Harmonic source positioning method and device, equipment and storage medium

Abstract

The invention relates to a harmonic source positioning method and device, equipment and a storage medium. The method comprises the following steps: acquiring measurement data at each node in a power system, wherein the measurement data comprises harmonic voltage measurement data and branch harmonic current measurement data; determining an estimation matrix of node harmonic injection current according to the measurement data and a preset blind source separation algorithm; determining a target conversion matrix according to the estimation matrix, wherein the target conversion matrix is used for representing a corresponding relation between the estimation matrix and an actual matrix of the node harmonic injection current; and carrying out harmonic source positioning according to the target conversion matrix, and carrying out estimation processing on the injection current of the positioned harmonic source. By adopting the method, the harmonic source positioning accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of harmonic source positioning, and in particular to a harmonic source positioning method, device, equipment and storage medium. Background Art

[0003] In existing methods, harmonic location is usually performed based on the harmonic state estimation method, which requires determining accurate network harmonic impedance parameters and then combining the network harmonic impedance parameters with measurement data for harmonic location. However, due to component modeling errors and the difficulty in obtaining some load parameters, the estimated network harmonic impedance parameters often deviate from the true values. This results in low harmonic location accuracy when the existing harmonic state estimation method is used to solve the harmonic current for harmonic location. Summary of the invention

[0004] Based on this, it is necessary to provide a harmonic source positioning method, device, equipment and storage medium that can improve the accuracy of harmonic positioning in response to the above technical problems.

[0005] In a first aspect, the present application provides a harmonic source positioning method, comprising:

[0006] Acquire measurement data at each node in the power system, the measurement data including harmonic voltage measurement data and branch harmonic current measurement data; determine the estimation matrix of node harmonic injection current based on the measurement data and a preset blind source separation algorithm; determine the target conversion matrix based on the estimation matrix, the target conversion matrix is ​​used to characterize the correspondence between the estimation matrix and the actual matrix of node harmonic injection current; locate the harmonic source based on the target conversion matrix, and estimate the injection current of the located harmonic source.

[0007] In one embodiment, determining an estimation matrix of node harmonic injection current according to measurement data and a preset blind source separation algorithm includes:

[0008] The number of harmonic sources is determined according to the measured data; and the estimation matrix of the node harmonic injection current is determined according to the number of harmonic sources, the measured data and a preset blind source separation algorithm.

[0009] In one embodiment, determining the number of harmonic sources according to the measured data includes:

[0010] The measurement noise in the measurement data is removed, and a measurement matrix is ​​determined according to the measurement data after the measurement noise is removed; and the rank of the measurement matrix is ​​determined to determine the number of harmonic sources according to the rank of the measurement matrix.

[0011] In one embodiment, the preset blind source separation algorithm is a fast independent component analysis algorithm, and the estimation matrix of the node harmonic injection current is determined according to the number of harmonic sources, the measurement data and the preset blind source separation algorithm, including:

[0012] The measurement matrix is ​​preprocessed, including decentralization and whitening. The preprocessed measurement matrix is ​​filtered based on a moving average filter to obtain fast-varying components and slow-varying components. The fast-varying components are used to describe the short-term rapid changes of each node in the power system, and the slow-varying components are used to describe the long-term change trends of each node in the power system. The separation matrix is ​​determined based on the fast independent component analysis algorithm and the fast-varying components, and the number of columns of the separation matrix is ​​the same as the number of harmonic sources. The estimation matrix of the node harmonic injection current is determined based on the conjugate transposed matrix of the separation matrix and the measurement matrix.

[0013] In one embodiment, after obtaining the measurement data at each node in the power system, the method further includes:

[0014] The harmonic state estimation equation is obtained. The harmonic state estimation equation is used to describe the relationship between the measurement matrix, the parameter matrix of the power system, and the actual matrix of the node harmonic injection current. The measurement matrix is ​​determined according to the measurement data, and the parameter matrix is ​​determined according to the network harmonic impedance parameters of the power system.

[0015] In one embodiment, determining a target transformation matrix according to the estimated matrix includes:

[0016] A first conversion matrix is ​​determined according to the harmonic state estimation equation, the measurement matrix and the estimation matrix; a second conversion matrix is ​​determined according to the first conversion matrix and the parameter matrix; the second conversion matrix is ​​modified to obtain a target conversion matrix, wherein each row of the target conversion matrix has only one non-zero element and each column has at most one non-zero element.

[0017] In one embodiment, locating the harmonic source according to the target conversion matrix and estimating the injected current of the located harmonic source include:

[0018] Obtain the position information (i, j) of the matrix elements with a value of 1 in the target conversion matrix; locate the harmonic source according to the position information, and the harmonic source exists at the i-th node of the power system;

[0019] The injection current of the located harmonic source is estimated according to the j-th row of the estimation matrix.

[0020] In a second aspect, the present application also provides a harmonic source locating device, comprising:

[0021] An acquisition module is used to acquire measurement data at each node in the power system, wherein the measurement data includes harmonic voltage measurement data and branch harmonic current measurement data;

[0022] A first determination module, used to determine an estimation matrix of node harmonic injection current according to measurement data and a preset blind source separation algorithm;

[0023] A second determination module is used to determine a target conversion matrix according to the estimated matrix, where the target conversion matrix is ​​used to characterize the corresponding relationship between the estimated matrix and the actual matrix of the node harmonic injection current;

[0024] The positioning module is used to locate the harmonic source according to the target conversion matrix and estimate the injected current of the located harmonic source.

[0025] In one of the embodiments, the first determination module is specifically configured to determine the number of harmonic sources according to the measurement data; and determine an estimation matrix of node harmonic injection current according to the number of harmonic sources, the measurement data and a preset blind source separation algorithm.

[0026] In one embodiment, the first determination module is specifically used to remove measurement noise from the measurement data, and determine a measurement matrix based on the measurement data after the measurement noise is removed; determine the rank of the measurement matrix to determine the number of harmonic sources based on the rank of the measurement matrix.

[0027] In one of the embodiments, the preset blind source separation algorithm is a fast independent component analysis algorithm, and the second determination module is specifically used to preprocess the measurement matrix, and the preprocessing includes decentralization and whitening processing; the preprocessed measurement matrix is ​​filtered based on a moving average filter to obtain fast-changing components and slow-changing components, and the fast-changing components are used to describe the short-term rapid changes of each node in the power system, and the slow-changing components are used to describe the long-term change trends of each node in the power system; the separation matrix is ​​determined according to the fast independent component analysis algorithm and the fast-changing components, and the number of columns of the separation matrix is ​​the same as the number of harmonic sources; the estimation matrix of the node harmonic injection current is determined according to the conjugate transposed matrix of the separation matrix and the measurement matrix.

[0028] In one of the embodiments, the acquisition module is also used to obtain the harmonic state estimation equation, which is used to describe the relationship between the measurement matrix and the parameter matrix of the power system and the actual matrix of the node harmonic injection current. The measurement matrix is ​​determined based on the measurement data, and the parameter matrix is ​​determined based on the network harmonic impedance parameters of the power system.

[0029] In one embodiment, the second determination module is specifically used to determine the first transformation matrix based on the harmonic state estimation equation, the measurement matrix and the estimation matrix; determine the second transformation matrix based on the first transformation matrix and the parameter matrix; and modify the second transformation matrix to obtain a target transformation matrix, wherein each row of the target transformation matrix has only one non-zero element and each column has at most one non-zero element.

[0030] In one of the embodiments, the positioning module is specifically used to obtain the position information (i, j) of the matrix elements with a value of 1 in the target conversion matrix; locate the harmonic source according to the position information, the harmonic source exists at the i-th node of the power system; and estimate the injected current of the located harmonic source according to the j-th row of the estimation matrix.

[0031] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, any of the methods described in the first aspect is implemented.

[0032] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the methods described in the first aspect above.

[0033] In a fifth aspect, the present application further provides a computer program product, including a computer program, which, when executed by a processor, implements any of the methods described in the first aspect above.

[0034] The above harmonic source location method, device, equipment and storage medium first obtain the measurement data at each node in the power system, the measurement data includes harmonic voltage measurement data and branch harmonic current measurement data; then, determine the estimated matrix of the node harmonic injection current based on the measurement data and the preset blind source separation algorithm; then, determine the target conversion matrix based on the estimated matrix, the target conversion matrix is ​​used to characterize the correspondence between the estimated matrix and the actual matrix of the node harmonic injection current; finally, locate the harmonic source based on the target conversion matrix, and estimate the injection current of the located harmonic source. In this way, by referencing the preset blind source separation algorithm, the influence of the network harmonic impedance parameters on the harmonic source location method can be eliminated to a certain extent during the harmonic source location process, thereby improving the accuracy of harmonic source location. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0036] FIG1 is a diagram showing an application environment of a harmonic source location method according to an embodiment;

[0037] FIG2 is a schematic diagram of a flow chart of a harmonic source locating method in one embodiment;

[0038] FIG3 is a flow chart of a step of determining an estimation matrix of node harmonic injection current in one embodiment;

[0039] FIG4 is a schematic flow chart of a step of determining the number of harmonic sources in one embodiment;

[0040] FIG5 is a flow chart of the step of determining an estimation matrix of node harmonic injection current in another embodiment;

[0041] FIG6 is a schematic flow chart of a step of determining a target conversion matrix in one embodiment;

[0042] FIG7 is a schematic diagram of a flow chart of a harmonic source location step in one embodiment;

[0043] FIG8 is a schematic flow chart of a harmonic source locating method in another embodiment;

[0044] FIG9 is a schematic diagram of an IEEE14-node circuit model in one embodiment;

[0045] FIG10 is a schematic diagram showing the variation of harmonic positioning accuracy with admittance parameter error in one embodiment;

[0046] FIG11 is a block diagram of a harmonic source locating device in one embodiment;

[0047] FIG. 12 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0049] The harmonic source positioning method provided in the embodiment of the present application can be applied to the application environment shown in Figure 1. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 can collect the measurement data at each node in the power system and upload it to the server 104. The server 104 obtains the measurement data at each node in the power system. The measurement data includes harmonic voltage measurement data and branch harmonic current measurement data. Then, the estimated matrix of the node harmonic injection current is determined according to the measurement data and the preset blind source separation algorithm, and then the target conversion matrix is ​​determined according to the estimated matrix. The target conversion matrix is ​​used to characterize the corresponding relationship between the estimated matrix and the actual matrix of the node harmonic injection current. Finally, the harmonic source is located according to the target conversion matrix, and the injection current of the located harmonic source is estimated and processed. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices, and data acquisition devices. The server 104 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0050] In an exemplary embodiment, as shown in FIG2 , a harmonic source location method is provided, which is described by taking the method applied to the server in FIG1 as an example, and includes the following steps 201 to 204. Among them:

[0051] S201, obtaining measurement data at each node in the power system.

[0052] The measurement data includes harmonic voltage measurement data and branch harmonic current measurement data.

[0053] Optionally, the measurement data may be obtained by monitoring by a power quality monitoring device, wherein the power quality monitoring device may be configured on a preset measurement node in the power system.

[0054] Optionally, the harmonic voltage measurement data of each measurement node and the branch harmonic current measurement data of the line connected to the node can be obtained through the power quality monitoring equipment. It can be understood that the power quality monitoring equipment can collect harmonic voltage and branch harmonic current at preset time intervals to obtain harmonic voltage measurement data and branch harmonic current measurement data.

[0055] S202, determining an estimation matrix of node harmonic injection current according to the measurement data and a preset blind source separation algorithm.

[0056] Optionally, the harmonic voltage and branch harmonic current collected in the power system are caused by node harmonic injection current. Therefore, the node harmonic injection current can be estimated based on the measurement data, and all estimated node harmonic injection currents are summarized into a matrix to obtain an estimated matrix of node harmonic injection current. It can be understood that the measurement matrix can be determined based on the measurement data, and the measurement matrix The dimension is (c+f)×N, that is, , where c is the number of measurement nodes in the power system, f is the number of branch harmonic currents, N is the number of sampling points, and the estimation matrix of node harmonic injection current is The dimension of can be s×N, that is, , where s is the number of harmonic injection currents, that is, the number of harmonic sources.

[0057] Optionally, a blind source separation algorithm is a method for separating an original signal from a mixed signal, and can decompose a mixed signal into several independent components without knowing the original signal and the mixing process. In a power system, the mixed signal can be measurement data, and the independent components obtained by decomposition correspond to node harmonic injection currents.

[0058] In one possible implementation, the preset blind source separation algorithm may be Independent Component Analysis (ICA), which is based on the statistical independence assumption of the original signals and achieves separation of the original signals by maximizing the non-Gaussianity of the separated signals.

[0059] In another possible implementation, the preset blind source separation algorithm may be a joint diagonalization algorithm, which realizes separation of the original signal by jointly diagonalizing the cumulant matrix.

[0060] In another possible implementation, the preset blind source separation algorithm may be second-order blind identification (SOBI), which is a blind source separation algorithm based on second-order statistics and uses the time delay correlation of the signal to separate the original signal.

[0061] In another possible implementation, the preset blind source separation algorithm may be an adaptive blind source separation algorithm, which can adjust separation parameters in real time according to changes in the mixed signal to separate the original information from the mixed signal.

[0062] S203, determining a target transformation matrix according to the estimated matrix.

[0063] The target conversion matrix is ​​used to characterize the correspondence between the estimated matrix and the actual matrix of the node harmonic injection current.

[0064] Optionally, there is a one-to-one correspondence between the non-zero vectors of the real matrix and the estimated matrix, but at the same time, there is uncertainty in the amplitude, phase and order. It is necessary to determine the correspondence between the two matrices, that is, the target conversion matrix. The target conversion matrix includes the topological information of the harmonic source and the node. The harmonic source can be located based on the target conversion matrix. The correspondence between the real matrix and the estimated matrix can be expressed by the following formula:

[0065]

[0066] in, The real matrix for node harmonic injection currents, The dimension is n×N, where n is the number of nodes in the power system. , is the estimated matrix of node harmonic injection current, is the target transformation matrix, , the target transformation matrix contains the change information of order, amplitude and phase, where , Can be and The order change matrix between ,like and Correspondingly, the corresponding matrix element value in the order change matrix is ​​1, that is , otherwise the matrix element value is 0, Can be and The amplitude and phase change matrix between , It is a diagonal matrix, and the diagonal elements represent the amplitude and phase difference.

[0067] In a possible implementation, the target conversion matrix may be determined according to the harmonic state estimation equation, the measurement matrix, and the estimation matrix.

[0068] In another possible implementation, the target conversion matrix can be obtained according to the pre-trained neural network model, wherein the estimated conversion matrix can be used as input and the target conversion matrix can be used as output. When the actual matrix of the node harmonic injection current is known, the measured data can be collected to obtain the estimated matrix of the node harmonic injection current, and the corresponding relationship can be determined according to the actual matrix and the estimated matrix. Through a large number of experiments, training sample data can be obtained, and the training sample data can include the estimated matrix and the corresponding relationship determined according to the actual matrix and the estimated matrix.

[0069] S204, locating the harmonic source according to the target conversion matrix, and estimating the injected current of the located harmonic source.

[0070] Optionally, since the target conversion matrix includes topological information of harmonic sources and nodes, the harmonic source can be located according to the matrix element values ​​and position information of the matrix elements in the target conversion matrix, and the injected current of the located harmonic source can be estimated according to the position information and the estimation matrix.

[0071] Optionally, harmonic source positioning may be determining the location where harmonics are generated, that is, at which node of the power system the harmonic source exists, and after determining at which node the harmonic source exists, estimating the injected current of the harmonic source in the node.

[0072] It is understandable that there may be multiple harmonic sources, and in a power system, there may be multiple nodes with harmonic sources.

[0073] The above harmonic source location method first obtains the measurement data at each node in the power system, and the measurement data includes harmonic voltage measurement data and branch harmonic current measurement data; then, the estimation matrix of the node harmonic injection current is determined according to the measurement data and the preset blind source separation algorithm; then, the target conversion matrix is ​​determined according to the estimation matrix, and the target conversion matrix is ​​used to characterize the correspondence between the estimation matrix and the actual matrix of the node harmonic injection current; finally, the harmonic source is located according to the target conversion matrix, and the injection current of the located harmonic source is estimated. In this way, compared with the prior art, by citing the preset blind source separation algorithm, the influence of the network harmonic impedance parameters on the harmonic source location method can be eliminated to a certain extent in the process of harmonic source location, thereby improving the accuracy of harmonic source location.

[0074] In an exemplary embodiment, as shown in FIG3 , optionally, determining an estimation matrix of node harmonic injection current according to measurement data and a preset blind source separation algorithm includes the following steps 301 to 302 . Among them:

[0075] S301, determining the number of harmonic sources according to measurement data.

[0076] Optionally, since the measured data is caused by node harmonic injection current, the number of harmonic sources can be determined based on the measured data.

[0077] In one possible implementation, the number of harmonic sources may be determined based on a pre-trained harmonic source number determination model, the measured data may be input as input into the harmonic source number determination model, and the number of harmonic sources may be output.

[0078] In another possible implementation manner, a measurement matrix is ​​determined according to the measurement data, and then the rank of the measurement matrix is ​​used as the number of harmonic sources.

[0079] Optionally, since the harmonic sources are independent of each other, the node harmonic injection current vectors are linearly independent, so ,in, is the rank of the real matrix, s is the number of harmonic sources; since the measured data are caused by the node harmonic injection current, the measured data are all linear changes of the node harmonic injection current vector, and the measurement matrix is ​​a part of it, therefore, , when the measurement network satisfies observability, we can get , so the number of harmonic sources is equal to the rank of the measurement matrix.

[0080] Optionally, as shown in FIG4 , determining the number of harmonic sources according to the measured data includes the following steps 401 to 402 . Among them:

[0081] S401, removing measurement noise from measurement data, and determining a measurement matrix according to the measurement data after the measurement noise is removed.

[0082] Optionally, due to the existence of measurement errors, the rank of the solved measurement matrix is ​​often greater than the actual value. Therefore, the tiny singular value components in the measurement data, namely the measurement noise, can be removed, and then the measurement matrix is ​​determined based on the measurement data after removing the measurement noise to obtain the precise rank of the measurement matrix and reduce the impact of errors.

[0083] Optionally, the measurement noise may be removed by any of the following methods: principal component analysis, mean filtering, Gaussian filtering, wavelet transform, linear regression model, etc., which is not limited in the embodiments of the present application.

[0084] S402, determining the rank of the measurement matrix, so as to determine the number of harmonic sources according to the rank of the measurement matrix.

[0085] Optionally, the rank of the measurement matrix is ​​solved. According to the above analysis, the rank of the measurement matrix is ​​the same as the number of harmonic sources. Therefore, after determining the rank of the measurement matrix, the value corresponding to the rank of the measurement matrix is ​​used as the number of harmonic sources.

[0086] Optionally, the rank of the measurement matrix can be calculated by calling a public function interface, or by using linear algebra methods such as QR decomposition, eigenvalue decomposition or singular value decomposition, which is not limited in the embodiments of the present application.

[0087] S302, determining an estimation matrix of node harmonic injection current according to the number of harmonic sources, measurement data and a preset blind source separation algorithm.

[0088] Optionally, the preset blind source separation algorithm may be a fast independent component analysis algorithm (Fast Independent Component Analysis, Fast ICA), as shown in FIG5 , determining the estimated matrix of node harmonic injection current according to the number of harmonic sources, measurement data and the preset blind source separation algorithm, including the following steps 501 to 504. Among them:

[0089] S501, preprocessing the measurement matrix.

[0090] Among them, preprocessing includes decentralization and whitening.

[0091] Optionally, decentralizing the measurement matrix, i.e., zero-meaning it, can eliminate the influence of the mean of the measurement data in the matrix, making subsequent independent signal extraction more accurate and effective, and can reduce redundant information in the measurement data, so that the ICA algorithm converges to the optimal solution faster.

[0092] Optionally, when the measurement data is decentralized, the mean of each column is calculated, and then the mean of the corresponding column is subtracted from each element in the measurement matrix, so that each column of the decentralized measurement matrix will have a zero mean.

[0093] Optionally, the measurement matrix may be whitened to remove the correlation of the measurement data, so that the measurement matrix after preprocessing has a unit equation and is uncorrelated.

[0094] Optionally, the measurement matrix can be whitened by the following steps: calculating the covariance matrix of the measurement matrix, then performing eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors, then using the square roots of the eigenvalues ​​to construct a new matrix, which is used to transform the original data matrix into a whitened data matrix, and finally, applying the whitening matrix to the measurement matrix to obtain the whitened measurement matrix.

[0095] Optionally, preprocessing the measurement matrix can improve the efficiency of solving the ICA. In the embodiment of the present application, the order of performing the decentralization processing and the whitening processing on the measurement matrix is ​​not limited.

[0096] S502 , filtering the preprocessed measurement matrix based on a moving average filter to obtain a fast-varying component and a slow-varying component.

[0097] Among them, the fast-changing component is used to describe the short-term rapid changes of each node in the power system, and the slow-changing component is used to describe the long-term change trend of each node in the power system.

[0098] Optionally, since power load data often does not have statistical independence, the measurement data does not meet the application conditions of the ICA algorithm. The load change can be the superposition of fast-changing components and slow-changing components, and the fast-changing components can be approximated as statistically independent. Therefore, the preprocessed measurement matrix can be filtered based on a moving average filter to obtain the fast-changing components.

[0099] Optional, fast-changing component and the slowly varying component Satisfies the following formula:

[0100]

[0101] Optionally, the measured data may be filtered first to obtain an initial fast-varying component and an initial slowly-varying component, and then the initial fast-varying component may be preprocessed to obtain the fast-varying component. The embodiment of the present application does not limit the sequence of the above steps.

[0102] S503, determining a separation matrix according to a fast independent component analysis algorithm and fast varying components.

[0103] The number of columns of the separation matrix is ​​the same as the number of harmonic sources.

[0104] Optionally, after the Fast ICA algorithm is executed on the fast-varying component to obtain a separation matrix, since the above average filtering is a linear process, the fast-varying component and the measurement signal have the same separation matrix.

[0105] Optionally, Fast ICA is a blind source separation algorithm based on the improvement of the ICA algorithm. It can estimate the source signal from the observed mixed signal when the source signal mixing mode is unknown. When noise is not considered, the mathematical model of ICA is X(t)=AS(t), where X(t) is the d-dimensional measurement signal vector (i.e. ), S(t) is the q-dimensional independent source signal vector (i.e. ), A is the signal mixing matrix, S(t) and A are unknown. The goal of Fast ICA is to obtain the estimate of the source signal using the following formula: ,Right now Y (t)= [ y 1 (t), y 2 (t),..., y q (t)] T , is a separation matrix with a dimension of d×q, that is, , H represents conjugate transpose. In the embodiment of the present application, d is the number of harmonic sources, and q is the number of rows of the measurement matrix.

[0106] The theoretical core of the FastICA algorithm is to obtain the separation matrix The non-Gaussianity of the source signal to be sought is maximized, where the non-Gaussianity can be judged by using negative entropy. Then the objective function is constructed by using the maximum negative entropy, which is expressed as follows:

[0107]

[0108] In the formula, represents the independent components to be extracted, , is the separation matrix , is a nonlinear function and can be selected according to the distribution characteristics of the source signal. represents the arithmetic mean, is a Gaussian random variable with mean 0 and variance 1. The above J(W) maximum solution involves a nonlinear process and can be calculated using the Newton method iteration method. The iteration formula is as follows:

[0109]

[0110] in, express The derivative of , P represents the column vector number of the separation matrix, the maximum value of P is the number of harmonic sources, until the formula converges, the Pth column vector of the separation matrix is ​​obtained, and the above process is repeated to obtain the separation matrix.

[0111] S504, determining an estimation matrix of node harmonic injection current according to the conjugate transposed matrix of the separation matrix and the measurement matrix.

[0112] Optionally, after determining the separation matrix, according to the formula in the above embodiment , we can get the estimation matrix of node harmonic injection current, .

[0113] The above method determines the number of harmonic sources based on the measured data, and then determines the estimation matrix of the node harmonic injection current based on the number of harmonic sources, the measured data and the preset blind source separation algorithm. The estimation matrix can be determined by the introduced blind source separation algorithm, which eliminates the influence of the network harmonic impedance parameters on the harmonic source locating method and improves the accuracy of harmonic source locating.

[0114] In an exemplary embodiment, optionally, after obtaining the measurement data at each node in the power system, the method further includes:

[0115] The harmonic state estimation equation is obtained. The harmonic state estimation equation is used to describe the relationship between the measurement matrix, the parameter matrix of the power system, and the actual matrix of the node harmonic injection current. The measurement matrix is ​​determined according to the measurement data, and the parameter matrix is ​​determined according to the network harmonic impedance parameters of the power system.

[0116] Alternatively, the harmonic state estimation equation can be as follows:

[0117]

[0118] in, is the parameter matrix.

[0119] Optionally, the construction process of the harmonic state estimation equation is introduced below:

[0120] In an n-node distribution network, taking the hth harmonic as an example, the node voltage and current equations are as follows:

[0121]

[0122] Among them, V h is the node voltage matrix, whose dimension is n×N, that is, , V h = [ u h 1 , u h 2 ,..., u h n ] T , N represents the number of sampling points, and the node harmonic current injection matrix is express, I h = [ i h 1 , i h 2 ,..., i h n ] T , represents the hth harmonic impedance matrix, represents the sampling time index, and Represent the harmonic voltage and injected current vector of node j respectively. When there is a harmonic source at node j, there is harmonic current injection The value is not zero. For linear load bus, its harmonic current injection can be approximated to zero. Considering that the harmonic impedance matrix is ​​a full-rank matrix, multiply both sides of the node voltage and current equations by The inverse of , we get the following formula:

[0123]

[0124] In the formula, The inverse of is the harmonic admittance matrix, which is express, is a diagonally dominant sparse matrix.

[0125] Optionally, when there are c measurement nodes, the following relationship exists between the node harmonic voltage measurement data and the node voltage:

[0126]

[0127] in, represents the node measurement voltage matrix, is the voltage measurement error matrix, for With V h The transformation matrix between the nodes is 1, and the rest are 0.

[0128] Based on the above analysis, .

[0129] Optionally, the branch harmonic current measurement data and the node voltage also satisfy the following relationship:

[0130]

[0131] represents the branch harmonic current measurement vector between node m and node j, represents the current measurement error vector, is the mutual admittance between two nodes, and all branch harmonic current measurement vectors are summarized into a matrix, using It is represented by, then it has the following relationship with the node voltage and the node injection current:

[0132]

[0133] in, , f is the number of branch harmonic currents, The conversion matrix between the branch harmonic measurement current and the node voltage can be determined based on the mutual admittance between two nodes.

[0134] Optionally, based on the above analysis, the following formula can be obtained:

[0135]

[0136] Therefore, when the measurement error is not considered, the harmonic state estimation equation can be constructed: , Determined according to network harmonic impedance parameters, the network harmonic impedance parameters may include a harmonic impedance matrix And the harmonic wave admittance matrix F and the transformation matrix T.

[0137] In an exemplary embodiment, as shown in FIG6 , optionally, determining the target transformation matrix according to the estimation matrix includes the following steps 601 to 603 . Among them:

[0138] S601, determining a first conversion matrix according to a harmonic state estimation equation, a measurement matrix and an estimation matrix.

[0139] S602: Determine a second transformation matrix according to the first transformation matrix and the parameter matrix.

[0140] Optionally, according to the analysis in step 203, , therefore, combined with the harmonic state estimation equation, the following formula can be obtained:

[0141]

[0142] Optional, let ,in, is the first transformation matrix, which can be obtained by using the least squares method and the above formula .

[0143] Optionally, after determining the first conversion matrix, Get the second transformation matrix :

[0144]

[0145] S603: Correct the second transformation matrix to obtain a target transformation matrix.

[0146] Among them, there is only one non-zero element in each row of the target transformation matrix, and at most one non-zero element in each column.

[0147] Optionally, considering measurement errors and calculation errors, the obtained second transformation matrix does not satisfy that there is only one non-zero element in each row and at most one non-zero element in each column. Therefore, the second transformation matrix can be corrected to obtain the target transformation matrix.

[0148] Alternatively, the correction process can be as follows:

[0149]

[0150] in, is the matrix element of the target transformation matrix, is a matrix element of the second conversion matrix. The correction process may be to process the second conversion matrix row by row, find the position information of the maximum value of the matrix element in a row, and correct the value of the matrix element to 1, and correct the values ​​of other matrix elements to 0.

[0151] In an exemplary embodiment, as shown in FIG7 , optionally, harmonic source positioning is performed according to the target conversion matrix, and the injection current of the harmonic source positioned is estimated, including the following steps 701 to 703. Among them:

[0152] S701, obtaining position information (i, j) of matrix elements with values ​​of 1 in a target transformation matrix.

[0153] Optionally, after determining the target conversion matrix, the matrix elements with values ​​of 1 in the target conversion matrix are obtained. It can be understood that there is one or no matrix element with a value of 1 in each row of the target conversion matrix, and in the target conversion matrix, there are s matrix elements with a value of 1, and the position information of each matrix element corresponds to a harmonic source.

[0154] S702, locating a harmonic source according to the position information, where the harmonic source exists at the i-th node of the power system.

[0155] S703: estimate the injected current of the located harmonic source according to the j-th row of the estimation matrix.

[0156] Optional, It can be indicated that the harmonic source exists at the i-th node of the power system, and the corresponding harmonic source injection current corresponds to the j-th row of the estimation matrix.

[0157] As an optional implementation, as shown in FIG8 , the harmonic source locating method provided in the embodiment of the present application may include the following specific steps:

[0158] S801, obtaining measurement data at each node in the power system.

[0159] The measurement data includes harmonic voltage measurement data and branch harmonic current measurement data.

[0160] S802, obtaining a harmonic state estimation equation.

[0161] Among them, the harmonic state estimation equation is used to describe the relationship between the measurement matrix and the parameter matrix of the power system and the actual matrix of the node harmonic injection current. The measurement matrix is ​​determined according to the measurement data, and the parameter matrix is ​​determined according to the network harmonic impedance parameters of the power system.

[0162] S803, removing measurement noise from the measurement data, and determining a measurement matrix according to the measurement data after the measurement noise is removed.

[0163] S804, determining the rank of the measurement matrix, so as to determine the number of harmonic sources according to the rank of the measurement matrix.

[0164] S805, preprocessing the measurement matrix, including decentralization and whitening processing.

[0165] S806 , filtering the preprocessed measurement matrix based on a moving average filter to obtain a fast-varying component and a slow-varying component.

[0166] Among them, the fast-changing component is used to describe the short-term rapid changes of each node in the power system, and the slow-changing component is used to describe the long-term change trend of each node in the power system.

[0167] S807, determining a separation matrix according to a fast independent component analysis algorithm and fast varying components.

[0168] The number of columns of the separation matrix is ​​the same as the number of harmonic sources.

[0169] S808, determining an estimation matrix of node harmonic injection current according to the conjugate transposed matrix of the separation matrix and the measurement matrix.

[0170] S809, determining a first conversion matrix according to the harmonic state estimation equation, the measurement matrix and the estimation matrix.

[0171] S810: Determine a second transformation matrix according to the first transformation matrix and the parameter matrix.

[0172] S811, modify the second transformation matrix to obtain a target transformation matrix.

[0173] Among them, there is only one non-zero element in each row of the target transformation matrix, and at most one non-zero element in each column.

[0174] S812, obtaining position information (i, j) of the matrix elements whose values ​​are 1 in the target conversion matrix.

[0175] S813, locating the harmonic source according to the position information, where the harmonic source exists at the i-th node of the power system.

[0176] S814: estimate the injected current of the located harmonic source according to the j-th row of the estimation matrix.

[0177] In an exemplary embodiment, a node circuit model is established according to the power system, as shown in FIG9 , the node circuit module can be an IEEE14 node circuit model, and the specific parameters in the circuit model are set as follows: the harmonic source nodes are set to 7, 11, and 13, nodes 2, 7, 11, and 13 are set as observation nodes, and the nodes are configured with synchronized power quality measurement equipment, and the monitoring covers the relevant fundamental and harmonic frequency components. Through the measurement equipment, the corresponding node voltage and the branch current connected to the node can be observed. At this time, the measurement network meets the global observability. The measurement equipment collects harmonic data once every 1 minute, and a total of 5000 points are collected. The measurement error of the measurement equipment is set according to the instrument standard of Table 1, where U m and I m are the measured values ​​of harmonic voltage and current, U n and I n It is the rated voltage and rated current of the measuring device.

[0178] Table 1

[0179]

[0180] Optionally, in the calculation, the fundamental power flow can be calculated first to obtain the load power. The harmonic current emission value is determined by the load power. Considering that different nonlinear loads may present different harmonic spectra, the harmonic spectra of each harmonic source node are shown in Table 2.

[0181] Table 2

[0182]

[0183] Optionally, in order to verify the positioning effect of the harmonic source positioning method in the embodiment of the present application, the harmonic source positioning method in the embodiment of the present application is used to locate the harmonic source with the traditional ICA method and weighted least squares (WLS), and the positioning results are compared, that is, the maximum deviation of the admittance matrix is ​​changed from 5% to 30%, with an interval of 5%, and the harmonic source is located 100 times in each case to study the influence of the network harmonic impedance parameter error on the accuracy of each harmonic source positioning method. Before the comparison, the positioning accuracy rate (CLR) can be introduced as an evaluation indicator of the algorithm.

[0184]

[0185] Among them, M represents the number of accurate positioning times, accurate positioning means that all harmonic sources are located without unnecessary misjudgment, and E represents the number of execution times of the positioning method.

[0186] Optionally, when the network harmonic impedance parameter error, that is, the maximum deviation of the admittance matrix, is 15%, the positioning performance of several methods at each harmonic frequency is shown in Table 3. It can be seen from Table 3 that when the error of the known network harmonic impedance parameter is large, the harmonic source positioning method provided in the embodiment of the present application has higher positioning accuracy.

[0187] Table 3

[0188]

[0189] Optionally, when the maximum deviation of the admittance matrix increases from 5% to 30%, the change in accuracy of each method under the 5th harmonic is shown in Figure 10. It can be clearly seen from Figure 10 that with the increase of the admittance parameter error, the harmonic source positioning accuracy of all methods shows a downward trend, among which the positioning accuracy of the method provided in the embodiment of the present application has a downward trend of 1001, which is the slowest decline, the positioning accuracy of the traditional ICA method has a downward trend of 1002, and the positioning accuracy of the WLS method has a downward trend of 1003. Since the positioning accuracy of the WLS method is completely dependent on the network harmonic impedance parameters, the accuracy decreases the fastest. Therefore, the method provided in the embodiment of the present application makes full use of the known power system topology and network harmonic impedance parameter information, and thus has a stronger resistance to parameter errors, and presents certain superiority in positioning effect.

[0190] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0191] Based on the same inventive concept, the embodiment of the present application also provides a harmonic source locating device for implementing the harmonic source locating method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more harmonic source locating device embodiments provided below can refer to the limitations of the harmonic source locating method above, and will not be repeated here.

[0192] In an exemplary embodiment, as shown in FIG11 , a harmonic source positioning device 900 is provided, comprising: an acquisition module 1101, a first determination module 1102, a second determination module 1103 and a positioning module 1104, wherein:

[0193] An acquisition module 1101 acquires measurement data at each node in the power system, the measurement data including harmonic voltage measurement data and branch harmonic current measurement data;

[0194] A first determination module 1102 is used to determine an estimation matrix of node harmonic injection current according to measurement data and a preset blind source separation algorithm;

[0195] A second determination module 1103 is used to determine a target conversion matrix according to the estimated matrix, where the target conversion matrix is ​​used to characterize the corresponding relationship between the estimated matrix and the actual matrix of the node harmonic injection current;

[0196] The positioning module 1104 is used to locate the harmonic source according to the target conversion matrix and estimate the injected current of the located harmonic source.

[0197] In an exemplary embodiment, the first determination module 1102 is specifically configured to determine the number of harmonic sources according to the measurement data; and determine an estimation matrix of node harmonic injection current according to the number of harmonic sources, the measurement data and a preset blind source separation algorithm.

[0198] In an exemplary embodiment, the first determination module 1102 is specifically configured to remove measurement noise from the measurement data, and determine a measurement matrix based on the measurement data after the measurement noise is removed; and determine the rank of the measurement matrix to determine the number of harmonic sources based on the rank of the measurement matrix.

[0199] In an exemplary embodiment, the preset blind source separation algorithm is a fast independent component analysis algorithm, and the second determination module 1103 is specifically used to preprocess the measurement matrix, and the preprocessing includes decentralization and whitening processing; the preprocessed measurement matrix is ​​filtered based on a moving average filter to obtain fast-varying components and slow-varying components, the fast-varying components are used to describe the short-term rapid changes of each node in the power system, and the slow-varying components are used to describe the long-term change trends of each node in the power system; the separation matrix is ​​determined according to the fast independent component analysis algorithm and the fast-varying components, and the number of columns of the separation matrix is ​​the same as the number of harmonic sources; the estimation matrix of the node harmonic injection current is determined according to the conjugate transposed matrix of the separation matrix and the measurement matrix.

[0200] In an exemplary embodiment, the acquisition module 1101 is also used to obtain a harmonic state estimation equation, which is used to describe the relationship between the measurement matrix and the parameter matrix of the power system and the actual matrix of the node harmonic injection current. The measurement matrix is ​​determined based on the measurement data, and the parameter matrix is ​​determined based on the network harmonic impedance parameters of the power system.

[0201] In an exemplary embodiment, the second determination module 1103 is specifically used to determine the first transformation matrix based on the harmonic state estimation equation, the measurement matrix and the estimation matrix; determine the second transformation matrix based on the first transformation matrix and the parameter matrix; and modify the second transformation matrix to obtain a target transformation matrix, wherein each row of the target transformation matrix has only one non-zero element and each column has at most one non-zero element.

[0202] In an exemplary embodiment, the positioning module 1104 is specifically used to obtain the position information (i, j) of the matrix elements with a value of 1 in the target conversion matrix; locate the harmonic source according to the position information, where the harmonic source exists at the i-th node of the power system; and estimate the injected current of the located harmonic source according to the j-th row of the estimation matrix.

[0203] Each module in the above harmonic source locating device can be implemented in whole or in part by software, hardware and a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module above.

[0204] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG12. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a harmonic source positioning method is implemented.

[0205] Those skilled in the art will understand that the structure shown in FIG. 12 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0206] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps described in any of the above method embodiments when executing the computer program.

[0207] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps described in any of the above method embodiments are implemented.

[0208] In one embodiment, a computer program product is provided, including a computer program, which implements the steps described in any of the above method embodiments when executed by a processor.

[0209] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0210] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0211] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A harmonic source locating method, characterized in that: The method comprises: Acquire measurement data at each node in the power system, wherein the measurement data includes harmonic voltage measurement data and branch harmonic current measurement data; Determine an estimation matrix of node harmonic injection current according to the measurement data and a preset blind source separation algorithm; Determine a target conversion matrix according to the estimated matrix, wherein the target conversion matrix is ​​used to characterize the corresponding relationship between the estimated matrix and the actual matrix of the node harmonic injection current; The harmonic source is located according to the target conversion matrix, and the injection current of the located harmonic source is estimated.

2. The method according to claim 1, characterized in that The method of determining an estimation matrix of node harmonic injection current according to the measurement data and a preset blind source separation algorithm includes: Determining the number of harmonic sources according to the measured data; An estimation matrix of node harmonic injection current is determined according to the number of harmonic sources, the measurement data and a preset blind source separation algorithm.

3. The method according to claim 2, characterized in that The determining the number of harmonic sources according to the measured data comprises: removing measurement noise from the measurement data, and determining a measurement matrix according to the measurement data after the measurement noise is removed; The rank of the measurement matrix is ​​determined, so as to determine the number of harmonic sources according to the rank of the measurement matrix.

4. The method according to claim 3, characterized in that The preset blind source separation algorithm is a fast independent component analysis algorithm, and the estimation matrix of node harmonic injection current is determined according to the number of harmonic sources, the measurement data and the preset blind source separation algorithm, including: Preprocessing the measurement matrix, wherein the preprocessing includes decentralization and whitening processing; The preprocessed measurement matrix is ​​filtered based on a moving average filter to obtain a fast-changing component and a slow-changing component, wherein the fast-changing component is used to describe the short-term rapid changes of each node in the power system, and the slow-changing component is used to describe the long-term change trend of each node in the power system; Determine a separation matrix according to a fast independent component analysis algorithm and the fast-varying components, wherein the number of columns of the separation matrix is ​​the same as the number of harmonic sources; An estimation matrix of the node harmonic injection current is determined according to the conjugate transposed matrix of the separation matrix and the measurement matrix.

5. The method according to claim 1, characterized in that After obtaining the measurement data at each node in the power system, the method further includes: Obtain a harmonic state estimation equation, wherein the harmonic state estimation equation is used to describe the relationship between a measurement matrix and a parameter matrix of the power system and an actual matrix of the node harmonic injection current, wherein the measurement matrix is ​​determined according to the measurement data, and the parameter matrix is ​​determined according to the network harmonic impedance parameters of the power system.

6. The method according to claim 5, characterized in that The determining of the target conversion matrix according to the estimated matrix comprises: Determine a first conversion matrix according to the harmonic state estimation equation, the measurement matrix and the estimation matrix; Determine a second transformation matrix according to the first transformation matrix and the parameter matrix; The second transformation matrix is ​​modified to obtain the target transformation matrix, wherein each row of the target transformation matrix has only one non-zero element and each column has at most one non-zero element.

7. The method according to claim 6, characterized in that The locating the harmonic source according to the target conversion matrix and estimating the injected current of the located harmonic source includes: Obtain position information (i, j) of the matrix element whose value is 1 in the target transformation matrix; Positioning a harmonic source according to the position information, the harmonic source existing at an i-th node of the power system; The injection current of the located harmonic source is estimated according to the j-th row of the estimation matrix.

8. A harmonic source locating device, characterized in that: The device comprises: An acquisition module is used to acquire measurement data at each node in the power system, wherein the measurement data includes harmonic voltage measurement data and branch harmonic current measurement data; A first determination module, used to determine an estimation matrix of node harmonic injection current according to the measurement data and a preset blind source separation algorithm; A second determination module is used to determine a target conversion matrix according to the estimated matrix, wherein the target conversion matrix is ​​used to characterize the corresponding relationship between the estimated matrix and the actual matrix of the node harmonic injection current; A positioning module is used to locate the harmonic source according to the target conversion matrix and estimate the injected current of the located harmonic source.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images