Power distribution network fault detection method and system based on interval identification and branch determination

Abstract

The application discloses a power distribution network fault detection method and system based on interval identification and branch determination, which comprises the following steps: normalizing the length of the power distribution network line, and obtaining a voltage line mode beta component signal by improved phase-mode transformation; obtaining a plurality of intrinsic mode function components by successively variational mode decomposition of the voltage line mode beta component signal, calculating the modal kurtosis value, selecting the modal component corresponding to the maximum kurtosis value as the characteristic mode, and calibrating the fault wave head time by using a symmetric difference energy operator; determining the fault interval based on the fault wave head time of adjacent detection points on the main line; constructing a fault branch determination matrix according to the distance relationship in the fault interval, and determining the fault section by using the corresponding determination criterion; and determining the fault point position by using the principle of double-end traveling wave method. The application determines the fault interval first, then determines the fault branch, and finally determines the fault point position, so that the accuracy and efficiency of fault detection and positioning are improved while the interference of false fault points is reduced.

Description

Technical Field

[0001] This application relates to the field of power system distribution network relay protection technology, specifically to a distribution network fault detection method and system based on section identification and branch determination. Background Technology

[0002] In power distribution systems, fault location technology is crucial for ensuring power supply stability and achieving rapid fault isolation. However, with the expansion of power distribution networks and the increasing complexity of their structures, traditional fault location methods based on the traveling wave principle face limitations in practical engineering, struggling to maintain high location accuracy in multi-branch, multi-node structures. Particularly in identifying false fault points, existing methods often rely on network-wide detection devices, leading to cumbersome data processing, complex fault determination matrix construction, and significant engineering implementation difficulties.

[0003] In recent years, the synergistic development of traveling wave detection technology and integrated primary and secondary switching equipment has provided fundamental support for the widespread deployment of intelligent switching devices with detection functions in power distribution systems. These integrated devices not only enhance the status monitoring capabilities of the power distribution network but also possess certain fault early warning and location capabilities. How to achieve accurate identification of faulty branches and precise determination of fault points in a power distribution network environment with numerous access points and flexible topologies has become a key technical challenge that urgently needs to be overcome in the field of power distribution automation. Summary of the Invention

[0004] The purpose of this application is to provide a distribution network fault detection method based on interval identification and branch determination, which solves the problems of difficulty in eliminating fault points and low positioning accuracy in existing distribution network fault location methods.

[0005] This application is achieved through the following technical solution:

[0006] Firstly, this application provides a method for detecting distribution network faults based on section identification and branch determination, comprising the following steps:

[0007] The normalized distribution network line length is obtained by using an improved phase-mode transformation to extract the voltage line mode β component signal from the three-phase voltage signal;

[0008] Successive variational mode decomposition is performed on the voltage line mode β component, characteristic modes are selected based on kurtosis, and the fault wavefront timing is calibrated using a symmetric differential energy operator;

[0009] The fault interval is determined based on the fault wavefront timing of adjacent detection points on the main line.

[0010] A fault branch determination matrix is ​​constructed based on the distance relationship within the fault interval, and the line segment where the fault is located is determined based on the corresponding determination criteria using the branch determination matrix.

[0011] Based on the identified fault section, the final measurement is performed using the double-ended traveling wave method to determine the location of the fault.

[0012] A further optimization scheme involves using an improved phase-mode transformation to obtain the voltage line mode β component signal from the three-phase voltage signal for the normalized distribution network line length. This specifically includes the following steps:

[0013] The distribution network lines are normalized, and the cable length in the equivalent distribution network is the same as the overhead line length:

[0014]

[0015] In the formula, For the equivalent length of the cable, For the original length of the cable, , These are the line mode wave velocities of traveling waves in overhead lines and cable lines, respectively.

[0016] The district distribution network consists of a main line and several branch lines. Based on the normalized cable length, traveling wave detection devices are deployed at the ends of branch lines whose length exceeds a set threshold.

[0017] When a fault occurs in the distribution network, the three-phase voltage traveling wave signals collected by each traveling wave detection device are improved by phase mode transformation, and the voltage line mode β component is extracted from them.

[0018] A further optimized scheme involves performing successive variational mode decomposition on the voltage line mode β component, selecting characteristic modes based on kurtosis, and using a symmetric differential energy operator to calibrate the fault wavefront timing. This specifically includes the following steps:

[0019] Successive variational mode decomposition is performed on the voltage line β component signal to obtain multiple intrinsic mode function components;

[0020] Calculate the kurtosis value of each intrinsic mode function component, and select the intrinsic mode function component corresponding to the maximum kurtosis value as the characteristic mode representing the fault mutation information;

[0021] The characteristic modes are calculated using a symmetric differential energy operator to determine the wavefront time of the fault traveling wave arriving at the measurement point.

[0022] A further optimized solution involves determining the fault interval based on the fault wavefront times of adjacent detection points on the main line, specifically including the following steps:

[0023] Based on the recorded fault wave head times of each detection point, for all adjacent detection point pairs on the main line, the projected distance from the fault point to the two detection points is calculated using the double-ended traveling wave method, and the fault interval is initially measured in combination with the inherent line distance between the two detection points to obtain the initial interval determination result of the fault occurrence.

[0024] Traverse all adjacent main line detection point pairs, and determine the fault range based on the relationship between the maximum value of the projected distance from the fault point to the two detection points and the actual line distance between the detection points.

[0025] A further optimized solution involves constructing a fault branch determination matrix based on distance relationships within the fault interval, and determining the line segment where the fault occurs based on the corresponding determination criteria using the branch determination matrix. This specifically includes the following steps:

[0026] Select all branch end detection points within the fault interval and construct a two-dimensional judgment matrix with rows corresponding to main line detection points and columns corresponding to branch end detection points;

[0027] Revise the two-dimensional decision matrix to obtain the correction matrix;

[0028] For the correction matrix, the fault branch determination criterion is used to determine the line section where the fault is located.

[0029] A further optimization scheme is as follows: for the modified matrix, the fault branch determination criterion is used to determine the line section where the fault occurs, specifically including the following steps:

[0030] If all elements in a certain column of the correction matrix are greater than zero, and all elements in the other columns are not greater than zero, then the fault is determined to be located on the first-level branch line corresponding to that column.

[0031] If there are two columns in the correction matrix where all elements are greater than zero and all other columns where all elements are not greater than zero, then the fault is determined to be located on the secondary branch line corresponding to the second column.

[0032] If all columns of the correction matrix are not greater than zero, then the fault is determined to be located within the current main line interval.

[0033] If all elements in a column of the correction matrix are equal to zero, then the fault is determined to be located at the branch node of the corresponding branch line in that column.

[0034] A further optimized solution involves determining the location of the fault point by performing a final measurement using the double-ended traveling wave method based on the identified fault section of the line. This specifically includes the following steps:

[0035] Based on the differences in fault location, a corresponding dual-end ranging strategy is adopted to obtain the projected distance from the fault point to the measurement point, and the calculation results of each path are unified as the equivalent fault distance.

[0036] The equivalent fault distance is restored and normalized to obtain the actual fault distance, and the location of the fault point is finally determined.

[0037] A further optimized solution involves using a corresponding dual-end ranging strategy to obtain the projected distance from the fault point to the measurement point based on the differences in fault location, and unifying the calculation results of each path as the equivalent fault distance. This specifically includes the following steps:

[0038] When the fault is determined to be located on a primary or secondary branch line, the detection point at the end of the fault branch and two adjacent measurement points on the main line are selected to form a double-end ranging path. The projected distance from the fault point to the measurement point on the main line is calculated, and the average of the two calculation results is taken as the equivalent fault distance.

[0039] When the fault is determined to be located in the main line section, the projected distance from the fault point to the upstream measurement point is calculated directly using the double-ended distance measurement path formed by two adjacent measurement points on the main line, and the calculation result is directly used as the equivalent fault distance.

[0040] When the fault is determined to be located at a branch node, the end detection point of the branch where the node is located and two adjacent measurement points on the main line are selected to form a double-end distance measurement path. The projected distance from the fault point to the measurement point on the main line is calculated, and the average of the two calculation results is taken as the equivalent fault distance.

[0041] Secondly, this application provides a distribution network fault detection system based on section identification and branch determination, including:

[0042] The component signal acquisition module is used to normalize the length of the distribution network line and uses an improved phase mode transformation to obtain the voltage line mode β component signal from the three-phase voltage signal.

[0043] The fault wavefront timing calibration module is communicatively connected to the component signal acquisition module. It is used to perform successive variational mode decomposition on the voltage line mode β component, select characteristic modes according to kurtosis, and use a symmetric differential energy operator to calibrate the fault wavefront timing.

[0044] The fault interval preliminary determination module is communicatively connected to the fault wavefront time calibration module and is used to determine the fault interval based on the fault wavefront times of adjacent detection points on the main line.

[0045] The fault branch determination module is communicatively connected to the fault section preliminary determination module. It is used to construct a fault branch determination matrix based on the distance relationship within the fault section, and to determine the line section where the fault is located based on the corresponding determination criteria using the branch determination matrix.

[0046] The fault location determination module is communicatively connected to the fault branch determination module. It is used to determine the location of the fault point by performing a final measurement based on the determined line section where the fault is located, using the principle of the double-ended traveling wave method.

[0047] Thirdly, this application provides a computer-readable storage medium storing a distribution network fault detection program based on interval identification and branch determination, wherein when the distribution network fault detection program based on interval identification and branch determination is executed by a processor, it implements the steps of the distribution network fault detection method based on interval identification and branch determination as described above.

[0048] Compared with the prior art, this application has the following advantages and beneficial effects:

[0049] This application proposes to combine the successive variational mode decomposition method with the symmetric differential energy operator for voltage traveling wave signal processing. The improved line-mode voltage signal is adaptively decomposed by successive variational mode decomposition, the main feature modes are extracted based on the kurtosis evaluation index, and the symmetric differential energy operator is used to achieve high-precision calibration of the wavefront moment, thus providing a reliable data foundation for subsequent accurate fault location.

[0050] This application addresses the difficulty in eliminating false fault points in multi-branch power grids by constructing a three-stage fault location strategy: "initial section measurement - branch identification - final measurement and location". This strategy initially delineates the fault section using traveling wave data from main line detection points, then constructs a fault branch identification matrix to identify the faulty branch, and finally achieves precise location by combining the double-ended traveling wave method. This method effectively improves the accuracy of fault branch identification and the precision of the location results while mitigating wavefront dispersion interference caused by long-distance ranging. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the exemplary embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0052] Figure 1 A flowchart of a power distribution network fault detection method based on section identification and branch determination provided in this application embodiment;

[0053] Figure 2 Another flowchart of distribution network fault detection based on section identification and branch determination is provided in the embodiments of this application;

[0054] Figure 3 This is a schematic diagram of a simulation model of a 10kV resonant grounding system provided in an embodiment of this application;

[0055] Figure 4 This is a topology diagram of feeder 1 provided in an embodiment of this application;

[0056] Figure 5 Line-mode voltage diagram of the measurement point at end B of the main line provided in the embodiments of this application;

[0057] Figure 6 This is a fault signal diagram of successive variational mode decomposition provided in an embodiment of this application;

[0058] Figure 7 Symmetric differential energy operator calibration wavefront diagrams for each measurement point provided in the embodiments of this application. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this application are only for explaining this application and are not intended to limit this application.

[0060] First, some of the technical terms used in this application will be explained to help those skilled in the art understand this application.

[0061] SVMD: Successive Variational Mode Decomposition

[0062] SDEO: Symmetric Difference Energy Operator;

[0063] TEO: Teager Energy Operator.

[0064] Firstly, such as Figures 1-2 As shown, this application provides a distribution network fault detection method based on section identification and branch determination, including the following steps:

[0065] Step S1: Normalize the distribution network line length and use an improved phase-mode transformation to obtain the voltage line mode β component signal from the three-phase voltage signal;

[0066] Step S2: Perform successive variational mode decomposition on the voltage line mode β component, select characteristic modes based on kurtosis, and use the symmetric differential energy operator to calibrate the fault wavefront time;

[0067] Step S3: Determine the fault interval based on the fault wavefront times of adjacent detection points on the main line;

[0068] Step S4: Construct a fault branch determination matrix based on the distance relationship within the fault interval, and determine the line section where the fault is located based on the corresponding determination criteria using the branch determination matrix.

[0069] Step S5: Based on the determined line section where the fault is located, perform the final measurement using the principle of the double-ended traveling wave method to determine the location of the fault point.

[0070] This application proposes to combine the successive variational mode decomposition method with the symmetric differential energy operator for voltage traveling wave signal processing. The improved line-mode voltage signal is adaptively decomposed by successive variational mode decomposition, the main feature modes are extracted based on the kurtosis evaluation index, and the symmetric differential energy operator is used to achieve high-precision calibration of the wavefront moment, thus providing a reliable data foundation for subsequent accurate fault location.

[0071] This application addresses the difficulty in eliminating false fault points in multi-branch power grids by constructing a three-stage fault location strategy: "initial section measurement - branch identification - final measurement and location". This strategy initially delineates the fault section using traveling wave data from main line detection points, then constructs a fault branch identification matrix to identify the faulty branch, and finally achieves precise location by combining the double-ended traveling wave method. This method effectively improves the accuracy of fault branch identification and the precision of the location results while mitigating wavefront dispersion interference caused by long-distance ranging.

[0072] In one embodiment, step S1, normalizing the distribution network line length, involves obtaining the voltage line mode β component signal from the three-phase voltage signal using an improved phase-mode transformation. Specifically, this includes the following steps:

[0073] Step S11: Normalize the distribution network lines so that the cable length in the equivalent distribution network is the same as the overhead line length.

[0074]

[0075] In the formula, For the equivalent length of the cable, For the original length of the cable, , These are the line mode wave velocities of traveling waves in overhead lines and cable lines, respectively.

[0076] Step S12: The distribution network consists of a main line and several branch lines. Branch lines directly connected to the main line are defined as first-level branch lines, and further branch lines connected to them are defined as second-level branch lines. The locations of the traveling wave detection devices installed at the starting point, the primary / secondary fusion switch point, and the ending point on the main line are sequentially numbered as follows: , m is the total number of detection devices on the main line; traveling wave detection devices are deployed at the ends of branch lines whose length exceeds a set threshold; specifically, the locations of the traveling wave detection devices installed at the ends of branch lines whose length exceeds the threshold δ are sequentially numbered as follows: , s is the total number of detection devices on the branch line, and for each detection point The branch nodes in the interval are numbered sequentially as follows: The threshold δ is determined based on the line structure and fault location accuracy requirements; the threshold in this application... Take 5km;

[0077] Step S13: When a fault occurs in the distribution network, the three-phase voltage traveling wave signals collected by each traveling wave detection device are subjected to improved phase mode transformation to extract the voltage line mode. Quantity Let the fault occurrence time be t, the sampling start time be t-0.1 milliseconds, and the sampling end time be t+0.4 milliseconds, acquiring a traveling wave signal data window with a total duration of 0.5 milliseconds; further, the following steps are included:

[0078] The purpose of phase-mode transformation is to decouple phases, breaking down a coupled three-phase system into three independent subsystems for power system fault analysis. For transient electrical quantities, the Karenbauer transformation is commonly used. Taking phase voltage as an example, the Karenbauer transformation formula is:

[0079] Equation (1)

[0080] in, The three-phase voltage representing the fault signal. These are the modulus signals of mode 0, mode α, and mode β after Kelvin transformation, respectively. When phases a and b are short-circuited, Substituting into That is, it cannot reflect a short circuit fault between phases a and b, and cannot be used to detect and characterize such faults; when phases a and c are short-circuited, Substituting into equation (1), we get This means it cannot reflect a two-phase short-circuit fault in AC.

[0081] Therefore, to address the problem that a single modulus in existing phase mode transformations cannot characterize all fault types, an improved phase mode transformation matrix is ​​adopted, and its calculation formula is as follows:

[0082] Equation (2)

[0083] When a fault occurs, the three-phase voltage signal is transformed using the improved phase mode transformation equation (2) to obtain the voltage line mode β component U. β The calculation formula is as follows:

[0084]

[0085] In the formula, These are the voltages of phase a, phase b, and phase c collected at the detection point, respectively.

[0086] In one embodiment, step S2, adjusting the voltage line mode The components are subjected to successive variational mode decomposition, characteristic modes are selected based on kurtosis, and the fault wavefront time is calibrated using a symmetric differential energy operator.

[0087] For discrete signals :

[0088] The formula for calculating SDEO is:

[0089] Equation (3)

[0090] In the formula, For signal SDEO output, , , These represent the TEO operations at times n, n+1, and n-1, respectively. These are the sampling points.

[0091] The TEO operation formula is:

[0092] Equation (4)

[0093] The first abrupt change is the moment when the fault wavefront arrives at the measurement point.

[0094] This embodiment uses SDEO, which combines the traditional Teager Energy Operator (TEO) with a sliding weighted averaging mechanism to reduce the interference of demodulation error. It also replaces the forward differential structure in TEO with a central differential form, thereby enhancing the sensitivity and capture capability of local signal variation characteristics.

[0095] In one embodiment, step S3, determining the fault interval based on the fault wavefront times of adjacent detection points on the main line, specifically includes the following steps:

[0096] Step S31: Based on the recorded fault wavefront times of each detection point, calculate the projected distance from the fault point to the two detection points for all adjacent detection point pairs on the main line using the double-ended traveling wave method, and combine this with the inherent line distance between the two detection points to perform an initial fault interval determination, obtaining the initial interval determination result for the fault occurrence; specifically, define adjacent detection points on the main line. and The recorded wavehead times are as follows: and Within the range of i=1, 2, ..., m-1, the distance from the fault point to the detection point is calculated using the double-ended traveling wave method. and The projection distances are denoted as follows: and , and The distance between them is denoted as ;

[0097] Step S32: Traverse all adjacent main line detection point pairs, and determine the fault interval based on the relationship between the maximum projected distance from the fault point to the two detection points and the actual line distance between the detection points; specifically, in Within the range, if it exists The fault was determined to occur in Within the interval, the interval includes the main line and its connected branches; otherwise, the fault is detected at the detection point corresponding to the minimum value at the wavefront time. At the location. Among them, express and The maximum value in. The formula for the two-ended traveling wave method described in step S3 is as follows:

[0098]

[0099]

[0100] In the formula, For voltage line mode The wave velocity component on the overhead line is equivalent to the wave velocity vo in step S11. L and C are the positive sequence inductance and positive sequence capacitance parameters of the overhead line.

[0101] In one embodiment, step S4, constructing a fault branch determination matrix based on the distance relationship within the fault interval, and determining the line segment where the fault occurs based on the branch determination matrix and the corresponding determination criteria, specifically includes the following steps:

[0102] Step S41: Select the fault range Select k branch end detection points connected within the interval, considering all branch end detection points within the interval. Construct a two-dimensional decision matrix where rows correspond to main line detection points and columns correspond to branch end detection points. Its structure is as follows:

[0103]

[0104] In the formula, the matrix , indicating 2 lines A matrix of real numbers in columns. The Row corresponding detection point r=1, 2; where r=1 indicates r=2 means .matrix The The column corresponds to the end detection point of the c-th branch. , where c = 1, 2, ..., k. Matrix medium elements The calculation formula is:

[0105]

[0106] In the formula, Indicates the detection point and testing points The two-ended path is formed, and the fault point f is obtained using the two-ended traveling wave formula. The projection distance, Indicates the detection point With the testing point Branch nodes in the interval The inherent line distance between them;

[0107] Step S42: Revise the two-dimensional decision matrix, setting the correction margin μ=0.05, when element At that time, modify the element If the value is 0, the elements remain unchanged; the correction matrix is ​​denoted as . ;

[0108] Step S43: For the correction matrix, use the fault branch determination criteria to determine the line section where the fault occurs:

[0109] Step S431, if the correction matrix All elements in column c are greater than 0, and the elements in the remaining columns are greater than 0. All are not greater than 0, that is If the fault is located in the branch corresponding to the element in column c, then the fault is determined to be located in the branch corresponding to the element in column c. The above indicates a first-level branch line fault; among which, express The minimum value of the element in column c. Indicates the first The maximum value of the column elements.

[0110] Step S432, if the correction matrix The Middle , All column elements are greater than 0, and the remaining column elements are... All are not greater than 0, that is Then the fault is determined to be located in the first position. Branches corresponding to column elements The above belongs to the second-level branch fault; among them, the first The branch road is a first-level branch, the first The branch road is a secondary branch. , They represent The Middle , The minimum value of the column elements.

[0111] Step S433: If the correction matrix All column elements are not greater than 0, that is Then the fault is determined to be located in On the main line of the interval; among them, express Find the maximum value of the element in column c.

[0112] Step S434, if the correction matrix All elements in column c are equal to 0, that is If the fault is located in the branch node corresponding to the element in column c, then the fault is determined to be located in the branch node corresponding to the element in column c. Place; among them, Represents the correction matrix All elements in column c of the array are equal to 0.

[0113] In one embodiment, step S5, determining the location of the fault point by performing a final measurement based on the determined line section where the fault is located using the two-end traveling wave method, specifically includes the following steps:

[0114] Step S51: Based on the differences in fault location, use the corresponding dual-end ranging strategy to obtain the projected distance from the fault point to the measurement point, and unify the calculation results of each path as the equivalent fault distance; specifically including the following steps:

[0115] Step S511: If the fault occurs on a first-level branch or a second-level branch, calculate the detection point at the end of the faulty branch. and In the constructed two-way path, the fault point to The distance, where r = 1, 2, and the average of the two calculations, is defined as the equivalent fault distance. .

[0116] Step S512: If the fault occurs on the main line, calculate the... and The fault point in the constructed two-end path to The distance is denoted as ;

[0117] Step S513: If the fault occurs at a branch node At this point, calculations are performed by The detection point at the end of the branch and The fault point in the constructed two-end path to The distances, where r = 1 and 2, are taken as the average, denoted as . ;

[0118] Step S52: Perform a normalization calculation on the equivalent fault distance to obtain the actual fault distance, and finally determine the location of the fault point.

[0119]

[0120] In the formula, To restore the actual fault distance obtained from the normalization calculation, Equivalent fault distance, This refers to the length of overhead lines included in the fault path. These represent the line mode wave velocities of traveling waves in overhead lines and cable lines, respectively.

[0121] Figure 3 This is a simulation model of the 10kV resonant grounding system described in the embodiments of this application. The system has three feeders, and the specific parameters are shown in Table 1. The system operates in an overcompensated state, with the overcompensation degree set to 5%. The equivalent inductance of the arc suppression coil is L. P =0.654H, series resistance is Each feeder is connected to a constant power load at its end, with active power P = 1.62 MW and reactive power Q = 0.99 Mvar. Feeder 1 is the focus of this study, and its topology is as follows: Figure 4 As shown, the other two feeders are simplified. Figure 4 As shown, the locations of the traveling wave detection devices on the main line are numbered sequentially as follows: ,in The location is the segment point of the primary and secondary fusion switch with traveling wave acquisition function. The locations of the traveling wave detection devices on the branch lines are numbered sequentially as follows: .

[0122] Assume the fault occurs at 0.08s and the system sampling frequency is 10MHz. Set up 5 fault scenarios on feeder 1 topology:

[0123] ① Main line fault ,distance 6km at the end;

[0124] ② First-level branch line Fault ,distance 2km at the end;

[0125] ③ T3 node failure ,distance 6km at the end;

[0126] ④T5N5 branch line ,distance 3km at the end;

[0127] ⑤ Main line failure ,distance 2km from the end.

[0128] Table 1. Specific parameters of the simulation model of the 10kV resonant grounding system

[0129]

[0130] Traveling wave velocity in overhead lines and cables for:

[0131] Equation (5)

[0132] The line was normalized, 5km cable. Equivalent to 13.57km of overhead line, (In the distribution network topology, connecting branch nodes) and The 3km cable segment in the line is equivalent to 8.142km of overhead line.

[0133] by Taking a fault as an example, when When a phase-A ground fault occurs with an initial phase angle of 90° and a grounding resistance of 100Ω, an improved phase-mode transformation is performed on the three-phase voltage traveling waves at each detection point within the 79ms-81ms range to extract the voltage line-mode β component. , Figure 5 Voltage line model for the M2 terminal detection point Components. Decomposition using SVMD method. , Figure 6 The SVMD decomposition diagram of the line-mode voltage at terminal M2 is shown below. It reflects the overall transformation trend of the original signal. To reflect the abrupt changes in the signal, kurtosis is calculated for each mode, and the mode with the largest kurtosis value is selected. The fault wavefront time is determined by using the characteristic mode and performing SDEO calculations.

[0134] Figure 7 (a) is SDEO fault wavefront calibration diagram;

[0135] Figure 7 (b) is SDEO fault wavefront calibration diagram;

[0136] Figure 7 (c) is SDEO fault wavefront calibration diagram;

[0137] Depend on Figure 7 It can be seen that the initial fault wavefront arrives , and The timings of the end detection devices were 41.8 μs, 45.0 μs, and 100.3 μs, respectively.

[0138] Initial measurements are performed using the wavefront times of adjacent detection points on the main line to determine the fault range: the distance from the fault point to the detection point is calculated using the double-ended traveling wave formula. Projection distance =4.033km and to Projection distance =4.967km. Similarly, calculate the distance from the fault point to the detection point. Projection distance =0km and to Projection distance =16.142km. and Line distance between =9km, and Line distance =16.142km. Due to Therefore, the fault is located in Interval.

[0139] Fault branch identification: There are two branch end detection points connected within the interval. , A fault branch determination matrix is ​​further constructed based on the distance relationship within the interval. :

[0140] Equation (6)

[0141] For matrix After making corrections, the correction matrix can be obtained. :

[0142] Equation (7)

[0143] As can be seen from equation (7), the matrix All elements in the first column are greater than 0, and all elements in the second column are not greater than 0. If the fault is located in the branch corresponding to the element in column 1, then the fault is determined to be in the branch corresponding to the element in column 1. The above indicates a fault in the first-level branch line.

[0144] Fault final location: After determining that the fault occurred on the first-level branch line Based on this, calculate separately and , and Two sets of double-ended distance measurements and The average value is calculated and recorded as the equivalent fault distance. , .

[0145] right Perform a normalization calculation to obtain the actual fault distance:

[0146]

[0147] Based on this, the final measurement was completed, and the distance to the fault point was determined. The distance is 1.995km, with an absolute error of 5m. (Regarding...) Figure 3 The feeder 1 topology diagram is shown. Under the condition of a phase A ground fault with an initial phase angle of 90° and a grounding resistance of 100Ω, the location results at different locations are shown in Table 2.

[0148] Table 2 Location results under different fault scenarios

[0149]

[0150] Table 2 shows that when a single-phase ground fault occurs in a resonant grounding system, the method of interval identification and branch determination can accurately determine the fault interval under different fault location scenarios, and achieve high-precision fault point location with an absolute error controlled within 10m.

[0151] Different grounding resistance tests: to verify the grounding resistance The impact on the positioning results, Location, C-phase grounding faults are set at the locations, with the initial phase angle of the fault being... For a 90° angle, Rg was set to 0.01, 10, 100, and 1000Ω, and the positioning results are shown in Table 3.

[0152] Table 3. Location results under different grounding resistance test scenarios.

[0153]

[0154] Different initial phase angle tests: To verify the impact of the initial phase angle δ of the fault on the location results, in Main line distance Location at 11.142km branch distance Two-phase short-circuit faults (A and B) were set at a position 5km from the end of the fault, with δ set at 0°, 30°, 60°, and 90°. The location results are shown in Table 4.

[0155] Table 4. Positioning results under different initial phase angles in test scenarios.

[0156]

[0157] As shown in Table 3, when Rg is between 0.01 and 1000 Ω, The maximum absolute error of the location fault location result is 10m, and the minimum is 0m. The absolute error of the fault location result is 1m. As shown in Table 4, the location results for the same fault point are identical, unaffected by the initial phase angle, and accurate location is still possible even in the event of an interphase fault. In summary, this application can accurately locate faults under different transition resistances, initial phase angles, and different fault types, demonstrating strong adaptability.

[0158] Secondly, this application provides a distribution network fault detection system based on section identification and branch determination, including:

[0159] The component signal acquisition module is used to normalize the length of the distribution network line and uses an improved phase mode transformation to obtain the voltage line mode β component signal from the three-phase voltage signal.

[0160] The fault wavefront timing calibration module is communicatively connected to the component signal acquisition module. It is used to perform successive variational mode decomposition on the voltage line mode β component, select characteristic modes according to kurtosis, and use a symmetric differential energy operator to calibrate the fault wavefront timing.

[0161] The fault interval preliminary determination module is communicatively connected to the fault wavefront time calibration module and is used to determine the fault interval based on the fault wavefront times of adjacent detection points on the main line.

[0162] The fault branch determination module is communicatively connected to the fault section preliminary determination module. It is used to construct a fault branch determination matrix based on the distance relationship within the fault section, and to determine the line section where the fault is located based on the corresponding determination criteria using the branch determination matrix.

[0163] The fault location determination module is communicatively connected to the fault branch determination module. It is used to determine the location of the fault point by performing a final measurement based on the determined line section where the fault is located, using the principle of the double-ended traveling wave method.

[0164] The functions of each module in the above-mentioned distribution network fault detection system based on interval identification and branch determination correspond to the steps in the above-mentioned distribution network fault detection method based on interval identification and branch determination. Their functions and implementation processes will not be described in detail here.

[0165] Thirdly, embodiments of this application provide a power distribution network fault detection device based on interval identification and branch determination. The power distribution network fault detection device based on interval identification and branch determination can be a personal computer (PC), laptop computer, server, or other device with data processing capabilities.

[0166] In this embodiment of the application, the power distribution network fault detection device based on section identification and branch determination may include a processor, a memory, a communication interface, and a communication bus.

[0167] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0168] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces. These interfaces enable interconnection of internal components within the distribution network fault detection equipment based on interval identification and branch determination, and also enable interconnection between the equipment and other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0169] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0170] The processor can be a general-purpose processor, which can call the distribution network fault detection program based on interval identification and branch determination stored in the memory, and execute the distribution network fault detection method based on interval identification and branch determination provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the distribution network fault detection program based on interval identification and branch determination is called can refer to the various embodiments of the distribution network fault detection method based on interval identification and branch determination in this application, and will not be repeated here.

[0171] Fourthly, embodiments of this application also provide a readable storage medium.

[0172] The present application stores a distribution network fault detection program based on interval identification and branch determination on a readable storage medium. When the distribution network fault detection program based on interval identification and branch determination is executed by a processor, it implements the steps of the distribution network fault detection method based on interval identification and branch determination as described above.

[0173] The method implemented when the distribution network fault detection program based on interval identification and branch determination is executed can be referred to in the various embodiments of the distribution network fault detection method based on interval identification and branch determination in this application, and will not be repeated here.

[0174] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for detecting distribution network faults based on section identification and branch determination, characterized in that, Includes the following steps: The normalized distribution network line length is obtained by using an improved phase-mode transformation to extract the voltage line mode β component signal from the three-phase voltage signal; Successive variational mode decomposition is performed on the voltage line mode β component, characteristic modes are selected based on kurtosis, and the fault wavefront timing is calibrated using a symmetric differential energy operator; The fault interval is determined based on the fault wavefront timing of adjacent detection points on the main line. A fault branch determination matrix is ​​constructed based on the distance relationships within the fault interval, and the line segment where the fault is located is determined based on the corresponding determination criteria using the branch determination matrix. The specific steps include: Select the fault interval M i -M i+1 For all branch end detection points, construct a two-dimensional decision matrix Q that corresponds to the main line detection points in rows and the branch end detection points in columns. i Its structure is as follows: In the formula, the two-dimensional decision matrix Represents a 2xk real number matrix; a two-dimensional decision matrix Q. i The r-th row corresponds to the detection point M i+r-1 r = 1, 2; Two-dimensional decision matrix Q i The i-th column corresponds to the c-th branch end detection point N ic Where c = 1, 2, ..., k; and the two-dimensional decision matrix Q. i medium element Q i The formula for calculating (r,c) is: In the formula, l Mi+r-1,Nic Indicates detection point M i+r-1 and detection point N ic The two-ended path is formed, and the fault point f to M is obtained using the two-ended traveling wave formula. i+r-1 The projection distance, d Mi+r-1,Tic Indicates detection point M i+r-1 With detection point N ic Branch node T in the interval ic The inherent line distance between them; The two-dimensional decision matrix is ​​revised to obtain the correction matrix; specifically, when element Q... i When (r,c)∈[-μ,μ], the correction element Q i (r,c) is 0, otherwise element Q i (r,c) remain unchanged, and the correction matrix is ​​denoted as Q. i '; where μ is the correction margin; For the correction matrix, the fault branch determination criterion is used to determine the line section where the fault is located; Based on the identified fault section, the final measurement is performed using the double-ended traveling wave method to determine the location of the fault.

2. The distribution network fault detection method based on section identification and branch determination according to claim 1, characterized in that, The normalized distribution network line length is obtained by using an improved phase-mode transformation to acquire the voltage line mode β component signal from the three-phase voltage signal, specifically including the following steps: The distribution network lines are normalized, and the cable length in the equivalent distribution network is the same as the overhead line length: In the formula, L eq L is the equivalent length of the cable. cable v is the original length of the cable o v c These are the line mode wave velocities of traveling waves in overhead lines and cable lines, respectively. The district distribution network consists of a main line and several branch lines. Based on the normalized cable length, traveling wave detection devices are deployed at the ends of branch lines whose length exceeds a set threshold. When a fault occurs in the distribution network, the three-phase voltage traveling wave signals collected by each traveling wave detection device are improved by phase mode transformation, and the voltage line mode β component is extracted from them.

3. The distribution network fault detection method based on section identification and branch determination according to claim 1, characterized in that, The process of performing successive variational mode decomposition on the voltage line mode β component, selecting characteristic modes based on kurtosis, and calibrating the fault wavefront timing using a symmetric differential energy operator specifically includes the following steps: Successive variational mode decomposition is performed on the voltage line β component signal to obtain multiple intrinsic mode function components; Calculate the kurtosis value of each intrinsic mode function component, and select the intrinsic mode function component corresponding to the maximum kurtosis value as the characteristic mode representing the fault mutation information; The characteristic modes are calculated using a symmetric differential energy operator to determine the wavefront time of the fault traveling wave arriving at the measurement point.

4. The distribution network fault detection method based on section identification and branch determination according to claim 1, characterized in that, The method of determining the fault interval based on the fault wavefront times of adjacent detection points on the main line specifically includes the following steps: Based on the recorded fault wave head times of each detection point, for all adjacent detection point pairs on the main line, the projected distance from the fault point to the two detection points is calculated using the double-ended traveling wave method, and the fault interval is initially measured in combination with the inherent line distance between the two detection points to obtain the initial interval determination result of the fault occurrence. Traverse all adjacent main line detection point pairs, and determine the fault range based on the relationship between the maximum value of the projected distance from the fault point to the two detection points and the actual line distance between the detection points.

5. The distribution network fault detection method based on section identification and branch determination according to claim 1, characterized in that, The process of determining the faulty line segment using the fault branch determination criterion for the modified matrix includes the following steps: If all elements in a certain column of the correction matrix are greater than zero, and all elements in the other columns are not greater than zero, then the fault is determined to be located on the first-level branch line corresponding to that column. If there are two columns in the correction matrix where all elements are greater than zero and all other columns where all elements are not greater than zero, then the fault is determined to be located on the secondary branch line corresponding to the second column. If all columns of the correction matrix are not greater than zero, then the fault is determined to be located within the current main line interval. If all elements in a column of the correction matrix are equal to zero, then the fault is determined to be located at the branch node of the corresponding branch line in that column.

6. The distribution network fault detection method based on section identification and branch determination according to claim 1, characterized in that, The process of determining the location of the fault point by performing a final measurement using the two-end traveling wave method based on the identified fault section of the line includes the following steps: Based on the differences in fault location, a corresponding dual-end ranging strategy is adopted to obtain the projected distance from the fault point to the measurement point, and the calculation results of each path are unified as the equivalent fault distance. The equivalent fault distance is restored and normalized to obtain the actual fault distance, and the location of the fault point is finally determined.

7. The distribution network fault detection method based on section identification and branch determination according to claim 6, characterized in that, The step of obtaining the projected distance from the fault point to the measurement point using a corresponding dual-end ranging strategy based on the difference in fault location, and unifying the calculation results of each path as the equivalent fault distance, specifically includes the following steps: When the fault is determined to be located on a primary or secondary branch line, the detection point at the end of the fault branch and two adjacent measurement points on the main line are selected to form a double-end ranging path. The projected distance from the fault point to the measurement point on the main line is calculated, and the average of the two calculation results is taken as the equivalent fault distance. When the fault is determined to be located in the main line section, the projected distance from the fault point to the upstream measurement point is calculated directly using the double-ended distance measurement path formed by two adjacent measurement points on the main line, and the calculation result is directly used as the equivalent fault distance. When the fault is determined to be located at a branch node, the end detection point of the branch where the node is located and two adjacent measurement points on the main line are selected to form a double-end distance measurement path. The projected distance from the fault point to the measurement point on the main line is calculated, and the average of the two calculation results is taken as the equivalent fault distance.

8. A distribution network fault detection system based on section identification and branch determination, characterized in that, The system is used to implement the distribution network fault detection method based on section identification and branch determination as described in any one of claims 1-7; the system includes: The component signal acquisition module is used to normalize the length of the distribution network line and uses an improved phase mode transformation to obtain the voltage line mode β component signal from the three-phase voltage signal. The fault wavefront timing calibration module is communicatively connected to the component signal acquisition module. It is used to perform successive variational mode decomposition on the voltage line mode β component, select characteristic modes according to kurtosis, and use a symmetric differential energy operator to calibrate the fault wavefront timing. The fault interval preliminary determination module is communicatively connected to the fault wavefront time calibration module and is used to determine the fault interval based on the fault wavefront times of adjacent detection points on the main line. The fault branch determination module is communicatively connected to the fault section preliminary determination module. It is used to construct a fault branch determination matrix based on the distance relationship within the fault section, and to determine the line section where the fault is located based on the corresponding determination criteria using the branch determination matrix. The fault location determination module is communicatively connected to the fault branch determination module. It is used to determine the location of the fault point by performing a final measurement based on the determined line section where the fault is located, using the principle of the double-ended traveling wave method.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a distribution network fault detection program based on interval identification and branch determination, wherein when the distribution network fault detection program based on interval identification and branch determination is executed by a processor, it implements the steps of the distribution network fault detection method based on interval identification and branch determination as described in any one of claims 1 to 7.

Images