CN112782523A - Dynamic pattern matching distance-based single-phase earth fault line selection method for power distribution network - Google Patents

Dynamic pattern matching distance-based single-phase earth fault line selection method for power distribution network Download PDF

Info

Publication number
CN112782523A
CN112782523A CN202011537903.5A CN202011537903A CN112782523A CN 112782523 A CN112782523 A CN 112782523A CN 202011537903 A CN202011537903 A CN 202011537903A CN 112782523 A CN112782523 A CN 112782523A
Authority
CN
China
Prior art keywords
line
fault
sequence
matrix
dynamic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202011537903.5A
Other languages
Chinese (zh)
Other versions
CN112782523B (en
Inventor
倪良华
吴春阳
倪诚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Institute of Technology
Original Assignee
Nanjing Institute of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing Institute of Technology filed Critical Nanjing Institute of Technology
Priority to CN202011537903.5A priority Critical patent/CN112782523B/en
Publication of CN112782523A publication Critical patent/CN112782523A/en
Application granted granted Critical
Publication of CN112782523B publication Critical patent/CN112782523B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/08—Locating faults in cables, transmission lines, or networks
    • G01R31/088—Aspects of digital computing
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
    • Y04S10/52—Outage or fault management, e.g. fault detection or location

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Locating Faults (AREA)

Abstract

The invention belongs to the field of power distribution network fault identification and safe operation, particularly relates to a power distribution network single-phase earth fault line selection method, and particularly relates to a power distribution network single-phase earth fault line selection method based on a dynamic pattern matching distance. Firstly, dynamically segmenting transient zero-sequence current of a single-phase grounded line by taking equal segmentation points and extreme points as endpoints, realizing symbolic expression of a sub-segmentation sequence form mode according to each segmentation mean value and the variation trend thereof, defining a mode matching distance by using the heterogeneity among the modes, then obtaining a dynamic mode matching distance matrix among zero-sequence current waveforms of the line by using a Dynamic Time Warping (DTW) planning principle, and finally realizing single-phase grounded line selection by using a fuzzy C mean value clustering algorithm. The result shows that the method can effectively improve the accuracy and fault tolerance of fault line selection of the power distribution network, and has certain practical significance on safe operation of the power distribution network.

Description

Dynamic pattern matching distance-based single-phase earth fault line selection method for power distribution network
Technical Field
The invention belongs to the field of power distribution network fault identification and safe operation, particularly relates to a power distribution network single-phase earth fault line selection method, and particularly relates to a power distribution network single-phase earth fault line selection method based on a dynamic pattern matching distance.
Background
The resonant grounding is an important grounding operation mode in a low-voltage power grid in China, the single-phase grounding short-circuit can compensate capacitance current in the single-phase grounding short-circuit process, the circuit is allowed to operate for 1-2h after a fault, but the non-fault phase voltage is increased and operates for a long time, so that the circuit is easily damaged by insulation, the interphase short circuit is caused, the fault is enlarged, and therefore the quick judgment and removal of the fault circuit have important significance for the safe operation of the power distribution network.
The existing line selection method for single-phase earth faults is divided into a steady-state method and a transient-state method. The steady-state signal is easily influenced by factors such as the size of a transition resistor, an operation mode, the length of an outgoing line, the unbalanced current of a mutual inductor and the like, and the transient-state signal is not influenced by an arc suppression coil and has high sensitivity, so that the line selection by utilizing the transient-state signal is a research hotspot in the field in recent years. Many line selection characteristic quantities are obtained by extracting the characteristic frequency bands of the transient quantities, but the line selection may fail due to different time windows selected for different types of faults; by utilizing wavelet transformation and taking energy and polarity as criteria, if engineering problems such as reverse polarity connection or asynchronous sampling of a zero-sequence current transformer occur, line selection failure can be caused; the fault line selection is carried out by using the morphological peak-valley characteristics of the transient zero-sequence current, when small-angle faults occur, the transient information is not obvious, and the reliability of the method is not high.
Disclosure of Invention
1. The technical problem to be solved is as follows:
the existing line selection method for the single-phase earth fault has the possibility of causing line selection failure due to different time windows selected by different types of faults.
2. The technical scheme is as follows:
the invention provides a power distribution network single-phase earth fault line selection method for identifying line zero-sequence current waveform characteristics by using a dynamic mode matching distance; the method comprises the steps of dynamically segmenting transient zero-sequence current of a line after single-phase grounding by taking equal segmentation points and extreme points as endpoints, realizing symbolic expression of a sub-segmentation sequence form mode according to each segmentation mean value and the variation trend thereof, defining mode matching distance by using the heterogeneity among the modes, obtaining a dynamic mode matching distance matrix among zero-sequence current waveforms of the line by using a Dynamic Time Warping (DTW) planning principle, and finally realizing single-phase grounding line selection by using a fuzzy C mean value clustering algorithm. The method can improve the effect of transient zero-sequence current waveform shape depiction, reduce the time complexity of the traditional distance measurement method and improve the accuracy of fault line selection.
The invention provides a dynamic pattern matching distance-based power distribution network single-phase earth fault line selection method, which specifically comprises the following steps:
step 1: according to the network structure of the power distribution network, setting a single-phase earth fault of a line, acquiring zero-sequence current data under different sampling frequencies in a simulation mode, and determining the number k of each sub-sequence data and the number b of intervals equally divided by peak and valley values after a sampling sequence is segmented equally;
step 2: when the zero sequence voltage of the bus is greater than a set threshold value, starting a line selection protection device, and acquiring transient zero sequence current data of a line with a fault in real time, wherein the line comprises a fault line and a non-fault line;
and step 3: processing transient zero-sequence current data of each line;
and 4, step 4: according to the solving process according to the dynamic time bending distance, seeking an optimal path of a form mode matrix M through a dynamic programming method, and solving a form mode sequence accumulated distance matrix R, wherein the optimal path is a path which is found out of the form mode matrix M and is dynamically bent, so that the sum of all elements on the path is minimum;
and 5: defining the final accumulated distance d as r (m, n) as the dynamic pattern matching distance of the time series X and Y;
step 6: traversing n lines to form a dynamic mode matching distance between every two lines to obtain a dynamic mode matching matrix D:
Figure RE-GDA0003005552940000021
in the formula dijRepresents the dynamic pattern matching distance of the ith and jth lines, i ∈ [1, n ∈],j∈[1,n]；
Carrying out fuzzy C-means clustering analysis on the matrix D, and realizing single-phase earth fault line selection by using a membership matrix, wherein a target function of the fuzzy C-means clustering and the membership matrix are as follows:
Figure RE-GDA0003005552940000022
wherein M is fuzzy weighted index, M is 2, n is total number of lines, djFor the dynamic pattern matching distance of the jth line to all lines, dj＝[dj1,dj2,…,djn]；hiRepresenting the cluster center vector, hi＝[h1,h2](ii) a J denotes an objective function, U denotes a membership matrix, H denotes a cluster center, U denotes a cluster centerijRepresenting the membership degree of the jth sample belonging to the ith class;
after the objective function J converges, a membership degree matrix U and a clustering center H, U are obtainedijThe following conditions need to be satisfied:
Figure RE-GDA0003005552940000023
further obtaining a membership matrix U:
Figure RE-GDA0003005552940000024
3. has the advantages that:
after the equal segmentation points and the extreme points are inserted, the method not only reduces the dimensionality of the time sequence, but also can filter noise and simultaneously reserve the integral trend information of the time sequence to a great extent. The time sequence morphology is modeled, and the average value information of each segment is combined in the traditional ternary fluctuation mode, so that the local morphological characteristics of the time sequence can be carefully described, and the overall trend of the time sequence can be grasped. And finally, measuring the heterogeneity of the sequence form and the mode by using the mode matching distance, thereby effectively avoiding the influence of local mutation and abnormal conditions on the overall similarity of the sequence.
Drawings
FIG. 1 is a diagram of the fluctuation pattern of the time series of the present invention.
FIG. 2 is a dynamic time warping distance path diagram according to the present invention.
FIG. 3 is a flow chart of fuzzy C-means clustering in accordance with the present invention.
Fig. 4 is a simplified simulation model diagram of a distribution network according to the present invention.
Fig. 5 is a waveform diagram of a transient zero-sequence current according to an embodiment of the present invention.
FIG. 6 is a flow chart for determining the k and b values according to the present invention.
FIG. 7 is a k-f and b-f relationship diagram according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings.
The method for selecting the single-phase earth fault line of the power distribution network based on the dynamic mode matching distance comprises the following steps:
step 1: and (4) according to the network structure of the power distribution network, setting a single-phase earth fault of the line, simulating to obtain zero-sequence current data under different sampling frequencies, and determining the number k of each sub-sequence data and the number b of peak-valley equally-divided intervals after the sampling sequence is equally segmented according to the method in the step 7.
Step 2: when the bus zero sequence voltage is larger than a set threshold value, the line selection protection device is started to collect transient zero sequence current data of a line with a fault in real time, wherein the line comprises a fault line and a non-fault line.
And step 3: the method for processing the transient zero-sequence current data of each line specifically comprises the following steps:
the method comprises the following steps: firstly, carrying out normalization processing on data, and uniformly mapping the data between [0,1 ]; the normalization processing method is shown as the formula (1):
Figure RE-GDA0003005552940000031
in the formula (1), x represents data before normalization, and x*Represents the normalized data;
step two: dynamically segmenting the normalized data by taking equal segmentation points and extreme points as end points;
step three: equally dividing the peak-to-valley value of the normalized time series into b intervals as shown in table 1;
as can be seen from fig. 1, the fluctuation mode is simply used as the time series variation trend, and although the local state can be effectively represented, the overall state is difficult to accurately reflect, so that the fluctuation mode can be more refined by combining the mean information of each segment, and the overall trend variation of the time series can be better depicted.
Step four: calculating the variation trend and the mean value of each sub-segment; the change trend comprises rising (the slope is greater than 0), keeping unchanged (the slope is 0) and falling (the slope is less than 0).
Step five: changing the sampling sequence into a corresponding form mode sequence according to the change trend and the mean value interval of each sub-segment by referring to the table 1;
TABLE 1 morphological mode notation
Figure RE-GDA0003005552940000041
X in Table 1minAnd xmaxEach represents a valley and a peak of the normalized time series, and c ═ xmax-xmin)/b，
Figure RE-GDA0003005552940000042
Is the average value of all sampling points of the ith sub-sequence.
Step six: defining a pattern matching distance dp (X) in the dissimilarity of morphological patterns for two morphological pattern sequences X and Yi,yj) The value is as formula (2):
Figure RE-GDA0003005552940000043
in the formula xiAnd yjForm pattern elements representing sequences X and Y, respectively;
traversing all the form mode elements of the time sequence X and the time sequence Y to obtain a form mode matrix M as follows:
Figure RE-GDA0003005552940000044
in the formula, m and n represent the number of form pattern elements included in the form pattern sequence X, Y, respectively.
And 4, step 4: according to the solving process of the dynamic time bending distance, an optimal path of a form mode matrix M is sought through a dynamic programming method, and a form mode sequence accumulated distance matrix R is obtained, wherein a path schematic diagram is shown in figure 2, the optimal path refers to finding a dynamic bending path in the form mode matrix M, and all elements on the path are minimum;
Figure RE-GDA0003005552940000045
Figure RE-GDA0003005552940000046
where r (i, j) is the cumulative distance, r (0,0) is 0, r (j,0) is r (0, i) is infinity, and the effective path satisfies three constraints:
(1) the characteristics of the bounding: max (m, n) is less than or equal to S and less than or equal to m + n-1, and S is the total step number of walking;
(2) and (3) boundary limitation: the starting point is (1,1) and the end point is (m, n);
(3) continuity: the occurrence of local jumps is not allowed.
And 5: defining the final accumulated distance d as r (m, n) as the dynamic pattern matching distance of the time series X and Y.
Step 6: traversing n lines to form a dynamic mode matching distance between every two lines to obtain a dynamic mode matching matrix D:
Figure RE-GDA0003005552940000051
in the formula dijDynamic pattern matching for representing ith and jth linesDistance, i ∈ [1, n ]],j∈[1,n]。
Carrying out fuzzy C-means clustering analysis on the matrix D, and realizing single-phase earth fault line selection by using a membership matrix, wherein a target function of the fuzzy C-means clustering and the membership matrix are as follows:
Figure RE-GDA0003005552940000052
where M is fuzzy weighted index, generally, M is 2, n is total number of lines, d isjFor the dynamic pattern matching distance of the jth line to all lines, dj＝[dj1,dj2,…,djn]；hiRepresenting the cluster center vector, hi＝[h1,h2](ii) a J denotes an objective function, U denotes a membership matrix, H denotes a cluster center, U denotes a cluster centerijIndicating the degree of membership that the jth sample belongs to the ith class.
After the objective function J converges, a membership degree matrix U and a clustering center H, U are obtainedijThe following conditions need to be satisfied:
Figure RE-GDA0003005552940000053
further obtaining a membership matrix U:
Figure RE-GDA0003005552940000054
the flow of fuzzy C-means clustering is shown in FIG. 3;
step 7) a method for determining a k value and a b value, wherein the specific flow is shown in fig. 6, and the method is characterized by comprising the following steps:
the dynamic pattern matching distance between the lines is related to k and b, and is regarded as a function of k and b;
defining the mean dynamic pattern matching distance D between the faulty line and the non-faulty linegAverage dynamic pattern matching distance D between (k, b) and non-faulty linef(k, b), the expression of which is as follows:
Figure RE-GDA0003005552940000061
dijthe meaning is the same as formula (6), A is a fault line set, B is a non-fault line set, n is the number of feeder lines, p is the number of fault lines,
defining a fault discrimination Q (k, b), wherein the expression is as follows:
Figure RE-GDA0003005552940000062
the first step is as follows: setting a line fault for the determined network structure, and simulating to obtain zero-sequence current data under different sampling frequencies f;
the second step is that: selecting data of a certain specific sampling frequency as a sample;
the third step: setting an iteration initial value k equal to 2, b equal to 2, the step length is 1, the threshold value k is less than or equal to 20, and b is less than or equal to 10;
the fourth step: calculating positive integers k and b when the fault discrimination Q (k, b) is maximized by using a genetic algorithm;
the fifth step: changing the sampling frequency and jumping to the second step;
sixthly, changing a fault line and jumping to the first step;
the seventh step: and (5) drawing a k-f, b-f relation curve, and finding k and b for a specific sampling frequency comparison curve.
This invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, which are provided for a complete and complete disclosure of the invention and to convey the scope of the invention to those skilled in the art.
The validity of the method is checked in the following with a specific fault case. The specific process is as follows:
firstly, a distribution network simulation model diagram shown in the attached figure 4 is built, wherein specific parameters of a cable line and an overhead line are shown in a table 2:
TABLE 2 line parameters
Overhead line parameters
Figure RE-GDA0003005552940000063
Parameters of cable line
Figure RE-GDA0003005552940000064
The compensation degree of the arc suppression coil is set to be 10%, A-phase metal grounding fault occurs at a position 8km away from a bus in a circuit L2, the initial fault angle is 30 degrees, the fault time is 0.30167s, the fault duration is 0.5s, and the simulation duration is 1 s. Transient zero-sequence current of each line is sampled by a current transformer, the sampling frequency is 20kHz, according to a relation curve of k-f and b-f in the attached figure 7, k is 6, and b is 4.
And then collecting data of two periods of transient zero-sequence current after the fault starts, wherein the waveform of the transient zero-sequence current is as shown in figure 5, carrying out normalization processing on the data, dividing every 6 data into a segment, and inserting an extreme point to obtain a new sub-segment. The table look-up 1 equally divides the peak-to-valley value of the normalized time series into 4 intervals, and represents the time series by 12 morphological pattern symbols.
And calculating the variation trend and the mean value of each sub-segment, and changing the segment sequence into a morphological mode sequence according to the morphological mode symbol.
Calculating the mode matching distance of the two lines, calculating the dynamic mode matching distance between the two lines according to the solving process of the dynamic time bending distance of the form mode matrix M, traversing each line to obtain a dynamic mode matching distance matrix D:
Figure RE-GDA0003005552940000071
carrying out fuzzy C-means clustering on the obtained dynamic mode matching distance matrix D, taking M as 2, setting the iteration frequency to be 50 times at most, and setting the iteration convergence criterion to be 10-6After 5 iterations, the objective function J (U, H) 139.0864, membership matrix U:
Figure RE-GDA0003005552940000072
the rows of the membership degree matrix represent categories, the columns 1 to 6 represent lines L1 to L6, the row where the maximum element of each column is located is the category to which the line belongs, and the membership degree matrix can obviously judge that a fault line L2 and a non-fault line belong to different categories, so that the fault of the line L2 is judged.
In order to reflect the adaptability of the algorithm, taking the network structure in the case as an example, several different situations are simulated, and the line selection result is shown as follows:
TABLE 3 results of line selection
Figure RE-GDA0003005552940000073
Figure RE-GDA0003005552940000081
And (4) conclusion: according to the case line selection result, reliable line selection can be effectively carried out on the single-phase earth fault of the power distribution network by adopting the dynamic mode matching distance method, and the method has good adaptability and fault tolerance on high-resistance earth, noise interference, two-point earth, arc fault and the like.
Although the present invention has been described with reference to the preferred embodiments, it should be understood that various changes and modifications can be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (4)

1. A power distribution network single-phase earth fault line selection method based on dynamic pattern matching distance comprises the following steps:
step 1: according to the network structure of the power distribution network, setting a single-phase earth fault of a line, acquiring zero-sequence current data under different sampling frequencies in a simulation mode, and determining the number k of each sub-sequence data and the number b of intervals equally divided by peak and valley values after a sampling sequence is segmented equally;
step 2: when the zero sequence voltage of the bus is greater than a set threshold value, starting a line selection protection device, and acquiring transient zero sequence current data of a line with a fault in real time, wherein the line comprises a fault line and a non-fault line;
and step 3: processing transient zero-sequence current data of each line;
and 4, step 4: according to the solving process according to the dynamic time bending distance, seeking an optimal path of a form mode matrix M through a dynamic programming method, and solving a form mode sequence accumulated distance matrix R, wherein the optimal path is a path which is found out of the form mode matrix M and is dynamically bent, so that the sum of all elements on the path is minimum;
and 5: defining the final accumulated distance d as r (m, n) as the dynamic pattern matching distance of the time series X and Y;
step 6: traversing n lines to form a dynamic mode matching distance between every two lines to obtain a dynamic mode matching matrix D:
Figure FDA0002854074080000011
in the formula dijRepresents the dynamic pattern matching distance of the ith and jth lines, i ∈ [1, n ∈],j∈[1,n]；
Carrying out fuzzy C-means clustering analysis on the matrix D, and realizing single-phase earth fault line selection by using a membership matrix, wherein a target function of the fuzzy C-means clustering and the membership matrix are as follows:
Figure FDA0002854074080000012
wherein M is fuzzy weighted index, M is 2, n is total number of lines, djFor the dynamic pattern matching distance of the jth line to all lines, dj＝[dj1,dj2,…,djn]；hiRepresenting the cluster center vector, hi＝[h1,h2](ii) a J denotes an objective function, U denotes a membership matrix, H denotes a cluster center, U denotes a cluster centerijRepresenting the membership degree of the jth sample belonging to the ith class;
after the objective function J converges, a membership degree matrix U and a clustering center H, U are obtainedijThe following conditions need to be satisfied:
Figure FDA0002854074080000013
further obtaining a membership matrix U:
Figure FDA0002854074080000014
2. the method of claim 1, wherein: in step 1, a method of determining a value of k and a value of b, comprising the steps of: the dynamic pattern matching distance between the lines is related to k and b, and is regarded as a function of k and b;
defining the mean dynamic pattern matching distance D between the faulty line and the non-faulty linegAverage dynamic pattern matching distance D between (k, b) and non-faulty linef(k, b), the expression of which is as follows:
Figure FDA0002854074080000021
dijthe meaning is the same as formula (6), A is a fault line set, B is a non-fault line set, n is the number of feeder lines, and p is the number of fault lines;
defining a fault discrimination Q (k, b), wherein the expression is as follows:
Figure FDA0002854074080000022
the first step is as follows: setting a line fault for the determined network structure, and simulating to obtain zero-sequence current data under different sampling frequencies f;
the second step is that: selecting data of a certain specific sampling frequency as a sample;
the third step: setting an iteration initial value k equal to 2, b equal to 2, the step length is 1, the threshold value k is less than or equal to 20, and b is less than or equal to 10;
the fourth step: calculating positive integers k and b when the fault discrimination Q (k, b) is maximized by using a genetic algorithm;
the fifth step: changing the sampling frequency and jumping to the second step;
and a sixth step: changing a fault line and jumping to the first step;
the seventh step: and (5) drawing a k-f, b-f relation curve, and finding k and b for a specific sampling frequency comparison curve.
3. The method of claim 1, wherein: in the third step, the transient zero sequence current data of each line is processed, and the method specifically comprises the following steps:
the method comprises the following steps: firstly, carrying out normalization processing on data, and uniformly mapping the data between [0,1 ]; the normalization processing method is shown as the formula (1):
Figure FDA0002854074080000023
in the formula (1), x represents data before normalization, and x*Represents the normalized data;
step two: dynamically segmenting the normalized data by taking equal segmentation points and extreme points as end points;
step three: the peak-valley value of the normalized time series is taken as a boundary and equally divided into b intervals,
step four: calculating the variation trend and the mean value of each sub-segment; the trend of change comprises rising, and when the slope is greater than 0; remains unchanged when the slope is 0; decreasing when the slope is less than 0;
step five: changing the sampling sequence into a corresponding form mode sequence according to the change trend and the mean value interval of each sub-segment by referring to the table 1;
step six: defining a pattern matching distance dp (X) in the dissimilarity of morphological patterns for two morphological pattern sequences X and Yi,yj) The value is as formula (2):
Figure FDA0002854074080000031
in the formula xiAnd yjForm pattern elements representing sequences X and Y, respectively;
step 37) traversing all the form mode elements of the time sequence X and the time sequence Y to obtain a form mode matrix M as follows:
Figure FDA0002854074080000032
in the formula, m and n represent the number of form pattern elements included in the form pattern sequence X, Y, respectively.
4. The method of claim 1, wherein: in step 4, the distance matrix R is:
Figure FDA0002854074080000033
Figure FDA0002854074080000034
where r (i, j) is the cumulative distance, r (0,0) is 0, r (j,0) is r (0, i) is infinity, and the effective path satisfies three constraints:
where r (i, j) is the cumulative distance, r (0,0) is 0, r (j,0) is r (0, i) is infinity, and the effective path satisfies three constraints: the characteristics of the bounding: max (m, n) is less than or equal to S and less than or equal to m + n-1, and S is the total step number of walking; and (3) boundary limitation: the starting point is (1,1) and the end point is (m, n); continuity: the occurrence of local jumps is not allowed.
CN202011537903.5A 2020-12-23 2020-12-23 Power distribution network single-phase earth fault line selection method based on dynamic mode matching distance Active CN112782523B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011537903.5A CN112782523B (en) 2020-12-23 2020-12-23 Power distribution network single-phase earth fault line selection method based on dynamic mode matching distance

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011537903.5A CN112782523B (en) 2020-12-23 2020-12-23 Power distribution network single-phase earth fault line selection method based on dynamic mode matching distance

Publications (2)

Publication Number Publication Date
CN112782523A true CN112782523A (en) 2021-05-11
CN112782523B CN112782523B (en) 2022-12-02

Family

ID=75751914

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011537903.5A Active CN112782523B (en) 2020-12-23 2020-12-23 Power distribution network single-phase earth fault line selection method based on dynamic mode matching distance

Country Status (1)

Country Link
CN (1) CN112782523B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113723201A (en) * 2021-08-03 2021-11-30 三明学院 Method, device and system for identifying time series local trend and storage medium
CN114779016A (en) * 2022-05-07 2022-07-22 四川大学 Repetitive transient fault identification method based on improved DTW
CN117332619A (en) * 2023-11-30 2024-01-02 华南理工大学 Method and system for establishing arc root jump model based on field path coupling
CN117368799A (en) * 2023-12-07 2024-01-09 山西思极科技有限公司 Diagnosis method for short-circuit fault of power transmission line of power system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103941163A (en) * 2014-05-12 2014-07-23 福州大学 Resonant earthed system fault line selection method utilizing fuzzy K-means clustering
CN108181547A (en) * 2017-12-20 2018-06-19 珠海许继电气有限公司 A kind of dynamic time warping distance fault section location method based on Time Series Compression
CN108896875A (en) * 2018-07-16 2018-11-27 国网福建晋江市供电有限公司 A kind of fault line selection method for single-phase-to-ground fault and device
CN109307824A (en) * 2018-09-18 2019-02-05 福州大学 A kind of one-phase earthing failure in electric distribution network Section Location based on cluster
CN111751671A (en) * 2020-06-29 2020-10-09 三峡大学 VMD-DTW cluster-based low-current grounding system fault line selection method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103941163A (en) * 2014-05-12 2014-07-23 福州大学 Resonant earthed system fault line selection method utilizing fuzzy K-means clustering
CN108181547A (en) * 2017-12-20 2018-06-19 珠海许继电气有限公司 A kind of dynamic time warping distance fault section location method based on Time Series Compression
CN108896875A (en) * 2018-07-16 2018-11-27 国网福建晋江市供电有限公司 A kind of fault line selection method for single-phase-to-ground fault and device
CN109307824A (en) * 2018-09-18 2019-02-05 福州大学 A kind of one-phase earthing failure in electric distribution network Section Location based on cluster
CN111751671A (en) * 2020-06-29 2020-10-09 三峡大学 VMD-DTW cluster-based low-current grounding system fault line selection method

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
刘鹏辉: "基于动态时间弯曲距离的小电流接地故障区段定位方法", 《电网技术》 *
李海林: "基于分段聚合时间弯曲距离的时间序列挖掘", 《山东大学学报(工学版)》 *
李海林: "基于数值符号和形态特征的时间序列相似性度量方法", 《控制与决策》 *
王瑞 等: "基于形态模式的时间序列相似性度量算法", 《计算机应用与软件》 *
邵翔 等: "基于改进 DTW 的接地故障波形互相关度聚类选线方法", 《电力自动化设备》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113723201A (en) * 2021-08-03 2021-11-30 三明学院 Method, device and system for identifying time series local trend and storage medium
CN114779016A (en) * 2022-05-07 2022-07-22 四川大学 Repetitive transient fault identification method based on improved DTW
CN117332619A (en) * 2023-11-30 2024-01-02 华南理工大学 Method and system for establishing arc root jump model based on field path coupling
CN117332619B (en) * 2023-11-30 2024-02-20 华南理工大学 Method and system for establishing arc root jump model based on field path coupling
CN117368799A (en) * 2023-12-07 2024-01-09 山西思极科技有限公司 Diagnosis method for short-circuit fault of power transmission line of power system
CN117368799B (en) * 2023-12-07 2024-02-23 山西思极科技有限公司 Diagnosis method for short-circuit fault of power transmission line of power system

Also Published As

Publication number Publication date
CN112782523B (en) 2022-12-02

Similar Documents

Publication Publication Date Title
CN112782523B (en) Power distribution network single-phase earth fault line selection method based on dynamic mode matching distance
CN108957251B (en) Cable joint partial discharge mode identification method
CN109828184B (en) Voltage sag source identification method based on mutual approximate entropy
CN110596533A (en) Power distribution network single-phase earth fault section positioning method and system
CN111308272B (en) Positioning method for low-current ground fault section
CN111428755A (en) Non-invasive load monitoring method
CN112580471A (en) Non-invasive load identification method based on AdaBoost feature extraction and RNN model
CN110927527A (en) Transient fault sensing method based on transient energy double channels and information fusion
CN112255500A (en) Power distribution network weak characteristic fault identification method based on transfer learning
CN110244122A (en) Resonant Overvoltage in Power Systems Detection & Controling method
CN112732748A (en) Non-invasive household appliance load identification method based on adaptive feature selection
Song et al. A negative selection algorithm-based identification framework for distribution network faults with high resistance
CN114359674A (en) Non-invasive load identification method based on metric learning
Li et al. A fault pattern and convolutional neural network based single-phase earth fault identification method for distribution network
CN115587329A (en) Power distribution network fault classification method and system based on convolutional neural network extraction features
CN112964959A (en) Power distribution network earth fault line selection method based on correlation coefficient measurement criterion
CN111896890B (en) Micro-grid line fault diagnosis method and system based on Hilbert-Huang transform
CN113010985A (en) Non-invasive load identification method based on parallel AANN
CN115561575A (en) Method for distinguishing electrical abnormal state of offshore wind farm based on multi-dimensional matrix profile
CN113567803B (en) Low-current ground fault positioning method and system based on Tanimoto similarity
CN112986753B (en) Flexible direct-current power grid double-end fault location method grounded through metal return line
CN114755529A (en) Multi-feature fusion single-phase earth fault type identification method based on deep learning
CN114004285A (en) Non-invasive load identification method based on improved kNN algorithm
Zhu et al. Fault Phase Identification Method Based on Convolutional Neural Network for Double Circuit Transmission Lines
Wei et al. Fault location method for active distribution network based on SVM and feature search algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant