the relevance of the selected project portfolio is maximized, it is defined as the optimal comprehensive index;

[0222] Taking the optimal comprehensive index and the minimum investment as the objectives, an optimization model for the decision-making of the distribution network transformation project portfolio of the y-th type of standard virtual line under the typical line case library is established as follows:

[0223]

[0224]

[0225] In the formula, C is the investment limit; B i is the investment amount of each project; w i is the preset weight of each index; z1 is the cumulative sum of the relevance of the project portfolio; z2 is the total investment amount of the project portfolio decision-making; The relevance between the i-th transformation project of the y-th type of typical virtual line and the technical index j; x i is the project decision variable, which only takes 0 or 1. 1 means the i-th project is selected, and 0 means the i-th project is not selected.

[0226] In step S5, the optimization result of the distribution line is obtained by solving the optimization model for the decision-making of the distribution network transformation project portfolio based on the firefly optimization algorithm.

[0227] As Figure 2 shown, an embodiment corresponding to the solution of the present invention is provided:

[0228] (1) Construction of the distribution network typical line case library and extraction of typical characteristic quantities:

[0229] In order to typically analyze various different distribution network line projects, a distribution network typical virtual line case library can be established, and the typical virtual lines in the case library are used to equivalently analyze the numerous same-type distribution network lines they represent. Because there are obvious differences in equipment level, load level, and operation level among different distribution network lines, the equipment level, load level, and power grid operation level can be quantitatively evaluated by three index values of the proportion of high-loss distribution transformers, line average load rate, and distribution transformer average load rate respectively.

[0230] As Figure 3 shown, 100 sample lines are randomly selected from the distribution network lines in the field, and then the three index values of the proportion of high-loss distribution transformers J, line average load rate K, and distribution transformer average load rate L of each sample line are calculated. The three index values of all sample lines are used with clustering analysis technology to obtain several central values of the three indexes respectively. Among them, the central values clustered by the proportion of high-loss distribution transformers are distributed as J1, J2... J n and the central values clustered by the line average load rate are K1, K2... K m, the central values clustered by the average load rate of distribution transformers are L1, L2... L w . Selecting any central value of one index and randomly combining it with any central value of the other two indices can be expressed as y = {J n , K m , L w}, and the J n , K m , L w in the formula are used as the three typical characteristic quantities of the y-th typical virtual line in the case library.

[0231] After determining the typical characteristic quantities, it is necessary to obtain the specific structural parameters of the y-th typical virtual line. The previously selected 100 sample lines are approximately classified according to their three index values and the three typical characteristic quantities of the typical virtual lines in the case library, and the 100 sample lines are divided into different typical virtual lines.

[0232] Assume that the structural parameters of each sample line: the number of line segments, the total length of the line, the total number of distribution transformers, the total capacity of the distribution transformers, and the number of line sectional switches are known. The structural parameters of the corresponding typical virtual line can be calculated through the structural parameters of the sample lines under each type of typical virtual line. The calculation method of the structural parameters of the typical virtual line is as follows:

[0233] 1) The length of the y-th typical virtual line:

[0234]

[0235] 2) The number of distribution transformers of the y-th typical virtual line:

[0236]

[0237] 3) The capacity of the distribution transformers of the y-th typical virtual line:

[0238]

[0239] 4) The number of segments of the y-th typical virtual line:

[0240]

[0241] Through the above formula, the structural parameters of the y-th 10kV typical virtual line can be calculated.

[0242] (2) Calculation of technical index values

[0243] The present invention proposes to use three technical indicators, namely, the network loss rate, the voltage qualification rate, and the power factor qualification rate, to analyze the transformation effect. The following are the calculation steps for each index value:

[0244] 1) First, calculate the power flow distribution on the distribution line through the forward-backward substitution method

[0245] ① Initialization: Specify the magnitude and phase angle of the voltage at the balanced node (i.e., the power source node); specify the active power P of all PQ nodes in the network L and the reactive power Q L , as well as the initial voltage values of all PQ nodes in the network (Generally, the magnitude is set to the rated voltage and the phase angle is set to 0); specify the active power P and voltage magnitude U of all PV nodes in the network, and the initial reactive power injection of all PV nodes in the network is

[0246] ② Calculate the operational power of each node:

[0247]

[0248] Where: is the load power of node i; is the admittance to ground of node i; is the operational power of node i.

[0249] ③ Starting from the end nodes of the network, gradually push forward towards the head node to calculate the power distribution of each branch in the whole network:

[0250]

[0251]

[0252] Where: is the active power and reactive power of the branch between node i and node j; is the active component and reactive component of the operational power of node j; is the active power loss and reactive power loss of the branch between node i and node j; R ij , X ij is the resistance and reactance of the branch between node i and node j; C j is the set of all nodes connected to node j except i; ∑P jk , ∑Q jk is the sum of the active power and the sum of the reactive power of all branches connected to node j except the branch between node i and node j.

[0253] ④ Starting from the head node, push back to the end nodes section by section, and calculate the voltage magnitude correction value and phase angle correction value of each node in the whole network according to the power of each branch in the whole network obtained in the previous step. The calculation formulas are as follows:

[0254]

[0255]

[0256] ⑤ Use the voltage magnitude correction values of each node in the previous step and the phase angle correction values to calculate the voltage correction values and the reactive power correction values

[0257]

[0258]

[0259] ⑥ Check whether convergence has been achieved:

[0260]

[0261]

[0262] |ΔP i (1) | < ε1 (5 - 16)

[0263]

[0264] where: ΔP i (1) , is the difference between the previous power correction amount and the next power correction amount of node i; P i (0) , are the active power and reactive power correction amounts of the previous calculated power of node i; ε1 is the set convergence error.

[0265] ⑦ If the convergence condition is not met, use the last correction amount of each node voltage as the new initial value and start the next iteration from the second step until the correction error of the calculated power before and after meets the convergence condition.

[0266] 2) Calculation method of line loss rate

[0267] After obtaining the power flow distribution on the line through the previous calculation, the total active power loss ∑ΔP of the entire network can be calculated ij , then the calculation formula for the line loss rate is as follows:

[0268]

[0269] where: ΔW z is the line power loss, which is equal in magnitude to the total active power loss ∑ΔP of the entire network ij ; W1 is the active power injected at the head of the line.

[0270] 3) Calculation method of voltage qualification rate

[0271] The national standard stipulates that the allowable voltage deviation of a 10 kV line is between (-7% and +7%), that is, within the range of (-0.93 pu to 1.07 pu). Through power flow calculation, the active power P of each branch can be obtained ij and the reactive power Q ij , and further calculate the node voltages of each node in the whole network, where U i and U j are the voltages of the head and end nodes of the branch, and the formula is as follows:

[0272]

[0273]

[0274]

[0275] By calculating the deviation between the node voltages of each node in the whole network obtained and the 10 kV standard voltage, it can be judged whether the voltage deviation of each node is between (-7% and +7%), and whether the node voltage is qualified. The formula for calculating the voltage qualification rate of the whole network is as follows:

[0276]

[0277] (4) Calculation method of power factor qualification rate

[0278] The national standard stipulates that the power factor of the load node above 0.9 is qualified. Therefore, the formula for calculating the power factor qualification rate is as follows:

[0279]

[0280] (3) Simulation environment setting

[0281] In the first step, the line structure parameters of the typical virtual line are known. The load distribution on the typical virtual line is simulated according to five typical situations: concentrated at the end, evenly distributed, increasing gradually, decreasing gradually, increasing first and then decreasing, and decreasing first and then increasing. And on the basis of each distribution situation, the load rates on the line are simulated at 20%, 40%, 60%, and 80% of the wire current-carrying capacity respectively.

[0282] Using Digsilent simulation software, simulate the typical virtual lines in the case library according to the line structure parameters and the above load configuration method.

[0283] The present invention selects four typical distribution network transformation projects: wire capacity increase, distribution transformer transformation in the substation area, reactive power compensation in the substation area, and distributed power source access. As Figure 4 shown, the simulation parameter simulation method is as follows:

[0284] Project 1: Line conductor capacity increase. Mainly by replacing the conductor type to increase the line current-carrying capacity and reduce the line load rate. At the same time, the impedance parameters of the line will also change accordingly. Therefore, in the simulation, modify the corresponding maximum current-carrying capacity and impedance parameters of the replaced conductor type to simulate the line capacity increase transformation project.

[0285] Project 2: Distribution transformer renovation in the substation area. The renovation of the substation area mainly improves the load rate of the substation area by increasing the transformer capacity. At the same time, the replaced new energy-saving transformer also reduces the transformer loss. Therefore, in the simulation, the rated capacity and impedance parameters of the transformer can be modified to simulate the distribution transformer renovation project in the substation area.

[0286] Project 3: Reactive power compensation in the substation area. Reactive power compensation mainly stabilizes the voltage on the load side and reduces the loss in the substation area by installing reactive power compensation devices on the low-voltage side of the transformer. In the simulation, the active power can be kept unchanged and the reactive power can be adjusted to indirectly change the node power factor to simulate the reactive power compensation project in the substation area.

[0287] Project 4: Distributed power source access. The built-in distributed power source model of Digsilent can be used to access the load nodes of the distribution network, and by configuring a certain capacity S DG to simulate the distributed power source access project.

[0288] (4) Analyze and calculate the index correlation matrix based on the entropy value method

[0289] Use the Digsilent simulation software to simulate the unmodified y-type typical virtual line in the case library, and calculate the three technical index values when the distribution transformer load rates of the line nodes are 20%, 40%, 60%, and 80% respectively according to the calculation methods of the three technical indexes in step (2), and form the technical index value matrix of the y-type standard virtual line In the matrix, the columns represent the three technical indexes, and the rows represent four different load rate situations. 0 represents the unmodified line:

[0290]

[0291] Where: (i = 1, 2, 3, 4; j = 1, 2, 3), where i represents the four distribution transformer load rate situations of 20%, 40%, 60%, and 80% respectively; j represents the three technical indexes of line loss rate, voltage qualification rate, and power factor qualification rate; (0) represents the unmodified line situation.

[0292] Similarly, simulate the four distribution network transformation projects of line capacity increase transformation, distribution transformer renovation in the substation area, reactive power compensation in the substation area, and distributed power source access project on the y-type typical virtual line in turn, and calculate the corresponding technical index value matrix (that is, the technical index value matrix after the line capacity increase transformation) Technical index value matrix after the transformation of the distribution transformer in the substation area Technical index value matrix after the reactive power compensation in the substation area Technical index value matrix after the distributed power source access project )

[0293] 1) Data dimensionless processing

[0294] Using the technical index value matrix of the typical virtual line without transformation as the reference matrix, the technical index value matrix after four types of distribution network transformation projects, namely line transformation, distribution transformer transformation in the substation area, reactive power compensation in the substation area, and distributed power source access project, is After dimensionless processing, it is Its calculation formula is as follows:

[0295]

[0296] In the formula: k represents four types of distribution network transformation projects; represents the technical index value matrix after dimensionless processing; represents the technical index value matrix after the k-th type of distribution network transformation project; represents the technical index value matrix of the line without transformation.

[0297] 2) Absolute difference calculation

[0298] The calculation formula of the absolute difference is as follows:

[0299]

[0300] After the absolute difference calculation of the technical index comparison value, the technical index absolute difference is a dimensionless quantity, which represents the degree of deviation of the technical index value after the distribution network project transformation compared with that before the transformation, that is, the degree of improvement of the transformation project on the index.

[0301] 3) Calculate the correlation coefficient matrix

[0302] ① According to the absolute difference obtained in the previous step Calculate the degree of improvement of the j-th index under the i-th transformer load rate under the k-th type of distribution network transformation project The calculation formula is as follows:

[0303]

[0304] ② Calculate the information entropy value e of the degree of improvement of the j-th index under the k-th type of distribution network transformation project j , and its calculation formula is as follows:

[0305]

[0306] where: K is a constant,

[0307] ③ Calculate the information utility value of the improvement degree of the j-th index under the k-th type of distribution network transformation project The calculation formula is as follows:

[0308]

[0309] It reflects the contribution of the improvement degree of the j-th index to the evaluation of the k-th type of distribution network transformation project, and also reflects the degree of association between the j-th index and the k-th type of distribution network transformation project from the side.

[0310] ④ Combine the information utility values of the improvement degrees of various indicators for each type of distribution network transformation project to form the correlation matrix V of multiple projects and multiple indicators on the y-th type of 10kV standard virtual line y , which is expressed as follows:

[0311]

[0312] (5) Line transformation project portfolio decision optimization method

[0313] 1) Line transformation project portfolio decision optimization model

[0314] The correlation degree between project i and technical indicator j on the y-th type of typical virtual line in the case library That is, the correlation degree of project i to the improvement degree of technical indicator j. Therefore, the correlation coefficient can be used to establish a transformation project portfolio decision optimization model. When the cumulative sum of the correlation degrees of the selected project portfolio reaches the maximum, the comprehensive indicators generated by this project portfolio are optimal. At the same time, considering the economic benefit factor, a distribution network transformation project portfolio decision optimization model for the y-th type of standard virtual line under the typical line case library can be proposed with the goal of the best comprehensive indicators and the minimum investment under a certain investment limit. The following programming problem can be listed:

[0315]

[0316] where C is the investment limit; B i is the investment amount of each project; w i is the preset weight of each indicator; z1 is the cumulative sum of the correlation degrees of the project portfolio; z2 is the total investment amount of the project portfolio decision; the correlation degree between the y-th type of typical virtual line and technical indicator j after being transformed by project i; x iis an engineering decision variable, which can only take 0 or 1. 1 means the i-th project is selected, and 0 means the i-th project is not selected.

[0317] 2) Solve the optimization model

[0318] The firefly optimization algorithm can be used to solve this multi-objective 0-1 programming problem. As Figure 5 shown, the steps are as follows:

[0319] ① Initialize the positions, neighborhood ranges, and related parameters of all fireflies.

[0320] ② Feasibilize the infeasible solutions according to the greedy algorithm mechanism and update the non-dominated solution memory bank.

[0321] ③ Calculate the fitness of the fireflies.

[0322] ④ Update the luciferin value and neighborhood range of the fireflies.

[0323] ⑤ The fireflies update their positions according to the crossover strategy within their neighborhoods and update the dynamic decision domain range. If the maximum number of iterations is satisfied, the result is output and the process ends. Otherwise, return to step 2○.

[0324] Based on the same inventive concept, the present application also proposes a distribution network optimization system based on multi-dimensional index superposition, including:

[0325] An analysis module for equivalently analyzing the characteristic quantities of different types of distribution network lines based on a pre-established virtual line case library of the distribution network;

[0326] A determination module for determining the structural parameters and technical indicators of the virtual line according to the characteristic quantities;

[0327] A calculation module for simulating the virtual lines in the case library based on the structural parameters of the virtual line; and calculating the correlation matrix of the technical indicators by using the entropy method;

[0328] A construction module for establishing an optimal combination decision optimization model for the distribution network transformation project of the standard virtual line with the goals of optimal comprehensive indicators and minimum investment based on the correlation matrix of the technical indicators;

[0329] An optimization module for determining the optimal solution of the distribution line transformation plan by solving the optimal combination decision optimization model of the distribution network transformation project.

[0330] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0331] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of blocks.

[0332] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of blocks.

[0333] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of blocks.

[0334] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A method for optimizing a distribution line based on the superposition of multi-dimensional indicators, characterized in that, The method includes: Based on a pre-established virtual line case library of the distribution network, equivalently analyze the characteristic quantities of different types of distribution network lines; According to the characteristic quantities, determine the structural parameters and technical indicators of the virtual line; Based on the structural parameters of the virtual line, simulate the virtual lines in the case library; and use the entropy method to calculate the correlation matrix of the technical indicators; Based on the correlation matrix of the technical indicators, with the goals of optimal comprehensive indicators and minimum investment, establish an optimal model for portfolio decision-making of the distribution network renovation project of the standard virtual line; By solving the optimal model for portfolio decision-making of the distribution network renovation project, determine the optimal solution of the distribution line renovation plan; The equivalently analyzing the characteristic quantities of different types of distribution network lines based on a pre-established virtual line case library of the distribution network includes: Randomly extract multiple sample lines from the distribution network lines to be optimized, and perform clustering analysis on all sample lines based on pre-set indicators to obtain the central values of each indicator respectively; Based on the clustering results, randomly combine any clustering central value of one indicator with any clustering central value of the other two indicators to obtain the characteristic quantities of the virtual lines in the virtual line case library of the distribution network.

2. The method according to claim 1, characterized in that, The pre-set indicators include the proportion J of high-loss distribution transformers, the line average load rate K, and the distribution transformer average load rate L; The central values clustered from the proportion of high-loss distribution transformers are J1, J2... J n , and the central values clustered from the average line load rate are K1, K2... K m , and the central values clustered from the average distribution transformer load rate are L1, L2... L w .

3. The method according to claim 2, characterized in that, The characteristic quantities of the typical virtual lines in the virtual line case library of the distribution network can be expressed as y = {J n , K m , L w}; where J n , K m , L w respectively represent three typical characteristic quantities of the y-th virtual line in the case base.

4. The method according to claim 1, characterized in that, The determining the structural parameters of the virtual line according to the characteristic quantities includes: Approximately classify the selected multiple sample lines according to the characteristic quantities of the phase virtual line, and store them under different virtual lines according to the sample line types; According to the structural parameters of the sample lines under each type of virtual line, calculate the structural parameters of the corresponding virtual line.

5. The method according to claim 4, characterized in that, The structural parameters of the virtual line include: the number of lines, the total line length, the total number of distribution transformers, the total capacity of distribution transformers, and the number of line sectional switches; where Determine the virtual line length through the following formula: Determine the number of distribution transformers of the virtual line through the following formula: Determine the capacity of distribution transformers of the virtual line through the following formula: Determine the number of sections of the virtual line through the following formula: Wherein, y1 represents the length of the y-th type of virtual line, y2 represents the number of distribution transformers of the y-th type of virtual line, y3 represents the capacity of the distribution transformers of the y-th type of virtual line, y4 represents the number of segments of the y-th type of virtual line, y l represents the total length of the 10kV virtual line of the y-th type, y n represents the total number of 10kV virtual lines of the y-th type; y x represents the total number of distribution transformers of the 10kV virtual line of the y-th type; y c represents the total capacity of the distribution transformers of the 10kV virtual line of the y-th type; y z represents the number of sectional switches of the 10kV virtual line of the y-th type.

6. The method according to claim 1, characterized in that, The determining the technical indicators of the virtual line according to the characteristic quantities includes: Determine the power flow distribution of the distribution line through the forward-backward substitution method; Based on the power flow distribution of the distribution line, calculate the technical indicators of the virtual line, including the network power loss rate, the voltage qualification rate, and the power factor qualification rate.

7. The method according to claim 6, characterized in that, The determining the power flow distribution of the distribution line through the forward-backward substitution method includes: a. Based on the initial parameters of the whole network PQ nodes collected in advance, calculate the node operation power; b. According to the node operation power, determine the power of each branch in the whole network; c. According to the power of each branch in the whole network, determine the voltage amplitude correction value and phase angle correction value of each node in the whole network; d. According to the voltage amplitude correction value and phase angle correction value of each node in the whole network, calculate the voltage correction value and reactive power correction value of the PV nodes in the whole network, and judge whether the active power and reactive power correction values obtained from the previous two calculations meet the convergence condition; if not, take the last correction value of each node voltage as the new initial value, and return to step a for iteration until the correction error of the operation power before and after meets the convergence condition; if so, the calculation ends.

8. The method according to claim 7, characterized in that, Determine the operation power of the node through the following formula: In the formula, is the operating power of node i, is the load power of node i; is the initial voltage value of all PQ nodes in the network; is the initial reactive power injection of all PV nodes in the network; is the admittance of node i to the ground.

9. The method according to claim 7, characterized in that, Determine the power distribution of each branch in the whole network through the following formula: wherein, are respectively the active power and reactive power of the branch between node i and node j; are respectively the active component and reactive component of the calculated power of node j; are respectively the active power loss and reactive power loss of the branch between node i and node j; R ij , X ij are respectively the branch resistance and branch reactance between node i and node j; C j is the set of all nodes connected to node j except node i; ∑P jk , ∑Q jk are respectively the sum of the active power and the sum of the reactive power of all branches connected to node j except the branch between node i and node j.

10. The method according to claim 7, characterized in that, Determine the voltage amplitude correction value of each node in the whole network through the following formula: Determine the electrical phase angle correction values of all nodes in the whole network by the following formula: In the formula, is the voltage amplitude correction value of each node in the whole network, is the phase angle correction value of each node in the whole network.

11. The method according to claim 7, characterized in that, Determine the voltage correction values of PV nodes in the whole network by the following formula: Determine the reactive power correction values of PV nodes in the whole network by the following formula: wherein, is the voltage correction value of all the PV nodes in the network, is the reactive power correction value of all the PV nodes in the network.

12. The method according to claim 7, characterized in that, Determine the convergence condition by the following formula: where, ΔP i (1) , is the difference between the previous power correction amount and the next power correction amount of node i; P i (0) , are the correction amounts of the active power and reactive power of the previous operation power of node i; ε1 is a preset convergence error.

13. The method according to claim 6, characterized in that, Determine the network loss rate of the whole network by the following formula: where ΔW z is the line power loss, which is equal in magnitude to the total active power loss ∑ΔP ij ; W1 is the active power injected at the beginning of the line.

14. The method according to claim 6, characterized in that, Determining the qualified rate of the whole network voltage includes: Obtain the deviation between the voltage of each node and the 10 kV standard voltage. If the deviation does not exceed the predefined node voltage deviation range, define the current node as a qualified load node; Calculate the qualified rate of the whole network voltage according to the number of qualified load nodes; Among them, The voltage of each node is determined by the following formula: The qualified rate of the whole network voltage is determined by the following formula: Where, ΔU ij is the voltage of each node, P ij , Q ij are the active power and reactive power of each branch respectively, R ij , X ij are the branch resistance and branch reactance between node i and node j respectively; U ij is the branch voltage between node i and node j, U i and U j are the voltages of the head and tail nodes of the branch respectively.

15. The method according to claim 6, characterized in that, Determine the power factor qualified rate of the whole network by the following formula:

16. The method according to claim 1, wherein The virtual lines in the structure parameter simulation case library based on virtual lines include: According to typical situations, simulate the load distribution on the virtual line to obtain each distribution situation; On the basis of each distribution situation, then simulate the load rate on the virtual line according to the proportion of the wire current-carrying capacity, and establish a technical index value matrix of the standard virtual line; Use Digsilent simulation software to simulate the virtual lines in the case library according to the distribution network transformation project connected to the distribution network; Among them, the typical situations include end concentration, uniform distribution, gradual increase, gradual decrease, first increase then decrease, and first decrease then increase; The distribution network transformation project includes wire capacity increase, distribution transformer transformation in the substation area, reactive power compensation in the substation area, and distributed power sources.

17. The method according to claim 16, wherein Determine the technical index value matrix of the standard virtual line by the following formula: In the formula, represents the technical index value matrix of the y-th type of standard virtual line, represents the technical index value matrix of the unmodified line, where i ∈ M, j ∈ N, M represents different ratios of distribution transformer load rates; N represents technical indexes, including the whole-network power loss rate, voltage qualification rate, and power factor qualification rate; (0) represents the unmodified situation of the line.

18. The method according to claim 16, wherein The method of using entropy value to calculate the correlation degree matrix of the technical indexes includes: Carry out distribution network transformation project transformation on the pre-established technical index value matrix of the standard virtual line; Carry out data dimensionless processing on the technical index value matrix after transformation; Based on the dimensionless processed technical index value matrix, calculate the absolute difference of the technical indexes; Based on the absolute difference of the technical indexes, determine the correlation degree coefficient matrix of the technical indexes.

19. The method according to claim 18, wherein Determine the dimensionless processed technical index value matrix by the following formula: In the formula, k represents four types of distribution network renovation projects; represents the matrix of technical index values after dimensionless processing; represents the matrix of technical index values after the k-th type of distribution network renovation project; represents the matrix of technical index values of the unrenovated line.

20. The method according to claim 19, wherein Determine the absolute difference of the technical indexes by the following formula: In the formula, is the absolute difference of technical indicators, representing the degree of deviation of the technical indicator value after the transformation of the distribution network project compared with that before the transformation.

21. The method according to claim 20, wherein Determine the correlation degree coefficient matrix of the technical indexes by the following formula: where, V y represents the correlation matrix of multiple projects and multiple indicators on the 10 kV standard virtual line of the y-th category, represents the improvement degree of the j-th indicator under the i-th transformer load rate in the k-th distribution network transformation project; represents the information utility value of the improvement degree of the j-th indicator under the k-th distribution network transformation project, e j represents the information entropy value of the improvement degree of the j-th indicator under the k-th distribution network transformation project; K is a constant, 22. The method according to claim 21, wherein The distribution network transformation project combination decision optimization model of the standard virtual line under the line case library based on the correlation degree matrix of the technical indexes with the goal of the best comprehensive index and the minimum investment includes: When the cumulative sum of the correlation degrees of the selected project combination is maximized, it is defined as the best comprehensive index; With the goal of the best comprehensive index and the minimum investment, establish a distribution network transformation project combination decision optimization model of the y - type standard virtual line under the line case library.

23. The method according to claim 22, wherein Determine the distribution network transformation project combination decision optimization model by the following formula: Where C is the investment limit; B i is the investment amount of each project; w i is the preset weight of each index; z1 is the cumulative sum of the correlation degrees of the project portfolio; z2 is the total decision-making investment amount of the project portfolio; The correlation degree between the transformed project i of the y-th type of typical virtual line and the technical index j; x i is a project decision variable, which only takes 0 or 1. 1 means selecting the i-th project, and 0 means not selecting the i-th project.

24. The method according to claim 1, wherein The optimal solution of the distribution line transformation plan is obtained by solving the distribution network transformation project combination decision optimization model based on the firefly optimization algorithm.

25. A power distribution line optimization system based on multi-dimensional index superposition, characterized in that, Include: An analysis module for equivalently analyzing the characteristic quantities of different types of distribution network lines based on a pre-established distribution network virtual line case library; A determination module for determining the structural parameters and technical indexes of the virtual line according to the characteristic quantities; A calculation module for simulating the virtual lines in the case library based on the structural parameters of the virtual line; And the entropy value method is adopted to calculate the correlation matrix of the technical indicators; A construction module is used to establish an optimal combination decision-making optimization model for the distribution network transformation project of the standard virtual line based on the correlation matrix of the technical indicators with the goals of the best comprehensive indicators and the minimum investment; An optimization module is used to determine the optimal solution of the distribution line transformation plan by solving the optimal combination decision-making optimization model for the distribution network transformation project; The analysis module includes: Randomly extract multiple sample lines from the distribution network lines to be optimized, and perform clustering analysis on all sample lines based on the preset indicators to obtain the central values of each indicator respectively; Based on the clustering results, any clustering central value of one indicator is randomly combined with any clustering central value of the other two indicators to obtain the characteristic quantities of the virtual lines in the virtual line case library of the distribution network.

Images