Limited information distribution network fault positioning method based on voltage calculation deviation

Abstract

The invention discloses a finite information distribution network fault positioning method based on voltage calculation deviation. The method mainly comprises the following steps: firstly, constructing a node impedance matrix according to the structure and parameters of a distribution network; judging whether the power distribution network has a short-circuit fault or not according to the power distribution network node monitoring system based on a small number of monitoring points; when a fault occurs, a node current equation is solved through a basis tracking algorithm according to a small amount of measuring point data obtained by a monitoring system, and a fault interval is judged by observing an obtained reconstructed node current vector. Then, a fault positioning method based on interval downstream node voltage calculation deviation is provided by utilizing the relation between a measurement point interval downstream node voltage calculation value and a measurement voltage value where a fault interval is located, fault positioning is preliminarily realized, and then a post-fault full-node voltage value of a system is deduced according to a reconstructed node current vector, so that the fault positioning accuracy is improved. And calculating the deviation by using the downstream node voltage of the fault section to further locate the fault position, and finally realizing the fault location of the power distribution network in combination with the measurement information and the reconstruction information.

Description

Technical Field

[0001] The invention belongs to the field of relay protection of electric power systems, and in particular relates to a limited information distribution network fault location method based on voltage calculation deviation. Background Art

[0002] The rapid and accurate positioning of the faulty section of the distribution network is a prerequisite for timely isolating the faulty area and restoring power supply to the non-faulty area, which is of great significance for improving system reliability. According to statistics, more than 80% of the faults in the power grid occur on the distribution network side. With the access of a large number of distributed power sources to the distribution network, it becomes an active network with bidirectional flow of normal operating power and fault current, making the positioning of the faulty section complicated, and the traditional positioning algorithm is no longer applicable. At the same time, the traditional fault location method needs to meet the strict synchronization of data and high sampling frequency. The cost of devices that meet these two conditions is high, which makes many fault location methods difficult to use in practice. Therefore, the research of new technologies for locating distribution network fault sections that meet the accuracy is a technical problem that needs to be solved urgently. Summary of the invention

[0003] In order to solve the technical problems raised by the above background technology, the present invention provides a limited information distribution network fault location method based on voltage calculation deviation. It can obtain full node information through the basis pursuit compressed sensing algorithm based on the data information measured by a small number of measurement devices in the distribution network, so as to accurately perform fault distance measurement based on the voltage calculation deviation.

[0004] In order to achieve the above technical objectives, the technical solution of the present invention is:

[0005] A limited information distribution network fault location method based on voltage calculation deviation includes the following steps: (1) collecting the distribution network topology and related parameters and constructing the distribution network node impedance matrix;

[0006] (2) Build a distribution network node monitoring system based on a small number of monitoring points to monitor the voltage and current signals of a small number of nodes in the distribution network in real time;

[0007] (3) Determine whether a fault has occurred in the distribution network by monitoring the node line voltage;

[0008] (4) When a fault occurs in the distribution network, the modulus of the positive sequence line voltage change at each monitoring node is calculated;

[0009] (5) Using the basis pursuit compressed sensing algorithm, the node current vector and node current value are reconstructed, and the fault interval is determined based on the reconstructed node current vector;

[0010] (6) Using the original measurement point data of the downstream node of the measurement point interval where the fault interval is located, the voltage calculation deviation method is used to obtain the fault location of the measurement point information;

[0011] (7) Use the reconstructed node current vector to back-calculate the voltage value of all nodes in the system after the fault;

[0012] (8) Obtain the fault location of the reconstructed information by calculating the voltage calculation deviation of the downstream node of the fault interval;

[0013] (9) combining the measurement point information and the reconstruction information in step (6) and step (8);

[0014] (10) Accurately locate the fault.

[0015] Optionally, in step (4), when a fault occurs in the distribution network, the fault voltage U′ and the fault current I′ of the measuring point are obtained when the system fails, and the positive sequence voltage data of the system fault steady-state period and the normal operation period are used to obtain the corresponding node positive sequence line voltage change modulus ΔU m×1 , where the modulus of the positive sequence voltage change at the jth node is calculated as follows:

[0016]

[0017] Where: is the positive sequence measured voltage before the jth node fault, U j is the positive sequence measured voltage after the jth node fault.

[0018] Optionally, in step (5), an underdetermined node injection fault current equation is constructed and calculated by a basis pursuit-based compressed sensing algorithm to obtain a reconstructed node current vector ΔI and a node current calculation value I;

[0019] ΔU m×1 =ΔZ m×N ΔI N×1 ;

[0020] In the formula, ΔI N×1 To reconstruct the node current vector, ΔZ m×N is the node impedance matrix during normal operation of the distribution network, ΔU m×1 is the voltage variation of positive sequence fault;

[0021] Observe the reconstructed node current vector ΔI to determine the fault interval;

[0022]

[0023] ΔI N×1 is an n×1 matrix with non-zero elements ΔI i , ΔI j The order i and j occupied are the nodes of the fault interval, so the fault interval is ij.

[0024] Optionally, in step (6), considering the instability of the compressed sensing algorithm, the original measuring point data is first used to calculate the voltage calculation deviation of the measuring point downstream of the fault interval. The smaller the voltage calculation deviation, the better the voltage calculation deviation. 1 The closer the corresponding position is to the fault point; the following formula is the voltage calculation deviation of the downstream node in the measuring point interval;

[0025]

[0026] Among them, s 1 ,s 1 From 1 to m 1 The natural number of m 1 D is the ratio of the length of the measuring device section where the fault section is located to the required accuracy; q (s 1 ) 1 is the voltage calculation deviation of the downstream node q of the nearest measuring device interval on both sides of the fault interval pq, U(q) and U′(q) are the measured voltages of node q before and after the fault, respectively, and Z′ Tf The line impedance between node p and fault point f, Z qq , Z qp is the element value of the node impedance matrix between node p and node q during normal operation.

[0027] Optionally, step (8) calculates the voltage calculation deviation of the downstream node of the fault interval by the following formula. The smaller the voltage calculation deviation, the higher the voltage calculation deviation. 2 The closer the corresponding position is to the fault point;

[0028]

[0029] Among them, s 2 ,s 2 From 1 to m 2 The natural number of m 2 D is the ratio of the length of the interval where the fault point is located to the required accuracy; j (s 2 ) 2 is the voltage calculation deviation of node j, Z′ is the system node impedance matrix when the faulty node is considered after the fault occurs, I′ is the system node current matrix when the faulty node is considered after the fault occurs, U′ is the voltage value of all nodes after the fault occurs, U′, Z jj , Z ji is the element value of the node impedance matrix between node i and node j during normal operation, Z Tf =Z Ti +Z Lif , Z Ti is the equivalent impedance between node 0 and node i, Z Lif is the line impedance between node i and fault point f.

[0030] Optionally, step (9) combines the results of step (6) and step (8) to further locate the fault using the following formula:

[0031]

[0032] Among them, D j (s) 1 The voltage calculation deviation calculated using the original measurement point data, D j (p) 2 The voltage calculation deviation is calculated by reconstructing the data using the compressed sensing algorithm, l ij , l pq are the lengths of line ij and line pq respectively; a and b are the weights of the two results respectively.

[0033] Optionally, step (10) obtains the position s when the voltage calculation deviation D is the smallest. At this time, the proportion of the fault distance between the fault point and node i to the fault interval is

[0034] Where s is a natural number from 1 to m; m is the ratio of the length of the interval where the fault point is located to the required accuracy.

[0035] The beneficial effects brought by adopting the above technical solution are:

[0036] The present invention provides a new method for locating distribution network faults by using a small amount of measurement point data in the distribution network through voltage calculation deviation. The present invention can effectively solve the problems of inaccurate fault location in the distribution network and high cost of measuring devices. The present invention is widely applicable, and it has been verified that changing the location of the fault point, grid parameters, fault type, etc. will not affect the accuracy of fault location performed by the present invention. The present invention can accurately locate the fault, effectively reduce the scope of power outages, and improve the safety, economy and reliability of distribution network operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 is a flow chart of the present invention;

[0039] Figure 2 It is a schematic diagram of the IEEE33 node topology;

[0040] Figure 3 This is a schematic diagram of the reconstructed node vector when a two-phase AB short circuit occurs at 20% of section 12-13;

[0041] Figure 4 It is a schematic diagram of voltage calculation deviation at downstream measuring point 17 when a two-phase AB short circuit occurs at 20% of section 12-13;

[0042] Figure 5 A schematic diagram of voltage calculation deviation at the downstream node 13 of the fault interval when a two-phase AB short circuit occurs at 20% of the section 12-13;

[0043] Figure 6 A schematic diagram of voltage calculation deviation when a two-phase AB short circuit occurs at 20% of section 12-13;

[0044] Figure 7 The diagram shows the positioning results and errors when different fault types are set at 20% of each section of the IEEE33 node. DETAILED DESCRIPTION

[0045] The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0046] The present invention proposes a limited information distribution network fault location method based on voltage calculation deviation, such as Figure 1 As shown, the following steps are included:

[0047] Step (1) collects the distribution network topology and related parameters and constructs the distribution network node impedance matrix.

[0048] In the embodiment of the present invention, the topological structure of the distribution network is collected, and the data related to the topological structure are collected, including the branch admittance between each node, the admittance between each node and the zero potential, and the total number of nodes N, and all the nodes in the topological structure are numbered; the node impedance matrix Z of the distribution network is constructed according to the topological structure and line parameters of the distribution network. N×N , and for Z N×N All elements in the module are taken to get the node impedance module matrix Z m ;

[0049]

[0050]

[0051]

[0052] N is the total number of nodes in the distribution network topology, Y i·i is the self-admittance of the ith node, Y i·j is the mutual admittance between the i-th node and the j-th node, Z i·i is the self-impedance of the ith node, Z i·j is the mutual impedance between the ith node and the jth node, y i·0is the branch admittance between the ith node and the zero potential point, y i·j is the branch admittance between the i-th node and the j-th node.

[0053] Step (2) constructs a distribution network node monitoring system based on a small number of monitoring points to monitor the voltage and current signals of a small number of nodes in the distribution network in real time.

[0054] In an embodiment of the present invention, a distribution network node monitoring system based on a small number of monitoring points is constructed based on the installed μPMU device to monitor the voltage and current signals of a small number of nodes in the distribution network in real time; the measurement points that must be installed in the distribution network are: nodes at the outlet of the reference power supply of the distribution network; nodes with 3 branches or more; and end nodes in the topological structure of the distribution network.

[0055] Step (3) determines whether a fault occurs in the distribution network by monitoring the node line voltage.

[0056] In the embodiment of the present invention, whether a fault occurs in the distribution network is determined according to the following formula:

[0057] U j·min <<U z·d ;

[0058] Where U j·min is the minimum line voltage of the jth node, U z·d As the setting value, take U z·d =(0.5~0.8)U N , U N is the rated line voltage of the system.

[0059] Step (4) When a fault occurs in the distribution network, the modulus of the positive sequence line voltage change at each monitoring node is calculated.

[0060] In the embodiment of the present invention, when a fault occurs in the distribution network, the fault voltage U′ and the fault current I′ of the measuring point when the system fails are obtained, and the positive sequence voltage data of the system fault steady-state period and the normal operation period are used to obtain the corresponding node positive sequence line voltage change modulus ΔU m×1 , where the calculation method of the modulus value of the positive sequence voltage change at the jth node is as follows:

[0061]

[0062] Where: is the positive sequence measured voltage before the jth node fault, U j is the positive sequence measured voltage after the jth node fault.

[0063] Step (5) uses the basis pursuit compressed sensing algorithm to reconstruct the node current vector and the node current value, and determines the fault interval based on the reconstructed node current vector.

[0064] In the embodiment of the present invention, an underdetermined node injection fault current equation is constructed and calculated by a basis pursuit-based compressed sensing algorithm to obtain a reconstructed node current vector ΔI and a node current calculation value I;

[0065] ΔU m×1 =ΔZ m×N ΔI N×1 ;

[0066] In the formula, ΔI N×1 To reconstruct the node current vector, ΔZ m×N is the node impedance matrix during normal operation of the distribution network, ΔU m×1 is the voltage variation of positive sequence fault;

[0067] Observe the reconstructed node current vector ΔI to determine the fault interval;

[0068]

[0069] ΔI N×1 is an n×1 matrix with non-zero elements ΔI i , ΔI j The order i and j occupied are the nodes of the fault interval, so the fault interval is ij.

[0070] Step (6) uses the original measuring point data of the downstream node of the measuring point interval where the fault interval is located to obtain the measuring point information fault location through the voltage calculation deviation method.

[0071] In the embodiment of the present invention, considering the instability of the compressed sensing algorithm, the voltage calculation deviation of the downstream measuring point of the measuring point interval where the fault interval is located is first calculated using the original measuring point data. The smaller the voltage calculation deviation, the greater the voltage calculation deviation. 1 The closer the corresponding position is to the fault point; the following formula is the voltage calculation deviation of the downstream node in the measuring point interval;

[0072]

[0073] Among them, s 1 ,s 1 From 1 to m 1 The natural number of m 1 D is the ratio of the length of the measuring device section where the fault section is located to the required accuracy; q (s 1 ) 1 is the voltage calculation deviation of the downstream node q of the nearest measuring device interval on both sides of the fault interval pq, U(q) and U′(q) are the measured voltages of node q before and after the fault, respectively, and Z′ Tf The line impedance between node p and fault point f, Z qq , Z qpis the element value of the node impedance matrix between node p and node q during normal operation.

[0074] Step (7) uses the reconstructed node current vector ΔI to back-calculate the post-fault full-node voltage value U′ of the system.

[0075] Step (8) calculates the voltage calculation deviation of the downstream node in the fault interval to obtain the reconstructed information fault location.

[0076] In the embodiment of the present invention, the voltage calculation deviation of the downstream node of the fault interval is calculated by the following formula. The smaller the voltage calculation deviation, the higher the voltage calculation deviation. 2 The closer the corresponding position is to the fault point;

[0077]

[0078] Among them, s 2 ,s 2 From 1 to m 2 The natural number, m 2 D is the ratio of the length of the interval where the fault point is located to the required accuracy; j (s 2 ) 2 is the voltage calculation deviation of node j, Z′ is the system node impedance matrix when the faulty node is considered after the fault occurs, I′ is the system node current matrix when the faulty node is considered after the fault occurs, U′ is the voltage value of all nodes after the fault occurs, U′, Z jj , Z ji is the element value of the node impedance matrix between node i and node j during normal operation, Z Tf =Z Ti +Z Lif , Z Ti is the equivalent impedance between node 0 and node i, Z Lif is the line impedance between node i and fault point f.

[0079] Step (9) combines the measurement point information and reconstruction information in steps (6) and (8).

[0080] In the embodiment of the present invention, the results in step (6) and step (8) are combined to further locate the fault using the following formula:

[0081]

[0082] Among them, D j (s) 1 The voltage calculation deviation calculated using the original measurement point data, D j (p) 2 The voltage calculation deviation is calculated by reconstructing the data using the compressed sensing algorithm, l ij , l pqare the lengths of line ij and line pq respectively; a and b are the weights of the two results respectively.

[0083] Step (10) is to accurately locate the fault.

[0084] In the embodiment of the present invention, the position s when the voltage calculation deviation D is the smallest is obtained. At this time, the proportion of the fault distance between the fault point and the node i to the fault interval is

[0085] Where s is a natural number from 1 to m; m is the ratio of the length of the interval where the fault point is located to the required accuracy.

[0086] The present invention is illustrated below by way of examples.

[0087] Build IEEE33 node simulation diagram in MATLAB / SIMULINK, the line topology is as follows Figure 2 As shown. The sampling frequency of the simulation is 4 kHz, the reference voltage at the head end of the power supply network in the simulation is 12.66 kV, the neutral point direct grounding operation mode is adopted, and the total network load is 5084.26+j2547.32 kVA. There are a total of 33 nodes, 32 branches, and 5 interconnecting switch branches in the system. For the area between nodes 0-1, the line impedance is very small and can be regarded as the equivalent impedance of the power supply. The measurement points are installed at 7 locations, namely nodes 1, 2, 5, 17, 21, 24 and 32. When generating the node admittance matrix, the reference node No. 0 needs to be eliminated, so the power-saving admittance matrix is ​​of order 32, and the areas that can be located are the 32 areas contained in nodes 1 to 33 in the distribution network;

[0088] In the embodiment of the present invention, Figure 3 As shown, the fault point is set at 20% of the section 12-13 in the example of the present invention, and the fault type is AB two-phase short circuit. At this time, the positive sequence fault voltage change measured by 7 measuring devices is obtained by the basis tracking algorithm. Figure 3 It can be seen that there are non-zero elements at nodes 12 and 13, so the fault occurs in segment 12-13.

[0089] From the topological structure of the distribution network, it can be seen that the two-side devices at both ends of the section 12-13 exist at nodes 5 and 17, so the data at the downstream node 17 is used for the next step of positioning. The positioning accuracy is set to 0.1% of the line length.

[0090] In the embodiment of the present invention, Figure 4 It can be seen that the proportion corresponding to the minimum calculation deviation of the voltage of the downstream node in the measuring point interval is 55.5%, that is, the fault location is 55.5% of the measuring point section 5-17, which is converted to 18.774% of the section 12-13, and the relative error of ranging is 1.226%.

[0091] In the embodiment of the present invention, Figure 5 It can be seen that the proportion corresponding to the minimum calculation deviation of the voltage of the downstream node in the fault section is 24.1%, that is, the fault position is 24.1% of the fault section 12-13, and the relative error of the distance measurement is 4.1%.

[0092] Combining the above two fault judgment criteria, the fault section 12-13 accounts for 0.075503 of the measuring point section 5-17, so a=0.075503,b=0.924491,such as Figure 6 The voltage deviation is calculated accurately as shown.

[0093] Depend on Figure 6 It can be accurately known that the fault location is 23.7% of the section 12-13, and the relative error of ranging is 3.7%, which proves that the positioning effect of the present invention is accurate.

[0094] In the embodiment of the present invention, Figure 7 It can be seen that this fault location method is not affected by the fault location and fault type, which shows that the fault location method is effective.

Claims

1. A limited information distribution network fault location method based on voltage calculation deviation, characterized in that: The steps include: Step (1) collecting the distribution network topology and related parameters and constructing the distribution network node impedance matrix; Step (2) constructing a distribution network node monitoring system based on a small number of monitoring points to monitor the voltage and current signals of a small number of nodes in the distribution network in real time; Step (3) determines whether a fault occurs in the distribution network by monitoring the node line voltage; Step (4) when a fault occurs in the distribution network, calculate the modulus of the positive sequence line voltage change at each monitoring node; Step (5) uses a basis pursuit compressed sensing algorithm to reconstruct the node current vector and the node current value, and determines the fault interval according to the reconstructed node current vector; Step (6) using the original measuring point data of the downstream node of the measuring point interval where the fault interval is located through the voltage calculation deviation method to obtain the measuring point information fault location; Step (7) using the reconstructed node current vector to back-calculate the voltage value of all nodes after the fault of the system; Step (8) obtaining the fault location of the reconstructed information by calculating the voltage calculation deviation of the downstream node of the fault interval; Step (9) combines the measurement point information and reconstruction information in step (6) and step (8); Step (10) is to accurately locate the fault.

2. The method for locating faults in a distribution network based on limited information and voltage calculation deviation according to claim 1, characterized in that: Step (4) When a fault occurs in the distribution network, the fault voltage U′ and the fault current I′ of the measuring point are obtained when the system fails, and the positive sequence voltage data of the system fault steady-state period and the normal operation period are used to obtain the corresponding node positive sequence line voltage change modulus ΔU m×1 , where the modulus of the positive sequence voltage change at the jth node is calculated as follows: Where: is the positive sequence measured voltage before the jth node fault, U j is the positive sequence measured voltage after the jth node fault.

3. The method for locating faults in a distribution network based on limited information and voltage calculation deviation according to claim 1, characterized in that: Step (5) constructing an underdetermined node injection fault current equation and calculating it through a basis pursuit-based compressed sensing algorithm to obtain a reconstructed node current vector ΔI and a node current calculation value I; ΔU m×1 =ΔZ m×N ·ΔI N×1 ; In the formula, ΔI N×1 To reconstruct the node current vector, ΔZ m×N is the node impedance matrix during normal operation of the distribution network, ΔU m×1 is the voltage variation of positive sequence fault; Observe the reconstructed node current vector ΔI to determine the fault interval; ΔI N×1 is an n×1 matrix with non-zero elements ΔI i , ΔI j The order i and j occupied are nodes in the fault interval, so the fault interval is ij.

4. The method for locating faults in a distribution network based on limited information and voltage calculation deviation according to claim 1, characterized in that: Step (6) Considering the instability of the compressed sensing algorithm, the voltage calculation deviation of the downstream measuring point of the measuring point interval where the fault interval is located is first calculated using the original measuring point data. The smaller the voltage calculation deviation, the closer the position corresponding to s1 is to the fault point. The following formula is the voltage calculation deviation of the downstream node of the measuring point interval; Where, s1, s1 are natural numbers from 1 to m1; m1 is the ratio of the length of the measuring device interval where the fault interval is located to the required accuracy; D q (s1)1 is the voltage calculation deviation of the downstream node q of the nearest measuring device interval on both sides of the fault interval pq, U(q) and U′(q) are the measured voltages of node q before and after the fault, respectively, and Z′ Tf The line impedance between node p and fault point f, Z qq , Z qp is the element value of the node impedance matrix between node p and node q during normal operation.

5. The method for locating faults in a distribution network based on limited information and voltage calculation deviation according to claim 1 is characterized in that: Step (8) calculates the voltage calculation deviation of the downstream node in the fault interval by the following formula. The smaller the voltage calculation deviation is, the closer the position corresponding to s2 is to the fault point; Where, s2, s2 are natural numbers from 1 to m2; m2 is the ratio of the length of the interval where the fault point is located to the required accuracy; D j (s2)2 is the voltage calculation deviation of node j, Z′ is the system node impedance matrix when the faulty node is considered after the fault occurs, I′ is the system node current matrix when the faulty node is considered after the fault occurs, U′ is the voltage value of all nodes after the fault occurs, U′, Z jj , Z ji is the element value of the node impedance matrix between node i and node j during normal operation, Z Tf =Z Ti +Z Lif , Z Ti is the equivalent impedance between node 0 and node i, Z Lif is the line impedance between node i and fault point f.

6. The method for locating faults in a distribution network based on limited information and voltage calculation deviation according to claim 1, characterized in that: Step (9) combines the results of step (6) and step (8) to further locate the fault using the following formula: Among them, D j (s)1 is the voltage calculation deviation calculated using the original measurement point data, D j (p)2 is the voltage calculation deviation calculated by using the compressed sensing algorithm to reconstruct the data, l ij , l pq are the lengths of line ij and line pq respectively; a and b are the weights of the two results respectively.

7. The method for locating faults in a distribution network based on limited information and voltage calculation deviation according to claim 1, characterized in that: Step (10) obtains the position s when the voltage calculation deviation D is the smallest. At this time, the proportion of the fault distance between the fault point and node i to the fault interval is Where s is a natural number from 1 to m; m is the ratio of the length of the interval where the fault point is located to the required accuracy.

Images