Distribution network current transformer metering error overrun identification method based on double decomposition

Abstract

The invention discloses a dual decomposition-based metering error overrun identification method and system for a current transformer for a distribution network. The method comprises the following steps: acquiring a three-phase current signal Xi of a secondary side of a current transformer; a variational mode decomposition constraint model is established according to constraint conditions, and a residual item signal ri of the three-phase current signal Xi is decomposed and extracted based on the variational mode decomposition constraint model; extracting an approximation coefficient of the residual item signal ri based on Haar wavelets, and determining that an approximation coefficient set xi is a detail wavelet coefficient CANi of the nth level; calculating the similarity Ki between the joint distribution probability Pmn in the high-dimensional space and the sample probability density qmn in the one-dimensional space based on the detail wavelet coefficient CANi of the nth level, and forming feature data Ci of the normal current signal; and determining an abnormal threshold T of the current transformer based on the characteristic data set C of the normal current signal, and judging whether the error of the current transformer exceeds the limit or not.

Description

Technical Field

[0001] The present invention relates to the technical field of current transformers, and more specifically, to a method for identifying over-limit measurement errors of current transformers for distribution networks based on double decomposition. Background Art

[0002] As a metering device directly associated with the user side, the accuracy of the current transformer used in the distribution network directly affects the lean management of the line loss of the distribution network and the economic interests of both the power supply and the power user. Therefore, it has received great attention from both the power supply and the power user and government departments. According to the verification regulations of JJG1189 "Measuring Transformers", the verification cycle of electromagnetic current transformers shall not exceed 10 years, and the transformers shall be fully verified with verification equipment. The technical difficulties of the traditional weekly inspection method include: the huge workload of the distribution network transformers to be inspected and the severe operating environment of the distribution network transformers. The commonly used offline detection methods are difficult to meet the requirements of high-quality services and accurate judgment. Therefore, in order to improve the automation and informatization level of on-site verification of power systems and reduce the workload of on-site staff, it is necessary to carry out research on the error over-limit identification method of the current transformer used in the distribution network. Summary of the invention

[0003] According to the present invention, a method for identifying over-limit measurement errors of current transformers for distribution networks based on double decomposition is provided to solve the technical difficulties of the traditional weekly inspection method: the workload of distribution transformers to be inspected is huge and the operating environment of distribution transformers is relatively severe. The commonly used offline detection method is difficult to meet the work requirements of high-quality service and accurate judgment.

[0004] According to a first aspect of the present invention, a method for identifying excessive measurement errors of current transformers for distribution networks based on double decomposition is provided, comprising:

[0005] Collect the three-phase current signal X on the secondary side of the current transformer i ;

[0006] A variational mode decomposition constraint model is established according to the constraint conditions, and the three-phase current signal X is decomposed and extracted based on the variational mode decomposition constraint model. i The residual signal r i ;

[0007] Extraction of residual signal based on Haar wavelet i The approximate coefficients of the approximate coefficient set ξ are determined as the detail wavelet coefficients CAn of the nth level. i ;

[0008] Based on the detail wavelet coefficients CAn at the nth level i , calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional spacemn The similarity K between i , forming the characteristic data C of the normal current signal i ;

[0009] Based on the characteristic data set C of the normal current signal, the abnormal threshold T of the current transformer is determined to determine whether the current transformer error exceeds the limit.

[0010] Optionally, a variational mode decomposition constraint model is established according to the constraint conditions, and based on the variational mode decomposition constraint model, the three-phase current signal X is decomposed and extracted. i The residual signal r i ,include:

[0011] A variational mode decomposition constraint model is established according to the constraint conditions, and an unconstrained model is established in combination with the concept of Lagrangian correlation data. All decomposed intrinsic mode signals are updated in combination with the alternating direction multiplier method to obtain an optimal set of intrinsic mode signals. The three-phase current signal is decomposed into a set of different center frequencies ω ik The intrinsic mode signal u(ω ik ), and select the residual signal r that contains the aging nature of the current transformer to the greatest extent:

[0012] X i =u(ω i1 )+u(ω i2 )+...+u(ω im )

[0013] r i =u(ω i1 )

[0014] Where m is the number of intrinsic mode signals, ω i1 The value is the smallest, then r i is the residual signal corresponding to the current signal.

[0015] Optionally, the residual signal r is extracted based on the Haar wavelet i The approximate coefficients of the approximate coefficient set ξ are determined as the detail wavelet coefficients CAn of the nth level. i ,include:

[0016] Based on Haar wavelet i Perform multiple decompositions;

[0017] The first decomposition obtains the first-level approximate coefficient CA1 i and detail coefficient CD1 i ;

[0018] Approximation coefficient CA1 for the first level i and detail coefficient CD1 iDecompose and get the second-level approximate coefficient CA2 i and detail coefficient CD2 i ;

[0019] Continuously decompose the approximate coefficient CAn-1 of the previous level i , get the approximate coefficient CAn of this level i and detail wavelet coefficients CDn i ;

[0020] Determine the approximate coefficient set as the detail wavelet coefficient CAn of the nth level i .

[0021] Optionally, based on the detail wavelet coefficients CAn at level n i , calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i , forming the characteristic data C of the normal current signal i ,include:

[0022] Based on the detail wavelet coefficients CAn at the nth level i , calculate the joint distribution probability P in high-dimensional space mn ;

[0023] Set the degrees of freedom and calculate the detail wavelet coefficients CAn i The sample probability density q mapped to one-dimensional space mn ;

[0024] Calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i ;

[0025] For similarity K i Perform optimization operations to obtain the detail wavelet coefficients CAn i The final result of mapping to one-dimensional space is the one-dimensional feature data C i .

[0026] Optionally, based on a characteristic data set C of a normal current signal, determining a current transformer abnormal threshold T and determining whether a current transformer error exceeds a limit include:

[0027] Based on the characteristic data set C of the normal current signal, the current signal below the abnormal threshold T of the current transformer is determined, and the one-dimensional characteristic data C corresponding to the current signal is extracted. 1 , C 2 , ..., C n} j, calculate its probability density distribution f(C) and probability distribution function F(C):

[0028]

[0029] Select the confidence level α and calculate the threshold T based on the probability distribution function F(C):

[0030] T=F -1 (α)

[0031] Wherein, n is the current signal lower than the abnormal threshold value T of the current transformer, ∫dx is the integral operation, and ∑ is the summation operation.

[0032] According to another aspect of the present invention, a system for identifying over-limit measurement errors of current transformers for distribution networks based on double decomposition is provided, comprising:

[0033] The three-phase current signal acquisition module is used to acquire the three-phase current signal X on the secondary side of the current transformer i ;

[0034] The residual signal extraction module is used to establish a variational mode decomposition constraint model according to the constraint conditions, and decompose and extract the three-phase current signal X based on the variational mode decomposition constraint model. i The residual signal r i ;

[0035] Determine the detail wavelet coefficient module, which is used to extract the residual signal r based on the Haar wavelet i The approximate coefficients of the approximate coefficient set ξ are determined as the detail wavelet coefficients CAn of the nth level. i ;

[0036] Forming a feature data module for the n-th level detail wavelet coefficient CAn i , calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i , forming the characteristic data C of the normal current signal i ;

[0037] The module for judging whether the current transformer exceeds the limit is used to determine the abnormal threshold value T of the current transformer based on the characteristic data set C of the normal current signal, and to judge whether the current transformer error exceeds the limit.

[0038] Optionally, the residual signal extraction module includes:

[0039] The residual term signal extraction submodule is used to establish a variational mode decomposition constraint model according to the constraint conditions, establish an unconstrained model in combination with the Lagrangian correlation data concept, and update all decomposed intrinsic mode signals in combination with the alternating direction multiplier method to obtain an optimal set of intrinsic mode signals, decomposing the three-phase current signal into a set of different center frequencies ω ik The intrinsic mode signal u(ω ik ), and select the residual signal r that contains the aging nature of the current transformer to the greatest extent:

[0040] X i =u(ω i1 )+u(ω i2 )+...+u(ω im )

[0041] r i =u(ω i1 )

[0042] Where m is the number of intrinsic mode signals, ω i1 The value is the smallest, then r i is the residual signal corresponding to the current signal.

[0043] Optionally, determining a detail wavelet coefficient module includes:

[0044] The residual signal decomposition submodule is used to decompose the residual signal r based on the Haar wavelet. i Perform multiple decompositions;

[0045] The first decomposition submodule is used to decompose the first level approximate coefficient CA1 for the first time i and detail coefficient CD1 i ;

[0046] The second decomposition submodule is used to approximate the first-level coefficient CA1 i and detail coefficient CD1 i Decompose and get the second-level approximate coefficient CA2 i and detail coefficient CD2 i ;

[0047] Continuously decompose submodules to continuously decompose the approximate coefficient CAn-1 of the previous level i , get the approximate coefficient CAn of this level i and detail wavelet coefficients CDn i ;

[0048] Determine the detail wavelet coefficient submodule, which is used to determine the approximate coefficient set as the detail wavelet coefficient CAn of the nth level i .

[0049] Optionally, a feature data module is formed, including:

[0050] Calculate the joint distribution probability submodule for the detail wavelet coefficients CAn based on the nth level i , calculate the joint distribution probability P in high-dimensional space mn ;

[0051] Calculate sample probability density submodule, used to set degrees of freedom and calculate detail wavelet coefficients CAn i The sample probability density q mapped to one-dimensional space mn ;

[0052] The similarity calculation submodule is used to calculate the joint distribution probability P in the high-dimensional space. mn And the sample probability density q in one-dimensional space mn The similarity K between i ;

[0053] Get a one-dimensional feature data submodule for similarity K i Perform optimization operations to obtain the detail wavelet coefficients CAn i The final result of mapping to one-dimensional space is the one-dimensional feature data C i .

[0054] Optionally, the module for determining whether the limit is exceeded includes:

[0055] The calculation probability submodule is used to determine the current signal below the abnormal threshold T of the current transformer based on the characteristic data set C of the normal current signal, and extract the one-dimensional characteristic data C corresponding to the current signal = {C 1 , C 2 , ..., C n} j , calculate its probability density distribution f(C) and probability distribution function F(C):

[0056]

[0057] Select the confidence level α and calculate the threshold T based on the probability distribution function F(C):

[0058] T=F -1 (α)

[0059] Wherein, n is the current signal lower than the abnormal threshold value T of the current transformer, ∫dx is the integral operation, and ∑ is the summation operation.

[0060] Therefore, it is proposed to use variational mode decomposition to extract the residual signal of the secondary current signal of the current transformer, which can effectively characterize the change process of the original signal. It is proposed to use wavelet transform to extract the approximate coefficients of each signal segment, use t-SNE to reduce the dimension of the approximate coefficient set, and finally determine the abnormal threshold of the current transformer based on normal data using density estimation method. This method is simple and effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] A more complete understanding of exemplary embodiments of the present invention may be obtained by referring to the following drawings:

[0062] Figure 1 A schematic flow chart of a method for identifying excessive measurement errors of current transformers for distribution networks based on double decomposition according to this embodiment;

[0063] Figure 2 is a schematic diagram of a secondary side current signal of a current transformer according to this embodiment;

[0064] Figure 3 A schematic diagram of the selection result of the residual term of the current signal according to this embodiment;

[0065] Figure 4 is a schematic diagram of the wavelet decomposition process described in this embodiment;

[0066] Figure 5 is a schematic diagram of the t-sne dimensionality reduction result described in this embodiment;

[0067] Figure 6 is a schematic diagram of a threshold determination result based on normal data according to this embodiment;

[0068] Figure 7 Schematic diagram of the judgment result of the error exceeding the limit of all ratio difference data of the current transformer described in this embodiment. ;

[0069] Figure 8 It is a schematic diagram of a system for identifying excessive measurement errors of current transformers for distribution networks based on double decomposition according to the present embodiment. DETAILED DESCRIPTION

[0070] Now, exemplary embodiments of the present invention are described with reference to the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to disclose the present invention in detail and completely and to fully convey the scope of the present invention to those skilled in the art. The terms used in the exemplary embodiments shown in the accompanying drawings are not intended to limit the present invention. In the accompanying drawings, the same units / elements are marked with the same reference numerals.

[0071] Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.

[0072] According to a first aspect of the present invention, a method 100 for identifying over-limit measurement error of a current transformer for a distribution network based on double decomposition is provided. Figure 1 As shown, the method 100 includes:

[0073] S101: Collecting three-phase current signal Xi on the secondary side of the current transformer;

[0074] S102: establishing a variational mode decomposition constraint model according to the constraint conditions, and decomposing and extracting the residual term signal ri of the three-phase current signal Xi based on the variational mode decomposition constraint model;

[0075] S103: extracting approximate coefficients of the residual term signal ri based on the Haar wavelet, and determining the approximate coefficient set ξ as the detail wavelet coefficients CAni of the nth level;

[0076] S104: Based on the detail wavelet coefficient CAni of the nth level, the similarity Ki between the joint distribution probability Pmn in the high-dimensional space and the sample probability density qmn in the one-dimensional space is calculated to form the characteristic data Ci of the normal current signal;

[0077] S105: Based on the characteristic data set C of the normal current signal, determine the current transformer abnormal threshold T and judge whether the current transformer error exceeds the limit.

[0078] Specifically, the present invention uses national metrology equipment to carry out experiments. The primary and secondary currents of the equipment are fixed values. A total of 67 types of secondary three-phase current data with different ratio differences are collected. The collection period is 1 second, and the current data collection time for each ratio difference is 2 minutes, that is, 67 types of current data with ratio differences, each with 120 sample points, a total of 8040 sample points were collected. This experiment simulates the gradual aging evolution process of the current transformer, that is, the current ratio difference gradually increases from 0.00% to 0.66%, and the current ratio difference changes by 0.01% each time. Among them, the current transformer is considered to be in a normal working state when the ratio difference is within 0.2%, and the rest can be regarded as the current transformer is in an error limit exceeding state.

[0079] The measured current transformer secondary current signal is as follows: Figure 2 shown.

[0080] Variational mode decomposition is used to denoise the current signals output by all current transformers, that is, the current data of 67 ratio differences and 8040 sample points. The number of variational mode decompositions is set to 4. Through this step of filtering, the quality of the data is improved, laying the foundation for subsequent work. The denoising results are as follows: Figure 3 shown.

[0081] Then, wavelet decomposition is used to perform multi-layer decomposition on the current signal of each ratio difference. The decomposition process is as follows: Figure 4 As shown in the figure, the wavelet decomposition layer is set to 5, the wavelet basis function is db1, and the three approximate coefficients obtained by decomposition are selected as preliminary aging indicators.

[0082] In order to better evaluate the aging state of the current transformer, the present invention uses t-SNE to reduce the dimension of the multi-dimensional approximate coefficients, and reduces the dimension of the approximate coefficients extracted based on wavelet to one-dimensional space. The generated feature data is as follows: Figure 5 shown.

[0083] Kernel density estimation is performed based on feature data with a ratio difference within 0.2%, and the significance level is set to 0.01. The final threshold T is -2.5852, as shown in Figure 6 shown.

[0084] Finally, the error of all current data of the current transformer is monitored by the determined threshold. It can be seen that all current data with a ratio difference higher than 0.2% are judged to be in an error-exceeding state. The specific experimental results are as follows: Figure 7 shown.

[0085] Therefore, it is proposed to use variational mode decomposition to extract the residual signal of the secondary current signal of the current transformer, which can effectively characterize the change process of the original signal. It is proposed to use wavelet transform to extract the approximate coefficients of each signal segment, use t-SNE to reduce the dimension of the approximate coefficient set, and finally determine the abnormal threshold of the current transformer based on normal data using density estimation method. This method is simple and effective.

[0086] Optionally, a variational mode decomposition constraint model is established according to the constraint conditions, and based on the variational mode decomposition constraint model, the three-phase current signal X is decomposed and extracted. i The residual signal r i ,include:

[0087] A variational mode decomposition constraint model is established according to the constraint conditions, and an unconstrained model is established in combination with the concept of Lagrangian correlation data. All decomposed intrinsic mode signals are updated in combination with the alternating direction multiplier method to obtain an optimal set of intrinsic mode signals. The three-phase current signal is decomposed into a set of different center frequencies ω ik The intrinsic mode signal u(ω ik), and select the residual signal r that contains the aging nature of the current transformer to the greatest extent:

[0088] X i =u(ω i1 )+u(ω i2 )+...+u(ω im )

[0089] r i =u(ω i1 )

[0090] Where m is the number of intrinsic mode signals, ω i1 The value is the smallest, then r i is the residual signal corresponding to the current signal.

[0091] Optionally, the residual signal r is extracted based on the Haar wavelet i The approximate coefficients of the approximate coefficient set ξ are determined as the detail wavelet coefficients CAn of the nth level. i ,include:

[0092] Based on Haar wavelet i Perform multiple decompositions;

[0093] The first decomposition obtains the first-level approximate coefficient CA1 i and detail coefficient CD1 i ;

[0094] Approximation coefficient CA1 for the first level i and detail coefficient CD1 i Decompose and get the second-level approximate coefficient CA2 i and detail coefficient CD2 i ;

[0095] Continuously decompose the approximate coefficient CAn-1 of the previous level i , get the approximate coefficient CAn of this level i and detail wavelet coefficients CDn i ;

[0096] Determine the approximate coefficient set as the detail wavelet coefficient CAn of the nth level i .

[0097] Optionally, based on the detail wavelet coefficients CAn at level n i , calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i , forming the characteristic data C of the normal current signal i ,include:

[0098] Based on the detail wavelet coefficients CAn at the nth level i , calculate the joint distribution probability P in high-dimensional space mn ;

[0099] Set the degrees of freedom and calculate the detail wavelet coefficients CAn i The sample probability density q mapped to one-dimensional space mn ;

[0100] Calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i ;

[0101] For similarity K i Perform optimization operations to obtain the detail wavelet coefficients CAn i The final result of mapping to one-dimensional space is the one-dimensional feature data C i .

[0102] Optionally, based on a characteristic data set C of a normal current signal, determining a current transformer abnormal threshold T and determining whether a current transformer error exceeds a limit include:

[0103] Based on the characteristic data set C of the normal current signal, the current signal below the abnormal threshold T of the current transformer is determined, and the one-dimensional characteristic data C corresponding to the current signal is extracted. 1 , C 2 , ..., C n} j , calculate its probability density distribution f(C) and probability distribution function F(C):

[0104]

[0105] Select the confidence level α and calculate the threshold T based on the probability distribution function F(C):

[0106] T=F -1 (α)

[0107] Wherein, n is the current signal lower than the abnormal threshold value T of the current transformer, ∫dx is the integral operation, and ∑ is the summation operation.

[0108] According to another aspect of the present invention, a system 800 for identifying over-limit measurement error of current transformer for distribution network based on double decomposition is provided. Figure 8 As shown, the system 800 includes:

[0109] The three-phase current signal acquisition module 810 is used to acquire the three-phase current signal X of the secondary side of the current transformer. i ;

[0110] The residual signal extraction module 820 is used to establish a variational mode decomposition constraint model according to the constraint conditions, and decompose and extract the three-phase current signal X based on the variational mode decomposition constraint model. i The residual signal r i ;

[0111] The detail wavelet coefficient determination module 830 is used to extract the residual signal r based on the Haar wavelet. i The approximate coefficients of the approximate coefficient set ξ are determined as the detail wavelet coefficients CAn of the nth level. i ;

[0112] The feature data module 840 is used to form the detail wavelet coefficient CAn based on the nth level. i , calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i , forming the characteristic data C of the normal current signal i ;

[0113] The module 850 for judging whether the current transformer error exceeds the limit is used to determine the abnormal threshold value T of the current transformer based on the characteristic data set C of the normal current signal, and to judge whether the current transformer error exceeds the limit.

[0114] Optionally, the residual signal extraction module includes:

[0115] The residual term signal extraction submodule is used to establish a variational mode decomposition constraint model according to the constraint conditions, establish an unconstrained model in combination with the Lagrangian correlation data concept, and update all decomposed intrinsic mode signals in combination with the alternating direction multiplier method to obtain an optimal set of intrinsic mode signals, decomposing the three-phase current signal into a set of different center frequencies ω ik The intrinsic mode signal u(ω ik ), and select the residual signal r that contains the aging nature of the current transformer to the greatest extent:

[0116] X i =u(ω i1 )+u(ω i2 )+...+u(ω im )

[0117] r i =u(ω i1 )

[0118] Where m is the number of intrinsic mode signals, ω i1 The value is the smallest, then r i is the residual signal corresponding to the current signal.

[0119] Optionally, determining a detail wavelet coefficient module includes:

[0120] The residual signal decomposition submodule is used to decompose the residual signal r based on the Haar wavelet. i Perform multiple decompositions;

[0121] The first decomposition submodule is used to decompose the first level approximate coefficient CA1 for the first time i and detail coefficient CD1 i ;

[0122] The second decomposition submodule is used to approximate the first-level coefficient CA1 i and detail coefficient CD1 i Decompose and get the second-level approximate coefficient CA2 i and detail coefficient CD2 i ;

[0123] Continuously decompose submodules to continuously decompose the approximate coefficient CAn-1 of the previous level i , get the approximate coefficient CAn of this level i and detail wavelet coefficients CDn i ;

[0124] Determine the detail wavelet coefficient submodule, which is used to determine the approximate coefficient set as the detail wavelet coefficient CAn of the nth level i .

[0125] Optionally, a feature data module is formed, including:

[0126] Calculate the joint distribution probability submodule for the detail wavelet coefficients CAn based on the nth level i , calculate the joint distribution probability P in high-dimensional space mn ;

[0127] Calculate sample probability density submodule, used to set degrees of freedom and calculate detail wavelet coefficients CAn i The sample probability density q mapped to one-dimensional space mn ;

[0128] The similarity calculation submodule is used to calculate the joint distribution probability P in the high-dimensional space. mn And the sample probability density q in one-dimensional space mn The similarity K between i ;

[0129] Get a one-dimensional feature data submodule for similarity K i Perform optimization operations to obtain the detail wavelet coefficients CAn i The final result of mapping to one-dimensional space is the one-dimensional feature data C i .

[0130] Optionally, the module for determining whether the limit is exceeded includes:

[0131] The calculation probability submodule is used to determine the current signal below the abnormal threshold T of the current transformer based on the characteristic data set C of the normal current signal, and extract the one-dimensional characteristic data C corresponding to the current signal = {C 1 , C 2 , ..., C n} j , calculate its probability density distribution f(C) and probability distribution function F(C):

[0132]

[0133] Select the confidence level α and calculate the threshold T based on the probability distribution function F(C):

[0134] T=F -1 (α)

[0135] Wherein, n is the current signal lower than the abnormal threshold value T of the current transformer, ∫dx is the integral operation, and ∑ is the summation operation.

[0136] A distribution network current transformer metering error exceeding limit identification system 800 based on double decomposition according to an embodiment of the present invention corresponds to a distribution network current transformer metering error exceeding limit identification method 100 based on double decomposition according to another embodiment of the present invention, and will not be described in detail here.

[0137] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0138] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0139] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0140] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0141] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0142] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for identifying excessive measurement error of current transformer for distribution network based on double decomposition, characterized in that: include: Collect the three-phase current signal X on the secondary side of the current transformer i ; A variational mode decomposition constraint model is established according to the constraint conditions, and the three-phase current signal X is decomposed and extracted based on the variational mode decomposition constraint model. i The residual signal r i ; Extracting residual signal based on Haar wavelet i The approximate coefficients of the approximate coefficient set ξ are determined as the detail wavelet coefficients CAn of the nth level. i ; Based on the detail wavelet coefficients CAn at the nth level i , calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i , forming the characteristic data C of the normal current signal i ; Based on the characteristic data set C of the normal current signal, the abnormal threshold T of the current transformer is determined to determine whether the current transformer error exceeds the limit.

2. The method according to claim 1, characterized in that A variational mode decomposition constraint model is established according to the constraint conditions, and the three-phase current signal X is decomposed and extracted based on the variational mode decomposition constraint model. i The residual signal r i ,include: A variational mode decomposition constraint model is established according to the constraint conditions, and an unconstrained model is established in combination with the concept of Lagrangian correlation data. All decomposed intrinsic mode signals are updated in combination with the alternating direction multiplier method to obtain an optimal set of intrinsic mode signals. The three-phase current signal is decomposed into a set of different center frequencies ω ik The intrinsic mode signal u(ω ik ), and select the residual signal r that contains the aging nature of the current transformer to the greatest extent: X i =u(ω i1 )+u(ω i2 )+...+u(ω im ) r i =u(ω i1 ) Where m is the number of intrinsic mode signals, ω i1 The value is the smallest, then r i is the residual signal corresponding to the current signal.

3. The method according to claim 1, characterized in that Extracting residual signal based on Haar wavelet i The approximate coefficients of the approximate coefficient set ξ are determined as the detail wavelet coefficients CAn of the nth level. i ,include: Based on Haar wavelet i Perform multiple decompositions; The first decomposition obtains the first-level approximate coefficient CA1 i and detail coefficient CD1 i ; The approximation coefficient CA1 for the first level i and detail coefficient CD1 i Decompose and get the second-level approximate coefficient CA2 i and detail coefficient CD2 i ; Continuously decompose the approximate coefficient CAn-1 of the previous level i , get the approximate coefficient CAn of this level i and detail wavelet coefficients CDn i ; Determine the approximate coefficient set as the detail wavelet coefficient CAn of the nth level i .

4. The method according to claim 1, characterized in that: Based on the detail wavelet coefficients CAn at the nth level i , calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i , forming the characteristic data C of the normal current signal i ,include: Based on the detail wavelet coefficients CAn at the nth level i , calculate the joint distribution probability P in high-dimensional space mn ; Set the degrees of freedom and calculate the detail wavelet coefficients CAn i The sample probability density q mapped to one-dimensional space mn ; Calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i ; For similarity K i Perform optimization operations to obtain the detail wavelet coefficients CAn i The final result of mapping to one-dimensional space is the one-dimensional feature data C i .

5. The method according to claim 1, characterized in that Based on the characteristic data set C of the normal current signal, the abnormal threshold T of the current transformer is determined to determine whether the current transformer error exceeds the limit, including: Based on the characteristic data set C of the normal current signal, the current signal below the abnormal threshold T of the current transformer is determined, and the one-dimensional characteristic data C corresponding to the current signal is extracted. n } j , calculate its probability density distribution f(C) and probability distribution function F(C): Select the confidence level α and calculate the threshold T based on the probability distribution function F(C): T=F -1 (a) Wherein, n is the current signal lower than the abnormal threshold value T of the current transformer, ∫dx is the integral operation, and ∑ is the summation operation.

6. A system for identifying over-limit measurement errors of current transformers for distribution networks based on double decomposition, characterized in that: include: The three-phase current signal acquisition module is used to acquire the three-phase current signal X on the secondary side of the current transformer i ; The residual signal extraction module is used to establish a variational mode decomposition constraint model according to the constraint conditions, and decompose and extract the three-phase current signal X based on the variational mode decomposition constraint model. i The residual signal r i ; Determine the detail wavelet coefficient module, which is used to extract the residual signal r based on the Haar wavelet i The approximate coefficients of the approximate coefficient set ξ are determined as the detail wavelet coefficients CAn of the nth level. i ; Forming a feature data module for the n-th level detail wavelet coefficient CAn i , calculate the joint distribution probability P in high-dimensional space mn And the sample probability density q in one-dimensional space mn The similarity K between i , forming the characteristic data C of the normal current signal i ; The module for judging whether the current transformer exceeds the limit is used to determine the abnormal threshold value T of the current transformer based on the characteristic data set C of the normal current signal, and to judge whether the current transformer error exceeds the limit.

7. The system according to claim 6, characterized in that Extract the residual signal module, including: The residual term signal extraction submodule is used to establish a variational mode decomposition constraint model according to the constraint conditions, establish an unconstrained model in combination with the Lagrangian correlation data concept, and update all decomposed intrinsic mode signals in combination with the alternating direction multiplier method to obtain an optimal set of intrinsic mode signals, decomposing the three-phase current signal into a set of different center frequencies ω ik The intrinsic mode signal u(ω ik ), and select the residual signal r that contains the aging nature of the current transformer to the greatest extent: X i =u(ω i1 )+u(ω i2 )+...+u(ω im ) r i =u(ω i1 ) Where m is the number of intrinsic mode signals, ω i1 The value is the smallest, then r i is the residual signal corresponding to the current signal.

8. The system according to claim 7, characterized in that Determine the detail wavelet coefficient module, including: The residual signal decomposition submodule is used to decompose the residual signal r based on the Haar wavelet. i Perform multiple decompositions; The first decomposition submodule is used to decompose the first level approximate coefficient CA1 for the first time i and detail coefficient CD1 i ; The second decomposition submodule is used to approximate the first-level coefficient CA1 i and detail coefficient CD1 i Decompose and get the second-level approximate coefficient CA2 i and detail coefficient CD2 i ; Continuously decompose submodules to continuously decompose the approximate coefficient CAn-1 of the previous level i , get the approximate coefficient CAn of this level i and detail wavelet coefficients CDn i ; Determine the detail wavelet coefficient submodule, which is used to determine the approximate coefficient set as the detail wavelet coefficient CAn of the nth level i .

9. The system according to claim 6, characterized in that Form a feature data module, including: Calculate the joint distribution probability submodule for the detail wavelet coefficients CAn based on the nth level i , calculate the joint distribution probability P in high-dimensional space mn ; Calculate sample probability density submodule, used to set degrees of freedom and calculate detail wavelet coefficients CAn i The sample probability density q mapped to one-dimensional space mn ; The similarity calculation submodule is used to calculate the joint distribution probability P in the high-dimensional space. mn And the sample probability density q in one-dimensional space mn The similarity K between i ; Get a one-dimensional feature data submodule for similarity K i Perform optimization operations to obtain the detail wavelet coefficients CAn i The final result of mapping to one-dimensional space is the one-dimensional feature data C i .

10. The system according to claim 6, characterized in that The module for judging whether the limit is exceeded includes: The calculation probability submodule is used to determine the current signal below the abnormal threshold T of the current transformer based on the characteristic data set C of the normal current signal, and extract the one-dimensional characteristic data C corresponding to the current signal = {C1, C2, ..., C n } j , calculate its probability density distribution f(C) and probability distribution function F(C): Select the confidence level α and calculate the threshold T based on the probability distribution function F(C): T=F -1 (a) Wherein, n is the current signal lower than the abnormal threshold value T of the current transformer, ∫dx is the integral operation, and ∑ is the summation operation.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

12. An electronic device, characterized in that: include: The computer readable storage medium as claimed in claim 11; as well as One or more processors are used to execute the program in the computer-readable storage medium.

Images