Method and system for determining power quality of power distribution network
By combining multiple analytical methods to calculate the weight and correlation of each quality parameter, the problem of difficulty in quantifying the overall power quality of the distribution network is solved, and the quantification and level reflection of the power quality of the distribution network is realized.
Patent Information
- Application Number
- CN202411040026.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-19
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing power quality analysis methods are difficult to quantify the overall power quality of the distribution network, and fail to effectively consider the impact of new source-load access on power quality, resulting in difficult control of power quality disturbances such as voltage fluctuations and frequency instability.
By obtaining the basic data and quality parameter sets of each node, the weight and correlation degree of each quality parameter are calculated by using network analysis ANP method, correlation analysis GRA method, entropy weight method, game equilibrium method and intuitive fuzzy multi-attribute decision-making VIKOR method, and combined with The overall impact level can realize the quantification of the power quality of the distribution network.
The overall power quality of the distribution network is quantified, reflects the power quality level, balances the weight gap, takes into account the correlation and data fusion between each quality parameter, and avoids one-sidedness of the results.
Smart Images

Figure CN118983813A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electric power grid, and in particular to a method and system for determining power quality of a distribution network. Background Art
[0002] In recent years, with the gradual industrialization of renewable energy and power electronic equipment, the penetration rate of new energy power generation equipment represented by photovoltaic power generation, wind power generation, and flexible source loads represented by electric vehicles and energy storage in the distribution system has increased year by year. However, the high proportion of new source loads connected has had a more severe impact on the stable operation and power quality of the distribution network. On the power supply side, new energy power generation is easily affected by the external natural environment such as wind, light, and temperature. Its output characteristics are highly intermittent, highly volatile, and highly random, which leads to power quality disturbances such as voltage fluctuations, frequency instability, and phase imbalance in the distribution system. On the load side, electric vehicles and energy storage devices no longer have the same regularity of electricity consumption as traditional loads, and it is difficult to use load forecasting and other means to regulate energy. Problems such as energy consumption and power balance in the distribution system are further highlighted. Therefore, it is necessary to analyze the overall power quality of the distribution network to ensure the safe and stable operation of the distribution system.
[0003] The power quality of the distribution system is a combination of multiple factors and multiple indicators. It is difficult to judge the overall power quality level of the distribution system from a certain parameter. Due to the characteristics of the distribution system, such as a large number of nodes, numerous load types, diverse topological structures, and complex spatiotemporal evolution, most of the existing power quality analysis is based on the measured power quality data of each node of the distribution network. The analysis of each node through diversified methods can only reflect the local power quality level of each node, but it is difficult to quantify the overall power quality of the distribution network. At the same time, the existing power quality analysis does not consider the impact of new sources and loads on power quality after they are connected to the distribution network. Summary of the invention
[0004] The purpose of the present invention is to provide a method and system for determining the power quality of a distribution network. The present invention takes into account the characteristics of the distribution network, such as a large number of nodes, numerous load types, diverse topological structures, and complex spatiotemporal evolution, and can quantify the power quality of the distribution network.
[0005] The present invention provides a method for determining power quality of a distribution network, which comprises:
[0006] Obtain basic data of each node in the distribution network;
[0007] Obtaining the influence parameter of each node based on the basic data of each node; obtaining the overall influence of each node based on the influence parameter of each node; the influence parameters include improved node degree, short circuit ratio, electrical betweenness and self-healing degree;
[0008] Acquire monitoring data of a quality parameter set of each node; the quality parameter set includes voltage deviation quality parameters, three-phase unbalance quality parameters, voltage sag and interruption quality parameters, voltage fluctuation and flicker quality parameters, harmonic parameters and frequency deviation quality parameters;
[0009] Based on the monitoring data of each quality parameter of each node, a first weight of each quality parameter is obtained:
[0010] Obtaining an initial first weight of each of the quality parameters based on a network analysis ANP method: Obtaining a stable weighted supermatrix based on the ANP method, wherein a row vector of the stable weighted supermatrix is an initial first weight assigned to each of the quality parameters;
[0011] Based on the GRA method, the initial first weights of the quality parameters are modified to obtain the first weights of the quality parameters:
[0012] Selecting the maximum value of each column in the stable weighted supermatrix and arranging them in sequence to obtain a reference sequence;
[0013] Subtract each row in the stable weighted supermatrix from the reference sequence and take the absolute value to obtain n sequence difference values; n is the total number of nodes in the distribution network;
[0014] Based on the n sequence differences, n correlation coefficient sequences are obtained; the expression is:
[0015]
[0016] Where: i represents the i-th correlation coefficient sequence, i = [1, 2, ..., n], the i-th correlation coefficient sequence is the i-th correlation coefficient sequence of each quality parameter, ΔQ is the difference of n sequences, min is the minimum value, max is the maximum value, ΔQ i is the ith sequence difference, ρ is the resolution coefficient, ρ∈(0,1);
[0017] Based on the n correlation coefficient sequences, the correlation degree sequence is obtained; the expression is:
[0018]
[0019] Where: E represents the correlation sequence, which is the correlation degree of each quality parameter;
[0020] Based on the association degree sequence, the first weight of each quality parameter is obtained; the expression is:
[0021]
[0022] Where: is the first weight of the bth quality parameter, B is the total number of quality parameters in the power quality classification model, G b is the correlation degree of the bth quality parameter;
[0023] Based on the monitoring data of each quality parameter of each node, an entropy weight method is used to obtain a second weight of each quality parameter;
[0024] Based on the first weight and the second weight of each of the quality parameters, a game equilibrium method is used to obtain the overall weight of each of the quality parameters;
[0025] Based on the overall weight of each quality parameter and combined with the monitoring data of each quality parameter of each node, the ratio value of each node is obtained by using the intuitive fuzzy multi-attribute decision-making VIKOR method;
[0026] Based on the ratio value and the overall influence of each node, the power quality parameter value of each node is obtained, and the power quality parameter value of the distribution network is obtained by summing the power quality parameter values of each node.
[0027] Optionally, the improved node degree expression is:
[0028]
[0029] Where: D i1 represents the improved node degree of the i-th node, K ie represents the degree of the i-th node, D avl represents the mean degree of n nodes, I i represents the propagation value of the i-th node, r ij represents the electrical distance between the i-th node and the j-th node;
[0030] The short circuit ratio expression is:
[0031]
[0032] Where: D i2 represents the short-circuit ratio of the i-th node, S ac,i represents the short-circuit capacity of the AC system at the i-th node, S eq,i represents the source-load equivalent grid-connected capacity of the i-th node;
[0033] The electrical betweenness expression is:
[0034]
[0035] Where: D i3represents the electrical betweenness of the ith node, V1 is the set of nodes in the distribution network that are traditional power generation nodes, V2 is the set of nodes in the distribution network that are distributed power generation nodes, V3 is the set of nodes in the distribution network that are load nodes, S x is the output power of the xth traditional power generation node, S y is the output power of the yth distributed power node, S k is the output power of the kth load node, C x,y,k (i) is the transit degree of the i-th node, I x,y,k (i,z) is the current generated between the i-th node and the z-th node after adding a unit current source between the x-th traditional generation node and the y-th distributed generation node, between the x-th traditional generation node and the k-th load node, and between the y-th distributed generation node and the k-th load node;
[0036] The self-healing degree expression is:
[0037]
[0038] Where: D i4 represents the self-healing degree of the i-th node, represents the capacity of the network-type device connected to the i-th node, The capacity of the i-th node connected to the networking device, S b Indicates the average capacity of all connected devices of n nodes.
[0039] Optionally, the overall influence expression is:
[0040]
[0041] Where: D i represents the overall influence of the i-th node, D ir represents the rth influence parameter of the ith node, D jr Represents the rth influence parameter of the jth node.
[0042] Optionally, the obtaining of the initial first weight of each quality parameter based on the ANP method of network analysis includes:
[0043] An ANP network structure is constructed based on the power quality classification model. The ANP network structure includes a control layer and a network layer. The control layer includes B element sets [P1, P2, ..., P b ,…,P B ], P b Indicates the classification of the bth quality parameter. The network layer includes B element sets [C1, C2, …, C b ,…,C B ], q represents the total number of decision makers in the ANP network, q = n, e bq It represents the assignment of the mth quality parameter by the qth decision maker, taking the B element sets in the control layer as the criterion and the B element sets in the network layer as the secondary criterion, and obtaining the important value between the assignments in the B element sets in the network layer;
[0044] constructing a weight matrix between each of the quality parameters based on the importance values between each of the assigned values, and further obtaining an unweighted supermatrix between each of the quality parameters;
[0045] Normalizing the unweighted supermatrix to obtain a weighted supermatrix;
[0046] The weighted supermatrix is stabilized to obtain the stabilized weighted supermatrix.
[0047] Optionally, the obtaining the second weight of each quality parameter based on the monitoring data of each quality parameter of each node by using an entropy weight method includes:
[0048] Normalizing the monitoring data of each quality parameter of each node to obtain n×B normalized data;
[0049] Based on the n×B normalized data, the information entropy of each quality parameter is obtained, and the effective value of the information of each quality parameter is further obtained;
[0050] A second weight of each of the quality parameters is obtained based on the information validity value of each of the quality parameters.
[0051] Optionally, the obtaining the overall weight of each quality parameter by adopting a game equilibrium method based on the first weight and the second weight of each quality parameter includes:
[0052] Construct the initial overall weight expression:
[0053]
[0054] Where: w b represents the overall weight of the b-th quality parameter, represents the second weight of the b-th quality parameter, a and s are linear combination coefficients;
[0055] Expression-based Solve to obtain the values of a and s, substitute the values of a and s into the initial overall weight expression to obtain the overall weight expression, and obtain the overall weight of each quality parameter based on the overall weight expression, where T is the transpose.
[0056] Optionally, the ratio value of each node is obtained by using the intuitive fuzzy multi-attribute decision making VIKOR method based on the overall weight of each quality parameter and the monitoring data of each quality parameter of each node, including:
[0057] The monitoring data of each quality parameter of each node is used to obtain n×B normalized data;
[0058] Based on n×B normalized data and combined with the overall weight of each quality parameter, the group value and individual value of each node are obtained; the expression is:
[0059]
[0060] Where: F i represents the group value of the i-th node, R i represents the individual value of the i-th node, max(Y b ) represents the maximum value of the normalized data of the bth quality parameter, min(Y b ) represents the minimum value of the normalized data of the bth quality parameter, Y ib represents the normalized data of the bth quality parameter of the i-th node;
[0061] Based on the group value and individual value of each node, the ratio value of each node is obtained, and the expression is:
[0062]
[0063] Where: X i represents the ratio value of the i-th node, minF represents the minimum value among the group values of n nodes, maxF represents the maximum value among the group values of n nodes, minR represents the minimum value among the individual values of n nodes, maxR represents the maximum value among the individual values of n nodes, and v is the decision coefficient.
[0064] Optionally, the power quality parameter value of each node is obtained based on the ratio value and the overall influence of each node, and the expression is:
[0065] G i =X i D i ;
[0066] Where: G i is the power quality parameter value of the ith node, D i is the overall influence of the i-th node, X i Represents the ratio value of the i-th node.
[0067] The present invention also provides a system applicable to the above-mentioned method for determining power quality of distribution network, comprising:
[0068] A first data acquisition module is used to acquire basic data of each node in the distribution network;
[0069] An influence module is used to obtain the influence parameter of each node based on the basic data of each node, and obtain the overall influence of each node based on the influence parameter of each node; the influence parameters include improved node degree, short circuit ratio, electrical betweenness and self-healing degree;
[0070] A second data acquisition module is used to acquire monitoring data of a quality parameter set of each node; the quality parameter set includes a voltage deviation quality parameter, a three-phase unbalance quality parameter, a voltage sag and interruption quality parameter, a voltage fluctuation and flicker quality parameter, a harmonic parameter and a frequency deviation quality parameter;
[0071] The first parameter weight module obtains the first weight of each quality parameter based on the monitoring data of each quality parameter of each node:
[0072] A first parameter weight unit is used to obtain an initial first weight of each quality parameter based on a network analysis ANP method: a stable weighted supermatrix is obtained based on the ANP method, and a row vector of the stable weighted supermatrix is an initial first weight assigned to each quality parameter;
[0073] The first weight correction unit is used to correct the initial first weights of the quality parameters based on the GRA method to obtain the first weights of the quality parameters:
[0074] A reference subunit, used for selecting the maximum value of each column in the stable weighted supermatrix and arranging them in sequence to obtain a reference sequence;
[0075] A difference subunit is used to make a difference between each row in the stable weighted supermatrix and the reference sequence, and take the absolute value to obtain n sequence difference values; n is the total number of nodes in the distribution network;
[0076] The correlation coefficient subunit is used to obtain n correlation coefficient sequences based on the n sequence differences; the expression is:
[0077]
[0078] Where: i represents the i-th correlation coefficient sequence, i = [1, 2, ..., n], the i-th correlation coefficient sequence is the i-th correlation coefficient sequence of each quality parameter, ΔQ is the difference of n sequences, min is the minimum value, max is the maximum value, ΔQ i is the ith sequence difference, ρ is the resolution coefficient, ρ∈(0,1);
[0079] The association degree subunit is used to obtain an association degree sequence based on the n association coefficient sequences; the expression is:
[0080]
[0081] Where: E represents the correlation sequence, which is the correlation degree of each quality parameter;
[0082] The first parameter weight subunit is used to obtain the first weight of each quality parameter based on the association degree sequence; the expression is:
[0083]
[0084] Where: is the first weight of the bth quality parameter, B is the total number of quality parameters in the power quality classification model, G b is the correlation degree of the bth quality parameter;
[0085] A second parameter weight module, used for obtaining a second weight of each quality parameter by adopting an entropy weight method based on monitoring data of each quality parameter of each node;
[0086] An overall weight module, used to obtain an overall weight of each quality parameter by using a game equilibrium method based on the first weight and the second weight of each quality parameter;
[0087] A ratio value module, for obtaining the ratio value of each node by using the VIKOR method based on the overall weight of each quality parameter and the monitoring data of each quality parameter of each node;
[0088] The power quality module is used to obtain the power quality parameter value of each node based on the ratio value and overall influence of each node, and sum the power quality parameter values of each node to obtain the power quality parameter value of the distribution network.
[0089] The beneficial technical effects of the present invention are as follows: the method for determining the power quality of the distribution network of the present invention takes into account the node degree, short-circuit ratio, electrical intermediate number and self-healing degree, and can obtain the overall power quality level of the distribution network. The method for determining the power quality of the distribution network of the present invention improves the ANP method based on the GRA method, further ensures the rationality of the first weight, avoids the difference characteristics brought by different decision makers, and adopts a balanced algorithm to obtain the overall weight, balancing the influence of the one-sidedness of the result caused by the large gap between the first weight and the second weight. The method for determining the power quality of the distribution network of the present invention takes into account the correlation between the quality parameters and the influence of data fusion, and uses the VIKOR method to obtain the ratio value of each node, further realizing the quantification of the power quality of the whole network, reflecting the overall power quality level of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS
[0090] Figure 1 is a flow chart of a method for determining power quality of a distribution network according to the present invention;
[0091] Figure 2 It is the IEEE13-node standard distribution network topology diagram of the present invention;
[0092] Figure 3 It is a schematic diagram of the influence parameters of each node and the overall influence of each node in the present invention;
[0093] Figure 4 It is a schematic diagram of the influence parameters of each node and the overall influence of each node after normalization of the present invention;
[0094] Figure 5 It is a diagram of monitoring data of each quality parameter of each node of the present invention;
[0095] Figure 6 is a schematic diagram of the first weight and the second weight of each quality parameter of the present invention;
[0096] Figure 7 It is a schematic diagram of the overall weights of various quality parameters of the present invention;
[0097] Figure 8 It is a schematic diagram of the ratio values of each node of the present invention;
[0098] Fig. 9 It is a schematic diagram of grading each node based on the ratio value of the node in the present invention;
[0099] Fig.10 It is a schematic diagram of the comparison results between the method of the present invention, the traditional TOPSIS method and the Extension Cloud Theory method. DETAILED DESCRIPTION
[0100] Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0101] Figure 1 Flow chart of the method for determining power quality of distribution network of the present invention. Figure 1 As shown, the present invention provides a method for determining power quality of a distribution network, which comprises:
[0102] Obtain basic data of each node in the distribution network.
[0103] The influence parameters of each node are obtained based on the basic data of each node; the overall influence of each node is obtained based on the influence parameters of each node. The influence parameters include improved node degree, short circuit ratio, electrical betweenness and self-healing degree.
[0104] Specifically, based on the local and global characteristics of the distribution network, the traditional node degree is combined with power quality propagation, and the improved node degree expression is:
[0105]
[0106] Where: D i1 represents the improved node degree of the i-th node, K ie represents the degree of the i-th node, D avl represents the mean degree of n nodes, I i represents the propagation value of the i-th node, r ij Represents the electrical distance between the i-th node and the j-th node.
[0107] The short-circuit ratio is used to measure the resistance of each node to power quality disturbances. The short-circuit ratio expression is:
[0108]
[0109] Where: D i2 represents the short-circuit ratio of the i-th node, S ac,i represents the short-circuit capacity of the AC system at the i-th node, S eq,i Represents the source-load equivalent grid-connected capacity of the i-th node.
[0110] Based on the power generation, power consumption and power flow of each node, according to the propagation characteristics of power quality disturbance, and taking into account the node capacity and branch impedance factors, the electrical betweenness expression is:
[0111]
[0112] Where: D i3 represents the electrical betweenness of the ith node, V1 is the set of nodes in the distribution network that are traditional power generation nodes, V2 is the set of nodes in the distribution network that are distributed power generation nodes, V3 is the set of nodes in the distribution network that are load nodes, S x is the output power of the xth traditional power generation node, S y is the output power of the yth distributed power node, S k is the output power of the kth load node, C x,y,k (i) is the transit degree of the i-th node, I x,y,k (i,z) is the current generated between the i-th node and the z-th node after adding a unit current source between the x-th traditional power generation node and the y-th distributed power generation node, between the x-th traditional power generation node and the k-th load node, and between the y-th distributed power generation node and the k-th load node.
[0113] Considering the control mode and output characteristics of each node connected to the source and load, the self-healing degree expression is:
[0114]
[0115] Where: D i4represents the self-healing degree of the i-th node, represents the capacity of the network-type device connected to the i-th node, The capacity of the i-th node connected to the networking device, S b It represents the average capacity of all connected devices of n nodes. i4 The larger it is, the worse the self-healing degree of the i-th node is, and vice versa.
[0116] The overall influence expression is:
[0117]
[0118] Where: D i represents the overall influence of the i-th node, D ir represents the rth influence parameter of the ith node, D jr Represents the rth influence parameter of the jth node.
[0119] Acquire monitoring data of the quality parameter set of each node. The quality parameter set includes voltage deviation quality parameters, three-phase unbalance quality parameters, voltage sag and interruption quality parameters, voltage fluctuation and flicker quality parameters, harmonic parameters and frequency deviation quality parameters.
[0120] Based on the monitoring data of each quality parameter of each node, a first weight of each quality parameter is obtained.
[0121] Preferably, based on the monitoring data of each quality parameter of each node, obtaining the first weight of each quality parameter includes:
[0122] Based on the network analysis ANP method, an initial first weight of each quality parameter is obtained: based on the ANP method, a stable weighted supermatrix is obtained, and a row vector of the stable weighted supermatrix is an initial first weight assigned to each quality parameter.
[0123] Based on the GRA method, each initial first weight of each quality parameter is corrected to obtain the first weight of each quality parameter.
[0124] Specifically, the initial first weight of each quality parameter is obtained based on the ANP method, including:
[0125] Based on the power quality classification model, an ANP network structure is constructed. The ANP network structure includes a control layer and a network layer. The control layer includes B element sets [P1, P2, …, P b ,…,P B ], P b Indicates the classification of the bth quality parameter. The network layer includes B element sets [C1, C2, …, C b ,…,C B ], q represents the total number of decision makers in the ANP network, q = n, e bq It represents the assignment of the mth quality parameter by the qth decision maker, which is obtained by the satty scaling method, with the B element sets in the control layer as the criterion and the B element sets in the network layer as the secondary criterion, and the important value between the assignments in the B element sets in the network layer is obtained.
[0126] A weight matrix between each quality parameter is constructed based on the importance value between each assignment, and an unweighted super matrix between each quality parameter is further obtained.
[0127] The unweighted supermatrix is normalized to obtain the weighted supermatrix.
[0128] The weighted supermatrix is stabilized to obtain a stable weighted supermatrix.
[0129] Furthermore, each initial first weight of each quality parameter is corrected based on the GRA method to obtain the first weight of each quality parameter, including:
[0130] The maximum value of each column in the stable weighted supermatrix is selected and arranged in sequence to obtain a reference sequence.
[0131] Each row in the stable weighted supermatrix is subtracted from the reference sequence and the absolute value is taken to obtain n sequence differences, where n is the total number of nodes in the distribution network.
[0132] Based on n sequence differences, n correlation coefficient sequences are obtained. The expression is:
[0133]
[0134] Where: i represents the i-th correlation coefficient sequence, i = [1, 2, ..., n], the i-th correlation coefficient sequence is the i-th correlation coefficient sequence of each quality parameter, ΔQ is the difference of n sequences, min is the minimum value, max is the maximum value, ΔQ i is the ith sequence difference, ρ is the resolution coefficient, ρ∈(0,1).
[0135] Based on the n correlation coefficient sequences, the correlation degree sequence is obtained. The expression is:
[0136]
[0137] Where: E represents the correlation degree sequence, and the correlation degree sequence is the correlation degree of each quality parameter.
[0138] The first weight of each quality parameter is obtained based on the association degree sequence. The expression is:
[0139]
[0140] Where: is the first weight of the bth quality parameter, B is the total number of quality parameters in the power quality classification model, G b is the correlation degree of the bth quality parameter.
[0141] Based on the monitoring data of each quality parameter of each node, the entropy weight method is used to obtain the second weight of each quality parameter. In the new distribution network, the power quality disturbance has the characteristics of multi-scale spatiotemporal changes. The single first weight cannot reflect the dynamic change information and the degree of disturbance of each quality parameter. Therefore, the present invention is based on the objective attributes of the measured power quality data itself, and uses the entropy weight method to obtain the second weight of each quality parameter.
[0142] Preferably, based on the monitoring data of each quality parameter of each node, an entropy weight method is used to obtain a second weight of each quality parameter, including:
[0143] The monitoring data of each quality parameter of each node is normalized to obtain n×B normalized data.
[0144] The information entropy of each quality parameter is obtained based on n×B normalized data, and the information effective value of each quality parameter is further obtained.
[0145] A second weight of each quality parameter is obtained based on the information validity value of each quality parameter.
[0146] Based on the first weight and the second weight of each quality parameter, the game equilibrium method is used to obtain the overall weight of each quality parameter. In the process of determining the power quality, only considering the first weight ignores the disturbance probability and change characteristics of each quality parameter itself under the operating condition of the distribution network, while only considering the second weight ignores the harmful effects of each quality parameter, both of which lead to a reduction in the accuracy of power quality. In order to fully integrate the respective advantages and characteristics of the two and avoid the defect of lack of accuracy of a single weight, the present invention adopts a game equilibrium algorithm to solve the overall weight, which not only reflects the importance of each quality parameter from the subjective judgment experience, but also reflects the probability information of objective data.
[0147] Specifically, based on the first weight and the second weight of each quality parameter, the overall weight of each quality parameter is obtained by using a game equilibrium method, including:
[0148] Construct the initial overall weight expression:
[0149]
[0150] Where: w b represents the overall weight of the b-th quality parameter, represents the second weight of the b-th quality parameter, and a and s are linear combination coefficients.
[0151] Expression-based Solve to obtain the values of a and s, substitute the values of a and s into the initial overall weight expression to obtain the overall weight expression, and obtain the overall weight of each quality parameter based on the overall weight expression, where T is the transpose.
[0152] Based on the overall weight of each quality parameter and the monitoring data of each quality parameter of each node, the intuitive fuzzy multi-attribute decision-making VIKOR method is used to obtain the ratio value of each node.
[0153] Furthermore, based on the overall weight of each quality parameter and combined with the monitoring data of each quality parameter of each node, the VIKOR method is used to obtain the ratio value of each node, including:
[0154] The monitoring data of each quality parameter of each node is used to obtain n×B normalized data.
[0155] Based on n×B normalized data and combined with the overall weight of each quality parameter, the group value and individual value of each node are obtained. The expression is:
[0156]
[0157] Where: F i represents the group value of the i-th node, R i represents the individual value of the i-th node, max(Y b ) represents the maximum value of the normalized data of the bth quality parameter, min(Y b ) represents the minimum value of the normalized data of the bth quality parameter, Y ib Represents the normalized data of the bth quality parameter of the ith node.
[0158] Based on the group value and individual value of each node, the ratio value of each node is obtained. The expression is:
[0159]
[0160] Where: X i represents the ratio value of the i-th node, minF represents the minimum value among the group values of n nodes, maxF represents the maximum value among the group values of n nodes, minR represents the minimum value among the individual values of n nodes, maxR represents the maximum value among the individual values of n nodes, v is the decision coefficient, v=0.5.
[0161] Based on the ratio value and overall influence of each node, the power quality parameter value of each node is obtained, and the power quality parameter value of each node is summed to obtain the power quality parameter value of the distribution network. The power quality parameter value expression of each node is:
[0162] G i =X iD i ;
[0163] Where: G i is the power quality parameter value of the ith node, D i is the overall influence of the i-th node, X i Represents the ratio value of the i-th node.
[0164] The present invention also provides a system applicable to the above-mentioned method for determining power quality of distribution network, comprising:
[0165] The first data acquisition module is used to acquire basic data of each node in the distribution network.
[0166] The influence module is used to obtain the influence parameters of each node based on the basic data of each node, and obtain the overall influence of each node based on the influence parameters of each node. The influence parameters include improved node degree, short circuit ratio, electrical betweenness and self-healing degree.
[0167] The second data acquisition module is used to acquire monitoring data of the quality parameter set of each node. The quality parameter set includes voltage deviation quality parameter, three-phase unbalance quality parameter, voltage sag and interruption quality parameter, voltage fluctuation and flicker quality parameter, harmonic parameter and frequency deviation quality parameter.
[0168] The first parameter weight module is used to obtain the first weight of each quality parameter based on the monitoring data of each quality parameter of each node.
[0169] The first parameter weight module includes:
[0170] The first parameter weight unit is used to obtain the initial first weight of each quality parameter based on the network analysis ANP method: a stable weighted super matrix is obtained based on the ANP method, and a row vector of the stable weighted super matrix is an initial first weight assigned to each quality parameter.
[0171] The first weight correction unit is used to correct each initial first weight of each quality parameter based on the GRA method to obtain the first weight of each quality parameter.
[0172] The reference subunit is used to select the maximum value of each column in the stable weighted supermatrix and arrange them in sequence to obtain a reference sequence.
[0173] The difference subunit is used to make a difference between each row in the stable weighted supermatrix and the reference sequence, and take the absolute value to obtain n sequence difference values. n is the total number of nodes in the distribution network.
[0174] The correlation coefficient subunit is used to obtain n correlation coefficient sequences based on n sequence differences. The expression is:
[0175]
[0176] Where: i represents the i-th correlation coefficient sequence, i = [1, 2, ..., n], the i-th correlation coefficient sequence is the i-th correlation coefficient sequence of each quality parameter, ΔQ is the difference of n sequences, min is the minimum value, max is the maximum value, ΔQ i is the ith sequence difference, ρ is the resolution coefficient, ρ∈(0,1).
[0177] The correlation subunit is used to obtain a correlation sequence based on n correlation coefficient sequences. The expression is:
[0178]
[0179] Where: E represents the correlation degree sequence, and the correlation degree sequence is the correlation degree of each quality parameter.
[0180] The first parameter weight subunit is used to obtain the first weight of each quality parameter based on the association degree sequence. The expression is:
[0181]
[0182] Where: is the first weight of the bth quality parameter, B is the total number of quality parameters in the power quality classification model, G b is the correlation degree of the bth quality parameter.
[0183] The second parameter weight module is used to obtain the second weight of each quality parameter by adopting the entropy weight method based on the monitoring data of each quality parameter of each node.
[0184] The overall weight module is used to obtain the overall weight of each quality parameter by adopting a game equilibrium method based on the first weight and the second weight of each quality parameter.
[0185] The ratio value module is used to obtain the ratio value of each node based on the overall weight of each quality parameter and the monitoring data of each quality parameter of each node using the VIKOR method.
[0186] The power quality module is used to obtain the power quality parameter value of each node based on the ratio value and overall influence of each node, and sum the power quality parameter values of each node to obtain the power quality parameter value of the distribution network.
[0187] Specifically, taking the IEEE 13-node standard distribution network as an example and carrying out source-load expansion, its topology is as follows: Figure 2 As shown, the actual operating condition of the IEEE 13-node standard distribution network is level III. According to the process of S2 of the present invention, the influence parameters of each node and the overall influence of each node are obtained as follows: Figure 3 The results after standardization are shown in Figure 4 As shown by Figure 4 It can be seen that the overall influence of nodes 1 and 2 is relatively large, mainly because the improved node degree, short-circuit ratio and electrical intermediate weight are relatively large. When the node here has power quality disturbance, many other nodes are affected, resulting in a decrease in the overall power quality level of the distribution network. Therefore, it occupies an obvious important position in the distribution network; while nodes 6, 9, and 11 are all at the end of the distribution network. The impact of the power quality disturbance of the node on the overall power quality level of the distribution network is relatively limited, so the overall influence is relatively low.
[0188] The monitoring equipment is used to monitor and record the quality parameters of each node, and the 95% probability maximum value is selected as the measured data sampling value of the quality parameter. Finally, the monitoring data of each quality parameter of each node can be obtained as follows: Figure 5 shown.
[0189] The first weight and the second weight of each quality parameter are as follows: Figure 6 As shown in Figure 7 The ratio values of each node are shown as Figure 8 As shown, after classification Fig. 9 As shown in the figure, the comparison results with the traditional TO PSIS method and the extension cloud theory method are as follows Fig.10 As shown, in the result of TOPSIS method, there is a certain error between the grading result of the ratio value of node 8 and node 12 and the grading result of the present invention, because TOPSIS method only uses a single Euclidean distance as the evaluation criterion, it is difficult to comprehensively evaluate the relationship between each quality parameter and the positive and negative ideal solutions, and it is easily affected by a single ratio value with a large disturbance amplitude. In the extension cloud theory method, the selection of the membership function is affected by subjective factors, so it is easy to cause errors in the results. The method not invented not only reflects the correlation between the ratio values, but also combines subjective factors with objective factors, reducing the error caused by a single weight to the result.
[0190] Finally, the overall power quality parameter value of the IEEE 13-node standard distribution network is 0.5854. Figure 8 It can be seen that the power quality of the distribution network is at level III, which is consistent with the actual operating conditions.
[0191] The embodiments described above are only descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.
Claims
1. A method for determining power quality of a distribution network, characterized in that: It includes: Obtain basic data of each node in the distribution network; Obtaining the influence parameter of each node based on the basic data of each node; Obtaining the overall influence of each node based on the influence parameters of each node; the influence parameters include improved node degree, short circuit ratio, electrical betweenness and self-healing degree; Acquire monitoring data of a quality parameter set of each node; the quality parameter set includes voltage deviation quality parameters, three-phase unbalance quality parameters, voltage sag and interruption quality parameters, voltage fluctuation and flicker quality parameters, harmonic parameters and frequency deviation quality parameters; Based on the monitoring data of each quality parameter of each node, a first weight of each quality parameter is obtained: Obtaining an initial first weight of each of the quality parameters based on a network analysis ANP method: Obtaining a stable weighted supermatrix based on the ANP method, wherein a row vector of the stable weighted supermatrix is an initial first weight assigned to each of the quality parameters; Based on the GRA method, the initial first weights of the quality parameters are modified to obtain the first weights of the quality parameters: Selecting the maximum value of each column in the stable weighted supermatrix and arranging them in sequence to obtain a reference sequence; Subtract each row in the stable weighted supermatrix from the reference sequence and take the absolute value to obtain n sequence difference values; n is the total number of nodes in the distribution network; Based on the n sequence differences, n correlation coefficient sequences are obtained; the expression is: Where: i represents the i-th correlation coefficient sequence, i = [1, 2, ..., n], the i-th correlation coefficient sequence is the i-th correlation coefficient sequence of each quality parameter, ΔQ is the difference of n sequences, min is the minimum value, max is the maximum value, ΔQ i is the ith sequence difference, ρ is the resolution coefficient, ρ∈(0,1); Based on the n correlation coefficient sequences, the correlation degree sequence is obtained; the expression is: Where: E represents the correlation sequence, which is the correlation degree of each quality parameter; Based on the association degree sequence, the first weight of each quality parameter is obtained; the expression is: Where: is the first weight of the bth quality parameter, B is the total number of quality parameters in the power quality classification model, G b is the correlation degree of the bth quality parameter; Based on the monitoring data of each quality parameter of each node, an entropy weight method is used to obtain a second weight of each quality parameter; Based on the first weight and the second weight of each of the quality parameters, a game equilibrium method is used to obtain the overall weight of each of the quality parameters; Based on the overall weight of each quality parameter and combined with the monitoring data of each quality parameter of each node, the ratio value of each node is obtained by using the intuitive fuzzy multi-attribute decision-making VIKOR method; Based on the ratio value and the overall influence of each node, the power quality parameter value of each node is obtained, and the power quality parameter value of the distribution network is obtained by summing the power quality parameter values of each node.
2. The method for determining power quality of a distribution network according to claim 1, characterized in that: The improved node degree expression is: Where: D i1 represents the improved node degree of the i-th node, K ie represents the degree of the i-th node, D avl represents the mean degree of n nodes, I i represents the propagation value of the i-th node, r ij represents the electrical distance between the i-th node and the j-th node; The short circuit ratio expression is: Where: D i2 represents the short-circuit ratio of the i-th node, S ac,i represents the short-circuit capacity of the AC system at the i-th node, S eq,i represents the source-load equivalent grid-connected capacity of the i-th node; The electrical betweenness expression is: Where: D i3 represents the electrical betweenness of the ith node, V1 is the set of nodes in the distribution network that are traditional power generation nodes, V2 is the set of nodes in the distribution network that are distributed power generation nodes, V3 is the set of nodes in the distribution network that are load nodes, S x is the output power of the xth traditional power generation node, S y is the output power of the yth distributed power node, S k is the output power of the kth load node, C x,y,k (i) is the transit degree of the i-th node, I x,y,k (i,z) is the current generated between the i-th node and the z-th node after adding a unit current source between the x-th traditional generation node and the y-th distributed generation node, between the x-th traditional generation node and the k-th load node, and between the y-th distributed generation node and the k-th load node; The self-healing degree expression is: Where: D i4 represents the self-healing degree of the i-th node, represents the capacity of the network-type device connected to the i-th node, The capacity of the i-th node connected to the networking device, S b Indicates the average capacity of all connected devices of n nodes.
3. The method for determining power quality of a distribution network according to claim 2, characterized in that: The overall influence expression is: Where: D i represents the overall influence of the i-th node, D ir represents the rth influence parameter of the ith node, D jr Represents the rth influence parameter of the jth node.
4. The method for determining power quality of a distribution network according to claim 1, characterized in that: The method of obtaining the initial first weight of each quality parameter based on the network analysis ANP method includes: An ANP network structure is constructed based on the power quality classification model. The ANP network structure includes a control layer and a network layer. The control layer includes B element sets [P1, P2, ..., P b ,…,P B ], P b Indicates the classification of the bth quality parameter. The network layer includes B element sets [C1, C2, …, C b ,…,C B ], q represents the total number of decision makers in the ANP network, q = n, e bq It represents the assignment of the mth quality parameter by the qth decision maker, taking the B element sets in the control layer as the criterion and the B element sets in the network layer as the secondary criterion, and obtaining the important value between the assignments in the B element sets in the network layer; constructing a weight matrix between each of the quality parameters based on the importance values between each of the assigned values, and further obtaining an unweighted supermatrix between each of the quality parameters; Normalizing the unweighted supermatrix to obtain a weighted supermatrix; The weighted supermatrix is stabilized to obtain the stabilized weighted supermatrix.
5. The method for determining power quality of a distribution network according to claim 1, characterized in that: The method of obtaining the second weight of each quality parameter based on the monitoring data of each quality parameter of each node by using an entropy weight method includes: Normalizing the monitoring data of each quality parameter of each node to obtain n×B normalized data; Based on the n×B normalized data, the information entropy of each quality parameter is obtained, and the effective value of the information of each quality parameter is further obtained; A second weight of each of the quality parameters is obtained based on the information validity value of each of the quality parameters.
6. The method for determining power quality of a distribution network according to claim 1, characterized in that: The method of obtaining the overall weight of each quality parameter based on the first weight and the second weight of each quality parameter by using a game equilibrium method includes: Construct the initial overall weight expression: Where: w b represents the overall weight of the b-th quality parameter, represents the second weight of the b-th quality parameter, a and s are linear combination coefficients; Expression-based Solve to obtain the values of a and s, substitute the values of a and s into the initial overall weight expression to obtain the overall weight expression, and obtain the overall weight of each quality parameter based on the overall weight expression, where T is the transpose.
7. The method for determining power quality of a distribution network according to claim 1, characterized in that: Based on the overall weight of each quality parameter, combined with the monitoring data of each quality parameter of each node, the ratio value of each node is obtained by using the intuitive fuzzy multi-attribute decision-making VIKOR method, including: The monitoring data of each quality parameter of each node is used to obtain n×B normalized data; Based on n×B normalized data and combined with the overall weight of each quality parameter, the group value and individual value of each node are obtained; the expression is: Where: F i represents the group value of the i-th node, R i represents the individual value of the i-th node, max(Y b ) represents the maximum value of the normalized data of the bth quality parameter, min(Y b ) represents the minimum value of the normalized data of the bth quality parameter, Y ib represents the normalized data of the bth quality parameter of the i-th node; Based on the group value and individual value of each node, the ratio value of each node is obtained, and the expression is: Where: X i represents the ratio value of the i-th node, minF represents the minimum value among the group values of n nodes, maxF represents the maximum value among the group values of n nodes, minR represents the minimum value among the individual values of n nodes, maxR represents the maximum value among the individual values of n nodes, and v is the decision coefficient.
8. The method for determining power quality of a distribution network according to claim 1, characterized in that: Based on the ratio value and overall influence of each node, the power quality parameter value of each node is obtained, and the expression is: G i =X i D i ; Where: G i is the power quality parameter value of the ith node, D i is the overall influence of the i-th node, X i Represents the ratio value of the i-th node.
9. A system for determining the power quality of a distribution network according to any one of claims 1 to 8, characterized in that: It includes: A first data acquisition module is used to acquire basic data of each node in the distribution network; An influence module is used to obtain the influence parameter of each node based on the basic data of each node, and obtain the overall influence of each node based on the influence parameter of each node; the influence parameters include improved node degree, short circuit ratio, electrical betweenness and self-healing degree; A second data acquisition module is used to acquire monitoring data of a quality parameter set of each node; the quality parameter set includes a voltage deviation quality parameter, a three-phase unbalance quality parameter, a voltage sag and interruption quality parameter, a voltage fluctuation and flicker quality parameter, a harmonic parameter and a frequency deviation quality parameter; The first parameter weight module obtains the first weight of each quality parameter based on the monitoring data of each quality parameter of each node: A first parameter weight unit is used to obtain an initial first weight of each quality parameter based on a network analysis ANP method: a stable weighted supermatrix is obtained based on the ANP method, and a row vector of the stable weighted supermatrix is an initial first weight assigned to each quality parameter; The first weight correction unit is used to correct the initial first weights of the quality parameters based on the GRA method to obtain the first weights of the quality parameters: A reference subunit, used for selecting the maximum value of each column in the stable weighted supermatrix and arranging them in sequence to obtain a reference sequence; A difference subunit is used to make a difference between each row in the stable weighted supermatrix and the reference sequence, and take the absolute value to obtain n sequence difference values; n is the total number of nodes in the distribution network; The correlation coefficient subunit is used to obtain n correlation coefficient sequences based on the n sequence differences; the expression is: Where: i represents the i-th correlation coefficient sequence, i = [1, 2, ..., n], the i-th correlation coefficient sequence is the i-th correlation coefficient sequence of each quality parameter, ΔQ is the difference of n sequences, min is the minimum value, max is the maximum value, ΔQ i is the ith sequence difference, ρ is the resolution coefficient, ρ∈(0,1); The association degree subunit is used to obtain an association degree sequence based on the n association coefficient sequences; the expression is: Where: E represents the correlation sequence, which is the correlation degree of each quality parameter; A first parameter weight subunit, configured to obtain a first weight of each of the quality parameters based on the association degree sequence; The expression is: Where: is the first weight of the bth quality parameter, B is the total number of quality parameters in the power quality classification model, G b is the correlation degree of the bth quality parameter; A second parameter weight module, used for obtaining a second weight of each quality parameter by adopting an entropy weight method based on monitoring data of each quality parameter of each node; An overall weight module, used to obtain an overall weight of each quality parameter by using a game equilibrium method based on the first weight and the second weight of each quality parameter; A ratio value module, for obtaining the ratio value of each node by using the VIKOR method based on the overall weight of each quality parameter and the monitoring data of each quality parameter of each node; The power quality module is used to obtain the power quality parameter value of each node based on the ratio value and overall influence of each node, and sum the power quality parameter values of each node to obtain the power quality parameter value of the distribution network.
Citation Information
Patent Citations
Power distribution network power quality comprehensive evaluation method and device and storage medium
CN116629632A
Edge-cloud collaborative fault detection method for low-voltage distribution network based on random matrix theory
US20230176560A1
