A distribution network fault detection and location method based on feeder information

Abstract

The present invention provides a distribution network fault detection and positioning method based on feeder information, and the distribution network fault detection and positioning method comprises the following steps: (1): collecting the current signals of each phase, the voltage signals of each phase, the zero-sequence voltage signal, and the feeder overcurrent signals at each node of the distribution network line; (2): judging whether the distribution network line has a fault according to the phase current signal and the phase voltage signal, and if so, entering (3); if not, returning to (1); (3): determining the fault type; (4): determining the fault phase; (5): extracting the feeder overcurrent signals at each node of the fault phase; (6): determining the fault section. The present invention provides a distribution network fault detection and positioning method based on feeder information, which can quickly judge and determine the feeder and section where the fault point is located.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power detection, and in particular relates to a distribution network fault detection and positioning method based on feeder information. Background Art

[0002] The power grid is an important foundation for the development of the national economy, and the distribution network is an indispensable part of the power grid. As an indispensable commodity in daily production and life, people have put forward higher requirements for power quality, power supply reliability, and related work of the distribution network such as fault location, isolation, and emergency repair. The operation and maintenance of the distribution network involves multiple distribution networks, which have the characteristics of numerous distribution networks, complex detection, many distribution lines, rich equipment types, and changeable on-site operating environment.

[0003] When a power system fails, the measurement system will record data in a timely manner, and then confirm whether a fault has occurred through analysis. If a line fault is confirmed, staff will be arranged to go to the location of the fault to troubleshoot the fault. The staff needs to walk along the fault line and conduct a detailed inspection of the line until the specific location of the fault is found. This process takes too much time and will cause a long power outage in the area, causing unnecessary losses to the national economy. Therefore, it is necessary to quickly and accurately diagnose the fault point and remove or isolate the fault in time.

[0004] The present invention proposes a distribution network fault detection and positioning method based on feeder information, which determines whether a fault occurs and the fault phase through the current signal and voltage signal of each phase, and then determines the fault location through the feeder overcurrent signal, so as to quickly determine and locate the feeder and section where the fault point is located. Summary of the invention

[0005] The present invention provides a distribution network fault detection and positioning method based on feeder information, which can quickly judge and determine the feeder and section where the fault point is located.

[0006] The present invention is specifically a distribution network fault detection and positioning method based on feeder information, and the distribution network fault detection and positioning method comprises the following steps:

[0007] Step (1): collecting the current signals of each phase, the voltage signals of each phase, the zero-sequence voltage signals, and the feeder overcurrent signals at each node of the distribution network;

[0008] Step (2): judging whether the distribution network line is faulty according to the phase current signal and the phase voltage signal; if so, proceeding to step (3); if not, returning to step (1);

[0009] Step (3): determine the fault type;

[0010] Step (4): determine the fault phase;

[0011] Step (5): extracting feeder overcurrent signals at each node of the fault phase;

[0012] Step (6): Determine the fault segment.

[0013] The method for judging whether the distribution network line fails according to the phase current signal and the phase voltage signal is as follows: judging whether there is a phase in which the phase current signal becomes larger and the phase voltage signal decreases, and if so, the distribution network line fails.

[0014] The specific method to determine the fault type is:

[0015] Firstly, the zero-sequence voltage signal is subjected to wavelet decomposition;

[0016] Secondly, extract high frequency energy;

[0017] Again, the fault type is determined according to the high frequency of the zero-sequence voltage signal.

[0018] The algorithm for extracting the high-frequency energy of the zero-sequence voltage signal is: H =||B m (n)| 2 , where B m (n) is the high-frequency component of the wavelet coefficient of the zero-sequence voltage signal.

[0019] The method for determining the fault type based on the high frequency of the zero-sequence voltage signal is as follows: judging whether the high-frequency energy of the zero-sequence voltage signal is greater than a reference value of the high-frequency energy of the zero-sequence voltage; if so, a ground fault occurs in the distribution network; if not, a phase-to-phase short circuit fault occurs in the distribution network.

[0020] The specific method to determine the fault phase is:

[0021] Firstly, the phase voltage signal is subjected to wavelet decomposition;

[0022] Secondly, singular value decomposition is performed on the coefficient matrix after wavelet decomposition of the phase voltage signal;

[0023] Again, extract the information entropy of the diagonal matrix after singular value decomposition;

[0024] Finally, the phase with the largest information entropy is found to be the fault phase.

[0025] The algorithm for performing singular value decomposition on the coefficient matrix after wavelet decomposition of the phase voltage signal is: M = U∑V T, where M is the matrix composed of m wavelet coefficients after the phase voltage signal is decomposed by wavelet, U is an m×m-order orthogonal matrix, U is an m×n-order orthogonal matrix, ∑ is a semi-positive definite m×n-order diagonal matrix, and the elements on the ∑ diagonal are the singular values ​​of the coefficient matrix after the phase voltage signal is decomposed by wavelet.

[0026] The specific algorithm for extracting the information entropy of the diagonal matrix after singular value decomposition is: λ is the ∑ diagonal element, and L is the number of ∑ diagonal elements.

[0027] The specific method to determine the fault segment is:

[0028] Calculate the eigenvalue p at each parent node ii : If the overcurrent is detected at the parent node i and the direction is opposite to the positive direction of the whole network, p ii =-1; if the overcurrent is not detected at the parent node i, p ii =0; if the overcurrent is detected at the parent node i and the direction is the same as the positive direction of the entire network, p ii =1;

[0029] Calculate the eigenvalue p at each child node ij : If the overcurrent is detected at the child node j and the direction is opposite to the positive direction of the whole network, p ij = -1; if the overcurrent is not detected at the child node j, p ij =0; if the overcurrent is detected at the child node j and the direction is the same as the positive direction of the whole network, p ij =1;

[0030] When p ii =1 and p ij =0, the distribution network fault occurs at node i;

[0031] When p ii =1 and p ij =1, the distribution network fault occurs between nodes i and j;

[0032] When p ii = 0 and p ij =1, the distribution network fault occurs at node j.

[0033] Compared with the prior art, the beneficial effect is that the distribution network fault detection and positioning method determines whether a fault occurs and the fault phase through the current signal and voltage signal of each phase, and then determines the fault location through the feeder overcurrent signal, and can quickly determine and locate the feeder and section where the fault point is located. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1The present invention is a flowchart of a method for detecting and locating distribution network faults based on feeder information. DETAILED DESCRIPTION

[0035] The following is a detailed description of a specific implementation of a method for detecting and locating a distribution network fault based on feeder information according to the present invention in conjunction with the accompanying drawings.

[0036] like Figure 1 As shown, the distribution network fault detection and positioning method of the present invention includes the following steps:

[0037] Step (1): collecting the current signals of each phase, the voltage signals of each phase, the zero-sequence voltage signals, and the feeder overcurrent signals at each node of the distribution network;

[0038] Step (2): determine whether the phase current signal of one phase increases and the phase voltage signal decreases. If so, the distribution network line fails, and the process goes to step (3); if not, return to step (1);

[0039] Step (3): Determine the fault type: first, perform wavelet decomposition on the zero-sequence voltage signal; second, extract high-frequency energy; third, determine the fault type based on the high frequency of the zero-sequence voltage signal;

[0040] Step (4): Determine the fault phase: First, perform wavelet decomposition on the phase voltage signal; second, perform singular value decomposition on the coefficient matrix after the wavelet decomposition of the phase voltage signal; third, extract the information entropy of the diagonal matrix after the singular value decomposition; finally, find the phase with the largest information entropy as the fault phase;

[0041] Step (5): extracting feeder overcurrent signals at each node of the fault phase;

[0042] Step (6): Determine the fault segment.

[0043] The algorithm for extracting the high-frequency energy of the zero-sequence voltage signal is: H =||B m (n)| 2 , where B m (n) is the high-frequency component of the wavelet coefficient of the zero-sequence voltage signal.

[0044] The method for determining the fault type based on the high frequency of the zero-sequence voltage signal is as follows: judging whether the high-frequency energy of the zero-sequence voltage signal is greater than a reference value of the high-frequency energy of the zero-sequence voltage; if so, a ground fault occurs in the distribution network; if not, a phase-to-phase short circuit fault occurs in the distribution network.

[0045] The algorithm for performing singular value decomposition on the coefficient matrix after wavelet decomposition of the phase voltage signal is: M = U∑V T, where M is the matrix composed of m wavelet coefficients after the phase voltage signal is decomposed by wavelet, U is an m×m-order orthogonal matrix, U is an m×n-order orthogonal matrix, ∑ is a semi-positive definite m×n-order diagonal matrix, and the elements on the ∑ diagonal are the singular values ​​of the coefficient matrix after the phase voltage signal is decomposed by wavelet.

[0046] The specific algorithm for extracting the information entropy of the diagonal matrix after singular value decomposition is: λ is the ∑ diagonal element, and L is the number of ∑ diagonal elements.

[0047] The specific method to determine the fault segment is:

[0048] Calculate the eigenvalue p at each parent node ii : If the overcurrent is detected at the parent node i and the direction is opposite to the positive direction of the whole network, p ii =-1; if the overcurrent is not detected at the parent node i, p ii =0; if the overcurrent is detected at the parent node i and the direction is the same as the positive direction of the entire network, p ii =1;

[0049] Calculate the eigenvalue p at each child node ij : If the overcurrent is detected at the child node j and the direction is opposite to the positive direction of the whole network, p ij = -1; if the overcurrent is not detected at the child node j, p ij =0; if the overcurrent is detected at the child node j and the direction is the same as the positive direction of the whole network, p ij =1;

[0050] When p ii =1 and p ij =0, the distribution network fault occurs at node i;

[0051] When p ii =1 and p ij =1, the distribution network fault occurs between nodes i and j;

[0052] When p ii = 0 and p ij =1, the distribution network fault occurs at node j.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit the present invention. A person skilled in the art should understand that the specific implementation of the present invention can be modified or replaced by equivalents, but these modifications or changes are within the scope of protection of the pending claims.

Claims

1. A distribution network fault detection and positioning method based on feeder information, characterized in that: The distribution network fault detection and positioning method comprises the following steps: Step (1): collecting the current signals of each phase, the voltage signals of each phase, the zero-sequence voltage signals, and the feeder overcurrent signals at each node of the distribution network; Step (2): Determine whether a fault occurs in the distribution network line according to the phase current signal and the phase voltage signal: Determine whether there is a phase current signal that increases and a phase voltage signal that decreases. If so, the distribution network line fails, and the process goes to step (3); if not, the process goes back to step (1); Step (3): determining the fault type; the specific method for determining the fault type is: first, performing wavelet decomposition on the zero-sequence voltage signal; second, extracting high-frequency energy; third, determining the fault type according to the high frequency of the zero-sequence voltage signal: judging whether the high-frequency energy of the zero-sequence voltage signal is greater than the reference value of the high-frequency energy of the zero-sequence voltage; if so, a ground fault occurs in the distribution network; if not, a phase-to-phase short circuit fault occurs in the distribution network; The algorithm for extracting the high-frequency energy of the zero-sequence voltage signal is: ,in is the high-frequency component of the wavelet coefficient of the zero-sequence voltage signal; Step (4): determine the fault phase; Step (5): extracting feeder overcurrent signals at each node of the fault phase; Step (6): Determine the fault segment.

2. A distribution network fault detection and positioning method based on feeder information according to claim 1, characterized in that: The specific method to determine the fault phase is: Firstly, the phase voltage signal is subjected to wavelet decomposition; Secondly, singular value decomposition is performed on the coefficient matrix after wavelet decomposition of the phase voltage signal; Again, extract the information entropy of the diagonal matrix after singular value decomposition; Finally, the phase with the largest information entropy is found to be the fault phase.

3. A distribution network fault detection and positioning method based on feeder information according to claim 2, characterized in that: The algorithm for performing singular value decomposition on the coefficient matrix after wavelet decomposition of the phase voltage signal is: ,in After wavelet decomposition of the phase voltage signal The matrix of wavelet coefficients, for orthogonal matrix, for orthogonal matrix, is semi-positive definite A diagonal matrix of order, The elements on the diagonal are the singular values ​​of the coefficient matrix after the phase voltage signal is decomposed by wavelet.

4. A distribution network fault detection and positioning method based on feeder information according to claim 3, characterized in that: The specific algorithm for extracting the information entropy of the diagonal matrix after singular value decomposition is: , for Diagonal elements, for The number of diagonal elements.

5. A distribution network fault detection and location method based on feeder information according to claim 4, characterized in that: The specific method to determine the fault segment is: Calculate the eigenvalues ​​at each parent node :If the parent node The overcurrent is detected at the location and the direction is opposite to the positive direction of the entire network. ; If the parent node The overcurrent is not detected at ; If the parent node The overcurrent is detected at the location and its direction is the same as the positive direction of the entire network. ; Calculate the eigenvalues ​​at each child node :If the child node The overcurrent is detected at the location and the direction is opposite to the positive direction of the entire network. ; If the child node The overcurrent is not detected at ; If the child node The overcurrent is detected at the location and its direction is the same as the positive direction of the entire network. ; when and , the distribution network fault occurs at node Department; when and , the distribution network fault occurs at node and between; when and , the distribution network fault occurs at node Place.

Images