Electric energy quality comprehensive treatment method based on multifunctional grid-connected inverter

Through the comprehensive management method of the power quality of a multi-function grid-connected inverter, the Louvain community discovery algorithm and sequence relationship analysis method are used to solve the voltage distortion and three-phase imbalance caused by the distributed power supply to the power grid, and the power is realized Optimization of quality and improvement of capacity utilization.

CN119010098APending Publication Date: 2024-11-22NORTHEAST DIANLI UNIVERSITY
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410977321.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The large number of distributed power supplies connected to the power grid leads to an increase in the harmonic content of the grid-side current, causing voltage distortion at the network connection point, and the single-phase distributed power supply connected to the power grid will lead to the worsening of the three-phase voltage imbalance problem.

Method used

The comprehensive power quality management method based on multi-function grid-connected inverter is adopted, and the power grid partition is partitioned through the Louvain community discovery algorithm, the dominant governance node is determined, and the sequence relationship analysis method is established-improve the comprehensive power quality evaluation model of the open-grade method is calculated. Wave, reactive and imbalance coefficients are used to optimize the power quality governance model using genetic algorithms and mixed integer linear programming methods to achieve comprehensive management of power quality.

Benefits of technology

It effectively reduces investment in system power quality management, improves the level of power quality, makes full use of the remaining capacity of multi-function grid-connected inverters, and achieves the optimization of harmonic, reactive and three-phase imbalances, ensuring the economic and reasonable operation of the power grid.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119010098A_ABST
    Figure CN119010098A_ABST
Patent Text Reader

Abstract

An electric energy quality comprehensive treatment method based on a multifunctional grid-connected inverter belongs to the technical field of electric energy quality treatment, and comprises the following steps: partitioning a target power grid by using a Louvain community discovery algorithm, and determining dominant treatment nodes of each region according to a reactive sensitivity maximum principle; establishing an electric energy quality comprehensive evaluation model based on an order relation analysis method and an improved grade pulling method; an electric energy quality comprehensive treatment double-layer optimization model considering different residual capacities of the multifunctional grid-connected inverter is constructed, the upper-layer model takes the minimum new energy grid-connected active power reduction amount as the target, and the lower-layer model takes the optimal electric energy quality after treatment as the target; and solving the double-layer optimization model by using a solving method combining a genetic algorithm and mixed integer linear programming. According to the method, the residual capacity of the MFGCI can be fully utilized to optimize and treat harmonic waves, reactive power and three-phase imbalance, and the electric energy quality level of the system and the capacity utilization rate of the MFGCI are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of power quality management in power systems, and in particular relates to a comprehensive power quality management method based on a multifunctional grid-connected inverter. Background Art

[0002] With the construction of new power systems, a large number of distributed power sources connected to the grid will increase the harmonic content of the grid-side current. At the same time, the harmonic current acting on the line impedance will cause the voltage at the grid connection point to be distorted. In addition, the connection of single-phase distributed power sources to the grid will also cause the three-phase voltage imbalance problem to worsen. Therefore, it is of great practical significance to study how to comprehensively manage the power quality under the new power system. Summary of the invention

[0003] The technical problem to be solved by the present invention is: to provide a comprehensive power quality management method based on a multifunctional grid-connected inverter to solve the current technical problems that a large number of distributed power sources are connected to the power grid, resulting in an increase in the harmonic content of the grid-side current, thereby causing voltage distortion at the grid connection point; the connection of single-phase distributed power sources to the power grid will also cause the three-phase voltage imbalance problem to worsen, etc.

[0004] The comprehensive power quality control method based on the multifunctional grid-connected inverter includes the following steps, and the following steps are performed in sequence:

[0005] Step 1: Based on the reactive sensitivity index, the Louvain community discovery algorithm is used to automatically partition the target power grid;

[0006] Step 2: determine the leading governance node in each region according to the maximum reactive power sensitivity principle, and determine the leading governance node in each region as the node with the maximum average sensitivity;

[0007] Step 3: Establish a comprehensive power quality assessment model based on the sequence relationship analysis method-improved grading method to evaluate the power quality before treatment;

[0008] 1) The dq transformation method based on synchronous coordinate transformation is used to extract each current component to obtain the harmonic component, reactive component, fundamental negative sequence component and fundamental positive sequence component of the load current;

[0009] 2) Based on the harmonic components, reactive components, fundamental negative sequence components and fundamental positive sequence components obtained in step 1), the calculation formulas of various power quality index coefficients are given for a single node, and the harmonic coefficients, reactive coefficients and unbalance coefficients before compensation are obtained accordingly;

[0010] 3) Using harmonic coefficient, reactive coefficient and unbalance coefficient as power quality evaluation indicators, the relative importance of each power quality evaluation indicator is calculated by using the order relationship analysis method according to the obtained harmonic coefficient, reactive coefficient and unbalance coefficient before compensation, and the subjective weight of the power quality evaluation indicator is calculated;

[0011] 4) Based on the obtained harmonic coefficient, reactive coefficient and unbalance coefficient before compensation, the objective weights of various power quality evaluation indicators are calculated using the improved grading method;

[0012] 5) Using the “least squares method”, the combined weights of the power quality assessment indicators are calculated and obtained according to the subjective weight values ​​and objective weight values ​​of each power quality assessment indicator;

[0013] 6) Calculate and obtain comprehensive evaluation results of power quality

[0014] Define the compensation coefficient of each power quality evaluation index, and obtain the harmonic coefficient, reactive coefficient, unbalance coefficient and corresponding compensation coefficient before compensation, obtain the harmonic coefficient, reactive coefficient and unbalance coefficient after compensation and use them as the power quality index after compensation, and then obtain the power quality level after compensation through the combined weights of various power quality indicators obtained in step 5) and the calculation formula of the power quality level after compensation, and use it as the comprehensive evaluation result of the power quality before governance;

[0015] Step 4: construct a two-layer optimization model for comprehensive power quality management considering different residual capacities of multifunctional grid-connected inverters, which includes an upper optimization model and a lower optimization model;

[0016] ①Upper-level optimization model

[0017] The decision variable of the upper optimization model is the remaining capacity of the multifunctional grid-connected inverter after active grid connection. The total capacity of the multifunctional grid-connected inverter is set to S I , the grid-connected active power is P I , the harmonic control capacity is S H , reactive power compensation capacity is S Q , the unbalanced compensation capacity is S l , then the relationship between the parameters is as follows:

[0018]

[0019] The objective function of the upper optimization model is to maximize the grid-connected active power so that the active power of photovoltaics can be connected to the grid as much as possible, that is, to minimize the reduction of photovoltaic active power, as shown in the following formula:

[0020] minf1=P PV,max -P PV (18);

[0021] Where: f1 represents the objective function optimized by the upper layer; P PV,max Represents the actual photovoltaic power generation at the current moment; P PV Represents the optimized photovoltaic grid-connected active power;

[0022] Since the remaining capacity of the inverter is limited, the output power of the multifunctional grid-connected inverter should not exceed the maximum power generated by the photovoltaic power plant, that is, it should meet the upper limit constraint of the inverter capacity, as shown in the following formula:

[0023] 0≤P PV ≤P PV,max (19);

[0024] ② Lower layer optimization model

[0025] The lower optimization model allocates the compensation capacity of the multifunctional grid-connected inverter for harmonics, reactive power and imbalance according to the remaining capacity of the multifunctional grid-connected inverter determined by the upper optimization model. Therefore, the decision variable of the lower optimization model is the compensation capacity for the comprehensive power quality management function. Since the remaining capacity of the multifunctional grid-connected inverter that can be used for power quality management is different under different circumstances, the lower optimization model is divided into the following two cases:

[0026] Case 1: Remaining capacity S Y If formula (20) is satisfied, the remaining capacity of the multifunctional grid-connected inverter is large enough. At this time, the lower model aims to optimize the power quality after treatment, and its objective function is shown in formula (21):

[0027]

[0028] minf2=minM=ω1c′1+ω2c′2+ω3c′3 (21);

[0029] Where: U represents the effective value of voltage, I s1 represents the effective value of the fundamental component of the current, c1 represents the harmonic coefficient before compensation, c3 represents the unbalance coefficient before compensation, c′1 represents the harmonic coefficient after compensation, c′2 represents the reactive power coefficient after compensation, c′3 represents the unbalance coefficient after compensation, ω1, ω2, ω3 represent the weights of the harmonic, reactive power and three-phase unbalance indicators respectively;

[0030] Since the primary task of the multifunctional grid-connected inverter is to realize the active power grid connection of photovoltaic power, the compensation capacity invested by the multifunctional grid-connected inverter is constrained, as shown in formula (22):

[0031]

[0032] In the formula, η represents the capacity margin safety factor of the multifunctional grid-connected inverter, which is 0.95; Sc Represents the compensation capacity of the multifunctional grid-connected inverter; S I Represents the total capacity of the multifunctional grid-connected inverter; S P Represents the active grid-connected capacity of the multifunctional grid-connected inverter;

[0033] Case 2: Remaining capacity S Y If equation (23) is satisfied, the remaining capacity of the multifunctional grid-connected inverter is insufficient. At this time, the lower-level model aims to ensure that all indicators of power quality meet the standards after treatment and that the multifunctional grid-connected inverter has the minimum compensation capacity, so as to avoid the multifunctional grid-connected inverter running at full capacity, which may bring potential safety hazards to the power grid, and achieve the purpose of treating power quality.

[0034]

[0035] In order to selectively compensate the power quality of the grid-connected point, the compensation coefficient that minimizes the compensation capacity of the multifunctional grid-connected inverter is selected as the optimal solution. The compensation capacity is expressed by formula (24):

[0036]

[0037] Where: β1 is the compensation coefficient of harmonic coefficient; β2 is the compensation coefficient of reactive power coefficient; β3 is the compensation coefficient of unbalance coefficient; I sh Represents the effective value of the current harmonic component, I n Represents the effective value of the nth harmonic current, where n≠±1,0 and n is an integer; I sq is the effective value of the reactive component of current; I sfu Represents the effective value of the negative sequence fundamental component of the current;

[0038] Correspondingly, the objective function is shown in formula (25):

[0039]

[0040] While ensuring that the multifunctional grid-connected inverter has the minimum compensation capacity, the single power quality index is constrained, and the constraints of the harmonic coefficient c′1 and the unbalance coefficient c′3 after compensation are set as shown in formula (26), and the constraints of formula (22) need to be satisfied at the same time:

[0041]

[0042] In the formula, c l1 and c l3 They respectively represent the maximum limits allowed for the total harmonic distortion rate of the grid-side current and the three-phase unbalance after the multifunctional grid-connected inverter is put into compensation;

[0043] Step 5. Use genetic algorithm and mixed integer linear programming method to solve the two-level optimization model. Among them, the upper optimization model has fewer decision variables, and the genetic algorithm is used to optimize and solve the upper optimization model; the lower optimization model is a multi-constrained linear model with more decision variables. The lower optimization model is solved by a mixed integer linear programming method using a solver suitable for processing multi-constraint problems.

[0044] In step 1, the Louvain community discovery algorithm is based on the modularity function R mod The maximum value of the target power grid is automatically partitioned, and the modularity function R mod The calculation formula is as follows:

[0045]

[0046] Where z represents the number of nodes in the power grid; e = 1, 2, ..., z, u = 1, 2, ..., z, e ≠ u; A eu represents the edge weight between node e and node u; h represents the sum of the system edge weights; k e represents the sum of the weights of all edges connected to node e; k u represents the sum of the weights of all edges connected to node u; θ e represents the number of the community where node e is located, θ e =θ u When δ(θ e ,θ u )=1, otherwise δ(θ e ,θ u )=0;d eu represents the electrical distance between node e and node u; d op Represents the electrical distance between node o and node p; o=1,2,…,z, p=1,2,…,z, o≠p.

[0047] The calculation formula of the harmonic coefficient c1 in step 2) of step 3 is as follows:

[0048] c1=I sh / I s1 (5);

[0049] Where: I s1 Represents the effective value of the fundamental component of the current;

[0050] The calculation formula of reactive power coefficient c2 is as follows:

[0051] c2=I sq / I s1 (6);

[0052] Where: I sq Represents the effective value of the reactive component of current;

[0053] The calculation formula of the unbalance coefficient c3 is as follows:

[0054] c3=I sfu / I s1 (7);

[0055] Where: I sfu Represents the effective value of the negative sequence fundamental component of current.

[0056] In step 3), the relative importance of each power quality evaluation index is calculated by using the order relationship analysis method, and the specific steps of calculating the subjective weight of the power quality evaluation index are as follows:

[0057] The number of power quality indicators is recorded as m, the number of objects to be evaluated is recorded as n, and according to the order relationship analysis method, the most important or least important indicator among the m indicators is selected and recorded as X1; the most important or least important indicator among the remaining m-1 indicators is selected and marked as X2; and so on, the unique order relationship between the evaluation indicators of power quality is obtained;

[0058] According to the order relationship between the power quality evaluation indicators, the relative importance of the adjacent power quality indicators is calculated and obtained. The calculation formula is shown as follows:

[0059] r i =W i-1 / W i (8);

[0060] Where: r i Represents the adjacent index X i-1 With X i The ratio of importance between them, i = 2, 3, ... m-1, m; W i-1 Represents the weight of the i-1th indicator;

[0061] The relative importance can also be determined by experts based on experience or actual conditions, and then the average of the determined values ​​can be taken;

[0062] The subjective weight calculation of each power quality evaluation index is shown in formulas (9)-(11):

[0063] r i-1 ≥1 / r i (9);

[0064]

[0065] W i-1 =r i W i (11);

[0066] Where: ri-1 Represents the adjacent index X i-2 With X i-1 The importance ratio between m Represents the weight of the mth indicator, W i The weight of the i-th indicator is represented by W = (W1, W2, ..., W m ) represents the subjective weight vector of each power quality evaluation index, and W represents the subjective weight vector of each power quality evaluation index.

[0067] The specific steps of using the improved grading method to calculate the objective weights of each power quality evaluation index in step 3 (4) are as follows:

[0068] First, the historical data of m power quality indicators and n indicators of the objects to be evaluated are standardized to form a standardized matrix B n×m ;

[0069] Secondly, the normalized matrix B n×m Multiply each column of by the subjective weight value calculated by equations (9)-(11) to obtain the matrix B n×m * ;

[0070] Then, let H = (B n×m * ) T B n×m * , calculate the eigenvector D of the matrix H and the corresponding maximum eigenvalue λ max ;

[0071] Finally, the characteristic vector D is standardized according to formula (12) to obtain the objective weights of various power quality evaluation indicators;

[0072]

[0073] Where D i Represents the objective weight of each evaluation indicator, D max,i Represents the maximum value of each row of the feature matrix.

[0074] In step 5) of step 3, the specific steps of using the "least squares method" to calculate and obtain the combined weight of the power quality evaluation index according to the subjective weight value and objective weight value of each power quality evaluation index are as follows:

[0075] Establish the weight matrix as shown in formula (13):

[0076]

[0077] Where: g itrepresents the weight value of the ith indicator calculated using the tth indicator weighting method; l represents the number of weighting methods used; m represents the number of power quality indicators;

[0078] The subjective weight value and the objective weight value are combined to establish a model as shown in formula (14):

[0079]

[0080] Where: F represents the objective function; ω j Represents the combined weight value of the jth indicator; a ij Represents the historical data of the jth indicator of the i-th evaluation object;

[0081] The genetic algorithm is used to optimize and solve the model established by formula (14) to obtain the combined weight value of various power quality evaluation indicators.

[0082] The calculation formula of the power quality level after compensation in step 3 (6) is:

[0083] M=ω1c′1+ω2c′2+ω3c′3 (16);

[0084] in,

[0085] Where c2 is the reactive power coefficient before compensation.

[0086] Through the above design scheme, the present invention can bring the following beneficial effects:

[0087] The present invention proposes a comprehensive power quality management method based on Louvain community discovery algorithm, reactive power sensitivity maximum principle, order relationship analysis method, improved grading method, genetic algorithm and mixed integer linear programming method. The method is characterized in that before the comprehensive power quality management is carried out, the idea of ​​zoning management is introduced, the Louvain community discovery algorithm is used to divide the target power grid into regions, and the leading management nodes of each region are selected based on the reactive power sensitivity maximum principle. At the same time, the order relationship analysis method is used to determine the subjective weight, and the improved grading method is used to determine the objective weight. Then, in order to avoid using only the order relationship classification, the subjective weight is determined by the order relationship classification method. The analysis method leads to too many subjective influencing factors, and only using the improved grading method leads to objective one-sidedness. The "least square method" is used to combine and weight the subjective and objective weight results to achieve the subjective and objective unity of weights; in the process of comprehensive power quality management, a two-level optimization model for comprehensive power quality management that taps the potential of multifunctional grid-connected inverters is established, and the two-level optimization model is solved by combining genetic algorithm and mixed integer linear programming; after comprehensive power quality management, the coefficients of each index, comprehensive power quality evaluation results and compensation capacity used for power quality management output by the two-level optimization model are analyzed and obtained.

[0088] The present invention can make full use of the remaining capacity of the MFGCI to optimize and control harmonics, reactive power and three-phase imbalance, improve the power quality level of the system and the capacity utilization rate of the MFGCI, thereby effectively reducing the investment in power quality control of the system and enabling the power grid to achieve more economical and reasonable operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] The present invention is further described below with reference to the accompanying drawings and specific embodiments:

[0090] Figure 1 A schematic diagram of the basic principle of the comprehensive power quality management method based on the multifunctional grid-connected inverter of the present invention;

[0091] Figure 2 It is a schematic diagram of a current component detection algorithm in a comprehensive power quality management method based on a multifunctional grid-connected inverter of the present invention;

[0092] Figure 3 It is a flow chart for solving a double-layer optimization model in a comprehensive power quality management method based on a multifunctional grid-connected inverter of the present invention;

[0093] Figure 4 The partition result of the IEEE33 node system in the embodiment of the present invention;

[0094] Figure 5 is the harmonic coefficient of each node before and after treatment in the embodiment of the present invention;

[0095] Figure 6 is the reactive power coefficient of each node before and after treatment in the embodiment of the present invention;

[0096] Figure 7 is the three-phase unbalance coefficient of each node before and after treatment in the embodiment of the present invention;

[0097] Figure 8 It is an iterative convergence result diagram in an embodiment of the present invention;

[0098] Fig. 9 The comprehensive evaluation results of power quality of each node under the three schemes in the embodiment of the present invention;

[0099] Fig.10 This is a comparison curve of the compensation capacities used in the three treatment schemes in the embodiments of the present invention. DETAILED DESCRIPTION

[0100] The following uses the attached Figure 1-Figure 10 The present invention is further described with examples.

[0101] The purpose of the present invention is to use the remaining capacity of a multifunctional grid-connected inverter to comprehensively manage the power quality in the power grid, and the specific contents are as follows:

[0102] Step 1: Use Louvain community discovery algorithm to partition the target power grid

[0103] According to the reactive sensitivity index, the Louvain community discovery algorithm is used to partition the network. Compared with other partitioning algorithms, the optimal number of partitions determined by the Louvain community discovery algorithm does not need to be set in advance, but is based on the modularity function R mod The maximum value of is automatically generated and has the advantage of being able to discover hierarchical community structures.

[0104] Reactive sensitivity E in power system eu It can be obtained by the inverse operation of the Jacobian matrix in the power flow calculation process. The calculation formula is shown in formula (1):

[0105]

[0106] Where V e represents the fundamental voltage of node e; Q u Represents the reactive power of node u.

[0107] The relationship between the fundamental voltages of the nodes is shown in equation (2):

[0108]

[0109] Where, ΔV e Represents the change in the fundamental voltage of node e; ΔV u Represents the change in the fundamental voltage of node u; V u Represents the fundamental voltage of node u; α eu Represents fundamental reactive sensitivity.

[0110] Modularity function R mod The calculation formula is shown in formula (3):

[0111]

[0112] Where z represents the number of nodes in the power grid; e = 1, 2, ..., z, u = 1, 2, ..., z, e ≠ u; A eu represents the edge weight between node e and node u; h represents the sum of the system edge weights; k e represents the sum of the weights of all edges connected to node e; k u represents the sum of the weights of all edges connected to node u; θ e represents the number of the community where node e is located. e =θ u When δ(θ e ,θ u )=1, otherwise δ(θe ,θ u )=0;d eu represents the electrical distance between node e and node u; d op Represents the electrical distance between node o and node p; o=1,2,…,z, p=1,2,…,z, o≠p.

[0113] The Louvain community discovery algorithm is used to partition the IEEE 33-node system. According to equations (1), (2) and (3), the optimal partitioning result of the target power grid is calculated as follows: When the number of partitions is 7, the maximum value of the modularity function is 0.755, so the optimal number of partitions is 7. The partitioning result of the IEEE 33-node system is as follows: Figure 4 shown.

[0114] Step 2: Determine the leading governance nodes in each region according to the principle of maximum reactive power sensitivity

[0115] The leading governance node in each region needs to have a strong control over the nodes in the region. The stronger its control over the nodes in the region, the smaller the compensation current can be used to make the power quality indicators in the region meet the standards. The smaller the required compensation capacity is, the greater the capacity that can be used for active grid connection under the premise of ensuring the qualified power quality level. Therefore, the node with the largest average reactive sensitivity to all nodes in each region meets the above requirements, so the leading governance node in each region is determined as the node with the largest average sensitivity. Taking region T as an example, the calculation formula for the average sensitivity is given as shown in formula (4):

[0116]

[0117] In the formula, represents the average sensitivity between node e and other nodes in region T; γ represents the number of all nodes in region T; Ω Z Represents the set of nodes contained in region T; E T,eu Represents the reactive sensitivity of node e to node u in area T.

[0118] According to formula (4), the dominant governance nodes of each region are calculated as shown in Table 1.

[0119] Table 1 Results of leading governance node selection in each region

[0120]

[0121] Step 3: Establish a comprehensive power quality assessment model based on the sequence relationship analysis method-improved grading method Sub-step 1: Current component detection algorithm

[0122] For the multifunctional grid-connected inverter, its reference current consists of two parts: grid-connected active power tracking current and compensation current. The dq transformation method based on synchronous coordinate transformation is used to extract each current component.

[0123] For harmonic and reactive current components, first perform Park transformation on the load current to obtain i d 、i q , and then the DC component of the load current can be obtained through a low-pass filter Since the DC component of the load current is obtained by Parker transformation of the fundamental positive sequence component of the load current, and the AC component is obtained by Parker transformation of the fundamental negative sequence component and the harmonic component of the load current, the fundamental positive sequence component of the load current can be obtained by performing Parker inverse transformation on the filtered DC component. The sum of the fundamental negative sequence component and the harmonic component can be obtained by subtracting the fundamental positive sequence component from the total load current, and the harmonic current component can be obtained by subtracting the fundamental negative sequence component. Directly Performing an inverse Park transformation can yield the corresponding reactive current component.

[0124] For negative-sequence current, the system fundamental frequency is denoted as b. The fundamental negative-sequence component is converted into an AC component with twice the fundamental frequency after Park transformation. The AC component is extracted using a bandpass filter with a center frequency of 2b, and then the fundamental negative-sequence component is obtained by inverse transformation.

[0125] Then, the harmonic components, reactive components and corresponding negative sequence components are extracted, and after certain calculations, the current components to be compensated can be selectively obtained. Figure 2 This is the schematic diagram of the current component detection algorithm.

[0126] Sub-step 2: Calculation method of each power quality index coefficient

[0127] Sub-step 1 of step 3 can obtain harmonic, reactive, negative sequence and fundamental positive sequence components, and the calculation method of each index coefficient of power quality is given for a single node. For harmonic current, according to the total harmonic distortion rate, the calculation formula of the harmonic coefficient c1 is given as shown in formula (5):

[0128] c1=I sh / I s1 (5);

[0129] Where: Represents the effective value of the current harmonic component; I n Represents the effective value of the nth harmonic current; I s1 Represents the effective value of the fundamental component of the current; where n≠±1,0, and n is an integer.

[0130] The calculation formula of reactive power coefficient c2 is given as shown in formula (6):

[0131] c2=I sq / I s1 (6);

[0132] For unbalanced current, the calculation formula of the unbalance coefficient c3 is given by the ratio of the effective value of the negative sequence component of the current to the effective value of the fundamental component as shown in formula (7):

[0133] c3=I sfu / I s1 (7);

[0134] Where: I sfu Represents the effective value of the negative sequence fundamental component of current.

[0135] Sub-step 3: Calculate the subjective weight of power quality assessment indicators using the ordinal relationship analysis method

[0136] Based on the harmonic, reactive power and three-phase unbalance coefficients defined in formulas (5) to (7), the sequence relationship analysis method is used to calculate the subjective weights of the power quality assessment indicators. The calculation steps are as follows:

[0137] The number of power quality indicators is denoted as m, and the number of objects to be evaluated is denoted as n. According to the ordinal relationship analysis method, the most important or least important indicator among the m indicators is selected and denoted as X1; the most important or least important indicator among the remaining m-1 indicators is selected and marked as X2; and so on, a unique ordinal relationship between the various power quality evaluation indicators can be obtained.

[0138] The calculation formula for the relative importance of adjacent power quality indicators is shown in formula (8).

[0139] r i =W i-1 / W i (8);

[0140] Where: r i Represents the adjacent index X i-1 With X i The ratio of importance between them; i = 2, 3, ... m-1, m; W i-1 Represents the weight of the i-1th indicator.

[0141] According to the order relationship between the power quality evaluation indicators, the relative importance of the power quality evaluation indicators can be calculated. The relative importance can be determined by experts based on experience or actual conditions, and then the average value of the determined values ​​is taken.

[0142] The calculation of subjective weights of various power quality evaluation indicators is shown in equations (9)-(11).

[0143] r i-1≥1 / r i (9);

[0144]

[0145] W i-1 =r i W i (11);

[0146] Where: W=(W1,W2,…,W m ) represents the subjective weight vector of each power quality evaluation index.

[0147] The subjective weight of each power quality index calculated by the order relationship analysis method is W i =[0.49790.29290.2092].

[0148] Sub-step 4: Calculate the objective weights of power quality assessment indicators using the improved grading method

[0149] Based on the harmonics, reactive power and three-phase unbalance coefficients defined in formulas (5) to (7), the objective weights of various power quality evaluation indicators are calculated using the improved grading method.

[0150] First, the historical data of m power quality indicators and n indicators of the objects to be evaluated are standardized to obtain the standardized matrix B n×m ; Secondly, standardize the matrix B n×m Multiply each column of by the subjective weight value calculated by equations (9)-(11) to obtain the matrix B n×m * ; Then, let H = (B n×m * ) T B n×m * , calculate the eigenvector D of the matrix H and the corresponding maximum eigenvalue λ max ; Finally, the eigenvector D is standardized according to formula (12) to obtain the objective weights of various power quality evaluation indicators.

[0151]

[0152] The objective weight of each power quality index calculated by the improved grading method is D i =[0.5249 0.3698 0.1053]

[0153] Sub-step 5: Calculate the combined weights of power quality assessment indicators using the “least squares method”

[0154] The subjective and objective combined weighting model is established based on the least squares method. The subjective and objective weight values ​​of each power quality evaluation index are obtained by using the order relationship analysis method and the improved grading method to establish the weight matrix shown in formula (13):

[0155]

[0156] Where: g it Representatives use the g it The weight value of the ith indicator calculated by the indicator weighting method; l represents the number of weighting methods adopted; m represents the number of power quality indicators.

[0157] The subjective and objective weight values ​​are combined to establish the corresponding model as shown in formula (14).

[0158]

[0159] Where: F represents the objective function; ω j Represents the combined weight value of the jth indicator; a ij Represents the raw data of the jth indicator of the i-th evaluation object.

[0160] The genetic algorithm is used to optimize and solve the model established by formula (14) to obtain the combined weight value of various power quality evaluation indicators.

[0161] The combined weight of each power quality index calculated by the least squares method and genetic algorithm is ω i =[0.50130.3699 0.1288].

[0162] Sub-step 6: Calculation of comprehensive power quality assessment results

[0163] In order to selectively compensate the components to be compensated at the grid connection point, the compensation coefficient of each component to be compensated is defined as β i (i=1,2,3) and 0≤β i ≤1.

[0164] The various power quality indicators after compensation satisfy formula (15):

[0165]

[0166] Assuming the power quality level after compensation is M, the calculation formula of M is shown in formula (16).

[0167] M=ω1c′1+ω2c′2+ω3c′3 (16);

[0168] Step 4: Construct a two-layer optimization model for comprehensive power quality management considering different residual capacities of multifunctional grid-connected inverters

[0169] Based on the different residual capacities of the multifunctional grid-connected inverter, a two-layer optimization model is proposed. The specific idea is: first, the minimum grid-connected active power reduction is taken as the upper-level goal; secondly, on the basis of satisfying the upper-level objective function, compensation for harmonics, reactive power and unbalanced current is performed. When the sunshine time and intensity are insufficient, the residual capacity of the multifunctional grid-connected inverter is large. At this time, the optimal power quality after treatment is taken as the lower-level goal; when the sunshine time and intensity are sufficient, the residual capacity of the multifunctional grid-connected inverter is insufficient to support all the power quality indicators to reach the optimal level. At this time, the various power quality indicators after treatment and the minimum compensation capacity of the multifunctional grid-connected inverter are taken as the lower-level optimization goals.

[0170] ①Upper-level optimization model

[0171] The decision variable of the upper optimization model is the remaining capacity of the multifunctional grid-connected inverter after active grid connection. Assume that the total capacity of the multifunctional grid-connected inverter is S I , the grid-connected active power is P I , the harmonic control capacity is S H , reactive power compensation capacity is S Q , the unbalanced compensation capacity is S l , then the satisfied relationship is as shown in formula (17):

[0172]

[0173] The objective function of the upper optimization model is to maximize the grid-connected active power so that the active power of photovoltaics can be connected to the grid as much as possible (i.e., minimize the reduction of photovoltaic active power), as shown in formula (18).

[0174] minf1=P PV,max -P PV (18);

[0175] Where: f1 represents the objective function optimized by the upper layer; P PV,max Represents the actual photovoltaic power generation at the current moment; P PV Represents the optimized photovoltaic grid-connected active power.

[0176] Since the remaining capacity of the inverter is limited, the output power of the multifunctional grid-connected inverter should not exceed the maximum power generated by the photovoltaic power plant, that is, it must meet the upper limit constraint of the inverter capacity, as shown in formula (19).

[0177] 0≤P PV ≤P PV,max (19);

[0178] (② Lower layer optimization model

[0179] The lower optimization model reasonably allocates the compensation capacity of the multifunctional grid-connected inverter for harmonics, reactive power and imbalance according to the remaining capacity of the multifunctional grid-connected inverter determined by the upper optimization model. Therefore, the decision variable of the lower optimization model is the compensation capacity used for the comprehensive power quality management function. Since the remaining capacity of the multifunctional grid-connected inverter that can be used for power quality management is different under different circumstances, the lower optimization model is divided into the following two cases for discussion:

[0180] When the remaining capacity S Y When formula (20) is satisfied, the remaining capacity of the multifunctional grid-connected inverter is large enough. At this time, the lower model aims to optimize the power quality after treatment, and its objective function is shown in formula (21):

[0181]

[0182] minf2=minM=ω1c′1+ω2c′2+ω3c′3 (21);

[0183] Where U represents the effective value of voltage.

[0184] Since the primary task of the multifunctional grid-connected inverter is to realize the active power grid connection of photovoltaic power, the compensation capacity invested by the multifunctional grid-connected inverter is constrained as shown in formula (22):

[0185]

[0186] In the formula, η represents the capacity margin safety factor of the multifunctional grid-connected inverter, which is generally taken as 0.95; S c Represents the compensation capacity of multi-function grid-connected inverter; S I Represents the total capacity of the multifunctional grid-connected inverter; S P Represents the active grid-connected capacity of the multifunctional grid-connected inverter.

[0187] When the remaining capacity S Y When equation (23) is satisfied, the remaining capacity of the multifunctional grid-connected inverter is insufficient. At this time, the lower-level model aims to ensure that all indicators of power quality meet the standards after treatment and that the multifunctional grid-connected inverter has the minimum compensation capacity, so as to avoid the multifunctional grid-connected inverter running at full capacity and bringing safety hazards to the power grid, and at the same time achieve the purpose of treating power quality.

[0188]

[0189] In order to selectively compensate the power quality of the grid-connected point, there may be many different options for the compensation coefficient. The compensation coefficient that minimizes the compensation capacity of the multifunctional grid-connected inverter is selected as the optimal solution. Since the compensation capacity can be expressed by formula (24):

[0190]

[0191] Therefore, we choose formula (25) as the objective function:

[0192]

[0193] While ensuring that the multifunctional grid-connected inverter has the minimum compensation capacity, the single power quality index is constrained. Since the national standard has no clear requirements for the reactive coefficient c′2 of the grid-side current after compensation, and considering that the capacity of the multifunctional grid-connected inverter is limited, if c′2 is constrained, it may lead to optimization failure. Therefore, the constraints on c′1 and c′3 are set as shown in formula (26), and the constraints of formula (22) need to be satisfied at the same time.

[0194]

[0195] In the formula, c l1 and c l3 They respectively represent the maximum limits allowed for the total harmonic distortion rate of the grid-side current and the three-phase imbalance after the multifunctional grid-connected inverter is put into compensation.

[0196] In order to verify the effectiveness of the proposed comprehensive power quality management strategy for tapping the potential of multifunctional grid-connected inverters, the following two schemes are set:

[0197] Solution 1 does not consider the comprehensive power quality management function of the multifunctional grid-connected inverter, but only considers its new energy active grid-connected function;

[0198] Scheme 2 considers the two-layer optimization model of the multifunctional grid-connected inverter when the remaining capacity is sufficient, and analyzes whether the harmonic, reactive power and unbalance indicators are optimal.

[0199] Step 5: Solve the two-level optimization model using a solution method that combines genetic algorithm and mixed integer linear programming

[0200] The upper model has fewer decision variables, and a genetic algorithm is used to optimize and solve the upper model. The chromosomes in the genetic algorithm can better encode the grid-connected active power. The lower model is a multi-constrained linear model with more decision variables. A solver suitable for handling multi-constraint problems is called to solve the lower optimization model using a mixed integer linear programming method. Figure 3 It is the flow chart of solving the two-layer optimization model.

[0201] The harmonics, reactive power and unbalance coefficients of each node before and after treatment are as follows: Figure 5 , Figure 6 and Figure 7As shown in the figure. It can be seen that the harmonic, reactive power and three-phase imbalance problems of each node have been significantly improved after treatment. Before treatment, the overall harmonic coefficient of each node was about 13.03%; the overall reactive power coefficient of each node was about 77.72%; the overall three-phase imbalance coefficient of each node was about 3.02%. After using the remaining capacity of the multifunctional grid-connected inverter to treat the power quality of each node, the harmonic, reactive power and three-phase imbalance coefficients of all nodes decreased significantly, and the overall harmonic coefficient of each node dropped to about 0.6%; the overall reactive power coefficient of each node dropped to about 50.13%; the overall three-phase imbalance coefficient of each node dropped to about 0.2%. The above results show that the strategy of considering the use of the remaining capacity of the multifunctional grid-connected inverter to treat the power quality is feasible, which can effectively treat the harmonic, reactive power and three-phase imbalance problems and improve the power quality.

[0202] Taking node 1 as an example, the iterative convergence results of this model are as follows: Figure 8 As shown by Figure 8 It can be seen that the genetic algorithm used in the present invention can converge to the optimal value at a faster speed.

[0203] In order to verify the power quality management function of the multifunctional grid-connected inverter when the remaining capacity is insufficient, Scheme 3 is proposed to analyze whether the compensation capacity invested is the minimum and whether the harmonic, reactive power and unbalance indicators meet the standards when the remaining capacity is insufficient.

[0204] The comprehensive evaluation result curves and average value comparison results of power quality of each node under the three schemes are shown in Fig. 9 and the first row of Table 2.

[0205] Table 2 Comparison results of three schemes

[0206]

[0207] Depend on Fig. 9 It can be seen that when the MFGCI in Scheme 1 is not put into compensation capacity, the M value curve of the comprehensive evaluation result of the power quality of each node is above that of Schemes 2 and 3, indicating that after the comprehensive power quality management strategy considering the remaining capacity of MFGCI proposed in the present invention is put into compensation, the M value of the comprehensive power quality index of each node becomes smaller, and the power quality of each node is significantly improved compared with that before the compensation is put into use.

[0208] It can be further known from the first row of Table 2 that in Scheme 1, when MFGCI does not invest in compensation capacity, the average value of the M value of the comprehensive evaluation result of power quality of each node is 0.33007; Scheme 2 uses the minimum M value of the comprehensive evaluation result of power quality as the objective function to compensate for harmonics, reactive power and unbalanced components, and the average value of M value is reduced to 0.1887, which is 42.83% better than that of Scheme 1 without compensation; Scheme 3 uses the minimum compensation capacity as the objective function to compensate for harmonics, reactive power and unbalanced components while meeting the standards of various indicators, and the average value of M value is reduced to 0.27445, which is 16.85% better than that of Scheme 1 without compensation.

[0209] The compensation capacity curves of the three schemes are as follows: Fig.10 As shown. Fig. 9 and Fig.10 It can be seen that the power quality management effect of Scheme 2 is better than that of Scheme 3. At the same time, the compensation capacity invested in Scheme 2 is also higher than that of Scheme 3. This is due to the different objective functions corresponding to the two schemes, which correspond to different residual capacities of MFGCI. Scheme 2 aims at the optimal power quality management level, and the coefficients of various evaluation indicators are significantly reduced, and its power quality management effect is more significant; Scheme 3 can effectively suppress harmonics, reactive power and three-phase imbalance problems, so that the coefficients of the nodes exceeding the standard meet the standards, and at the same time meet the requirements of PV grid connection, and the management effect is better.

[0210] In order to verify the economic feasibility of the proposed comprehensive power quality management strategy considering different residual capacities of the multifunctional grid-connected inverter, taking the installation cost of the traditional active power filter as an example (3204 yuan / A), the compensation current provided by the multifunctional grid-connected inverter under Scheme 2 and Scheme 3 and the investment cost required to install traditional power quality management equipment to provide the same amount of compensation current are given respectively. The specific data are shown in the second and third rows of Table 2.

[0211] From the second and third rows of Table 2, it can be seen that the cost of installing a traditional active power filter to manage the power quality so that its various coefficients meet the standards is 40,066.02 yuan, and the cost of achieving the optimal power quality level is 96,683.91 yuan. Therefore, making full use of the remaining capacity of the multifunctional grid-connected inverter to participate in the comprehensive management of power quality can effectively reduce the system power quality management investment and enable the power grid to achieve more economical and reasonable operation.

[0212] The specific embodiments used in the present invention have been used to fully illustrate the contents of the present invention, but are not limited to the present embodiments. Any obvious changes made by those skilled in the art based on the teachings of the present invention are within the scope of protection of the present invention.

Claims

1. Comprehensive power quality management method based on multifunctional grid-connected inverter, Its characteristics are: The process includes the following steps, which are performed in sequence: Step 1: Based on the reactive sensitivity index, the Louvain community discovery algorithm is used to automatically partition the target power grid; Step 2: determine the leading governance node in each region according to the maximum reactive power sensitivity principle, and determine the leading governance node in each region as the node with the maximum average sensitivity; Step 3: Establish a comprehensive power quality assessment model based on the sequence relationship analysis method-improved grading method to evaluate the power quality before treatment; 1) The dq transformation method based on synchronous coordinate transformation is used to extract each current component to obtain the harmonic component, reactive component, fundamental negative sequence component and fundamental positive sequence component of the load current; 2) Based on the harmonic components, reactive components, fundamental negative sequence components and fundamental positive sequence components obtained in step 1), the calculation formulas of various power quality index coefficients are given for a single node, and the harmonic coefficients, reactive coefficients and unbalance coefficients before compensation are obtained accordingly; 3) Using harmonic coefficient, reactive coefficient and unbalance coefficient as power quality evaluation indicators, the relative importance of each power quality evaluation indicator is calculated by using the order relationship analysis method according to the obtained harmonic coefficient, reactive coefficient and unbalance coefficient before compensation, and the subjective weight of the power quality evaluation indicator is calculated; 4) Based on the obtained harmonic coefficient, reactive coefficient and unbalance coefficient before compensation, the objective weights of various power quality evaluation indicators are calculated using the improved grading method; 5) Using the "least squares method", the combined weights of the power quality assessment indicators are calculated and obtained according to the subjective weight values ​​and objective weight values ​​of each power quality assessment indicator; 6) Calculate and obtain comprehensive evaluation results of power quality Define the compensation coefficient of each power quality evaluation index, and obtain the harmonic coefficient, reactive coefficient, unbalance coefficient and corresponding compensation coefficient before compensation, obtain the harmonic coefficient, reactive coefficient and unbalance coefficient after compensation and use them as the power quality index after compensation, and then obtain the power quality level after compensation through the combined weights of various power quality indicators obtained in step 5) and the calculation formula of the power quality level after compensation, and use it as the comprehensive evaluation result of the power quality before governance; Step 4: construct a two-layer optimization model for comprehensive power quality management considering different residual capacities of multifunctional grid-connected inverters, which includes an upper optimization model and a lower optimization model; ①Upper-level optimization model The decision variable of the upper optimization model is the remaining capacity of the multifunctional grid-connected inverter after active grid connection. The total capacity of the multifunctional grid-connected inverter is set to S I , the grid-connected active power is P I , the harmonic control capacity is S H , reactive power compensation capacity is S Q , the unbalanced compensation capacity is S l , then the relationship between the parameters is as follows: S I 2 ≥P I 2 +S H 2 +S Q 2 +S l 2 (17); The objective function of the upper optimization model is to maximize the grid-connected active power so that the active power of photovoltaics can be connected to the grid as much as possible, that is, to minimize the reduction of photovoltaic active power, as shown in the following formula: min f 1 =P PV,max -P PV (18); Where: f 1 Represents the objective function optimized by the upper layer; P PV,max Represents the actual photovoltaic power generation at the current moment; P PV Represents the optimized photovoltaic grid-connected active power; Since the remaining capacity of the inverter is limited, the output power of the multifunctional grid-connected inverter should not exceed the maximum power generated by the photovoltaic power plant, that is, it should meet the upper limit constraint of the inverter capacity, as shown in the following formula: 0≤P PV ≤P PV,max (19); ② Lower layer optimization model The lower optimization model allocates the compensation capacity of the multifunctional grid-connected inverter for harmonics, reactive power and imbalance according to the remaining capacity of the multifunctional grid-connected inverter determined by the upper optimization model. Therefore, the decision variable of the lower optimization model is the compensation capacity for the comprehensive power quality management function. Since the remaining capacity of the multifunctional grid-connected inverter that can be used for power quality management is different under different circumstances, the lower optimization model is divided into the following two cases: Case 1: Remaining capacity S Y If formula (20) is satisfied, the remaining capacity of the multifunctional grid-connected inverter is large enough. At this time, the lower model aims to optimize the power quality after treatment, and its objective function is shown in formula (21): min f 2 =minM=ω 1 c′ 1 +oh 2 c′ 2 +oh 3 c′ 3 (21); Where: U represents the effective value of voltage, I s1 represents the effective value of the fundamental component of the current, c 1 represents the harmonic coefficient before compensation, c 3 Indicates the unbalance coefficient before compensation, c′ 1 Indicates the harmonic coefficient after compensation, c′ 2 Represents the reactive power coefficient after compensation, c′ 3 Represents the unbalance coefficient after compensation, ω 1 ,ω 2 ,ω 3 Respectively represent the weights of harmonic, reactive power and three-phase unbalance indicators; Since the primary task of the multifunctional grid-connected inverter is to realize the active power grid connection of photovoltaic power, the compensation capacity invested by the multifunctional grid-connected inverter is constrained, as shown in formula (22): In the formula, η represents the capacity margin safety factor of the multifunctional grid-connected inverter, which is 0.95; S c Represents the compensation capacity of the multifunctional grid-connected inverter; S I Represents the total capacity of the multifunctional grid-connected inverter; S P Represents the active grid-connected capacity of the multifunctional grid-connected inverter; Case 2: Remaining capacity S Y If equation (23) is satisfied, the remaining capacity of the multifunctional grid-connected inverter is insufficient. At this time, the lower-level model aims to ensure that all indicators of power quality meet the standards after treatment and that the multifunctional grid-connected inverter has the minimum compensation capacity, so as to avoid the multifunctional grid-connected inverter running at full capacity, which may bring potential safety hazards to the power grid, and achieve the purpose of treating power quality. In order to selectively compensate the power quality of the grid-connected point, the compensation coefficient that minimizes the compensation capacity of the multifunctional grid-connected inverter is selected as the optimal solution. The compensation capacity is expressed by formula (24): Where: β 1 is the compensation coefficient of the harmonic coefficient; β 2 is the compensation coefficient of reactive power coefficient; β 3 is the compensation coefficient of the unbalanced coefficient; I sh Represents the effective value of the current harmonic component, I n Represents the effective value of the nth harmonic current, where n≠±1,0 and n is an integer; I sq is the effective value of the reactive component of current; I sfu Represents the effective value of the negative sequence fundamental component of the current; Correspondingly, the objective function is shown in formula (25): While ensuring the minimum compensation capacity of the multifunctional grid-connected inverter, the single power quality index is constrained and the harmonic coefficient c after compensation is 1 ′ and unbalance coefficient c 3 ′Set the constraint conditions as shown in formula (26), and at the same time, the constraint conditions of formula (22) must be satisfied: In the formula, c l1 and c l3 They respectively represent the maximum limits allowed for the total harmonic distortion rate of the grid-side current and the three-phase unbalance after the multifunctional grid-connected inverter is put into compensation; Step 5. Use genetic algorithm and mixed integer linear programming method to solve the two-level optimization model. The upper optimization model has fewer decision variables, and the genetic algorithm is used to optimize and solve the upper optimization model; the lower optimization model is a multi-constrained linear model with more decision variables. A solver suitable for handling multi-constraint problems is used to solve the lower optimization model using mixed integer linear programming method.

2. According to claim 1, the comprehensive power quality control method based on the multifunctional grid-connected inverter, Its characteristics are: In step 1, the Louvain community discovery algorithm is based on the modularity function R mod The maximum value of the target power grid is automatically partitioned, and the modularity function R mod The calculation formula is as follows: Where z represents the number of nodes in the power grid; e = 1, 2, ..., z, u = 1, 2, ..., z, e ≠ u; A eu represents the edge weight between node e and node u; h represents the sum of the system edge weights; k e represents the sum of the weights of all edges connected to node e; k u Represents the sum of all edge weights connected to node u; θ e represents the number of the community where node e is located, θ e =θ u When δ(θ e ,θ u )=1, otherwise δ(θ e ,θ u )=0;d eu represents the electrical distance between node e and node u; d op Represents the electrical distance between node o and node p; o=1,2,…,z, p=1,2,…,z, o≠p.

3. According to claim 1, the comprehensive power quality control method based on the multifunctional grid-connected inverter, Its characteristics are: In step 2) of step 3, the harmonic coefficient c 1 The calculation formula is as follows: c 1 =I sh / I s1 (5); Where: I s1 Represents the effective value of the fundamental component of the current; Reactive power coefficient c 2 The calculation formula is as follows: c 2 =I sq / I s1 (6); Where: I sq Represents the effective value of the reactive component of current; Unbalance coefficient c 3 The calculation formula is as follows: c 3 =I sfu / I s1 (7); Where: I sfu Represents the effective value of the negative sequence fundamental component of current.

4. The comprehensive power quality management method based on the multifunctional grid-connected inverter according to claim 1, Its characteristics are: In step 3), the relative importance of each power quality evaluation index is calculated by using the order relationship analysis method, and the specific steps of calculating the subjective weight of the power quality evaluation index are as follows: The number of power quality indicators is recorded as m, the number of objects to be evaluated is recorded as n, and according to the order relationship analysis method, the most important or least important indicator among the m indicators is selected and recorded as X. 1 ; Select the most important or least important indicator from the remaining m-1 indicators and mark it as X 2 ; By analogy, the unique order relationship between the various evaluation indicators of power quality is obtained; According to the order relationship between the power quality evaluation indicators, the relative importance of the adjacent power quality indicators is calculated and obtained. The calculation formula is shown as follows: r i =W i-1 / W i (8); Where: r i Represents the adjacent index X i-1 With X i The ratio of importance between them, i = 2, 3, ... m-1, m; W i-1 Represents the weight of the i-1th indicator; The relative importance can also be determined by experts based on experience or actual conditions, and then the average of the determined values ​​can be taken; The subjective weight calculation of each power quality evaluation index is shown in formulas (9)-(11): r i-1 ≥1 / r i (9); W i-1 =r i W i (11); Where: r i-1 Represents the adjacent index X i-2 With X i-1 The importance ratio between m Represents the weight of the mth indicator, W i Represents the weight of the ith indicator, W = (W 1 ,W 2 ,…,W m ) represents the subjective weight vector of each power quality evaluation index, and W represents the subjective weight vector of each power quality evaluation index.

5. The comprehensive power quality control method based on the multifunctional grid-connected inverter according to claim 1, Its characteristics are: The specific steps of using the improved grading method to calculate the objective weights of each power quality evaluation index in step 3 (4) are as follows: First, the historical data of m power quality indicators and n indicators of the objects to be evaluated are standardized to form a standardized matrix B n×m ; Secondly, the normalized matrix B n×m Multiply each column of by the subjective weight value calculated by equations (9)-(11) to obtain the matrix B n×m * ; Then, let H = (B n×m * ) T B n×m * , calculate the eigenvector D of the matrix H and the corresponding maximum eigenvalue λ max ; Finally, the characteristic vector D is standardized according to formula (12) to obtain the objective weights of various power quality evaluation indicators; Where D i Represents the objective weight of each evaluation indicator, D max,i Represents the maximum value of each row of the feature matrix.

6. The comprehensive power quality control method based on the multifunctional grid-connected inverter according to claim 1, Its characteristics are: In step 5) of the third step, the "least square method" is used to calculate and obtain the combined weight of the power quality evaluation index according to the subjective weight value and objective weight value of each power quality evaluation index. The specific steps are: Establish the weight matrix as shown in formula (13): Where: g it represents the weight value of the ith indicator calculated using the tth indicator weighting method; l represents the number of weighting methods used; m represents the number of power quality indicators; The subjective weight value and the objective weight value are combined to establish a model as shown in formula (14): Where: F represents the objective function; ω j Represents the combined weight value of the jth indicator; a ij Represents the historical data of the jth indicator of the i-th evaluation object; The genetic algorithm is used to optimize and solve the model established by formula (14) to obtain the combined weight value of various power quality evaluation indicators.

7. The comprehensive power quality control method based on the multifunctional grid-connected inverter according to claim 1, Its characteristics are: The calculation formula of the power quality level after compensation in step 3 (6) is: M=ω 1 c′ 1 +oh 2 c′ 2 +oh 3 c′ 3 (16); in, In the formula, c 2 is the reactive power coefficient before compensation.

Citation Information

Cited By

  • Harmonic analysis method and control device for multi-infeed receiving end power grid system

    CN119275941A

  • Comprehensive evaluation method and system for power quality management capability of power distribution network

    CN120598442A