CN113625107A - Power distribution network single-phase earth fault line selection method - Google Patents
Power distribution network single-phase earth fault line selection method Download PDFInfo
- Publication number
- CN113625107A CN113625107A CN202110881486.4A CN202110881486A CN113625107A CN 113625107 A CN113625107 A CN 113625107A CN 202110881486 A CN202110881486 A CN 202110881486A CN 113625107 A CN113625107 A CN 113625107A
- Authority
- CN
- China
- Prior art keywords
- line
- power distribution
- distribution network
- zero sequence
- sequence current
- 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.)
- Pending
Links
- 238000010187 selection method Methods 0.000 title claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 40
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000012360 testing method Methods 0.000 claims abstract description 18
- 238000005070 sampling Methods 0.000 claims description 18
- 238000004891 communication Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000010801 machine learning Methods 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000001052 transient effect Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000001629 suppression Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009533 lab test Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
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
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/083—Locating faults in cables, transmission lines, or networks according to type of conductors in cables, e.g. underground
Abstract
The invention discloses a single-phase earth fault line selection method for a power distribution network, which comprises the following steps: s1, measuring zero sequence current of each line of the power distribution network; s2, calculating the mismatching degree and the zero sequence current polarity of each line of the power distribution network according to the zero sequence current of each line of the power distribution network; s3, constructing a training set and a testing set according to the mismatching degree and the zero sequence current polarity of each line of the power distribution network, and training a stack self-encoder through the training set; and S4, inputting a test set into the trained stack self-encoder, and selecting the power distribution network line with the single-phase earth fault. The method makes full use of the physical phenomenon that the zero sequence current waveform of the power distribution network line can change after a single-phase earth fault occurs, has high accuracy, is not influenced by the high-resistance earth phenomenon, and has good applicability, reliability and robustness.
Description
Technical Field
The invention relates to the technical field of power grid power, in particular to a single-phase earth fault line selection method for a power distribution network.
Background
The low-voltage distribution network in China generally adopts a low-current grounding mode, faults of the distribution network mainly account for 80% of total number of fault types by taking single-phase grounding faults as main parts, when the faults occur, fault current signals generated by the power grid are weak, and fault lines are difficult to detect. Therefore, how to improve the accuracy and reliability of fault line selection of the low-current grounding system has important significance on the stable operation of the power distribution network.
At present, the fault line selection method for a low-current grounding system mainly comprises a steady-state method and a transient-state method. When the system has single-phase earth fault, the distribution of current steady-state components can be changed by the existence of the arc suppression coil, so that the effect of fault line selection by adopting a steady-state method is poor. The transient component of the fault current contains rich information, and the transient component is not influenced by the arc suppression coil in a power frequency period after the fault, so that the line selection by utilizing the transient component of the system fault is the direction of the current extensive research.
At present, a fault line selection method based on a transient state method mainly comprises a time frequency analysis method, a similarity analysis method, a 5 th harmonic component analysis method and an energy method. However, due to the fact that high-resistance grounding exists in the power distribution network fault, the identification accuracy of the method needs to be improved.
Disclosure of Invention
Aiming at the defects in the prior art, the single-phase earth fault line selection method for the power distribution network provided by the invention solves the problem that the identification accuracy of the current transient-method-based single-phase earth fault line selection method for the power distribution network is low.
In order to achieve the purpose of the invention, the invention adopts the technical scheme that: a single-phase earth fault line selection method for a power distribution network comprises the following steps:
s1, measuring zero sequence current of each line of the power distribution network;
s2, calculating the mismatching degree and the zero sequence current polarity of each line of the power distribution network according to the zero sequence current of each line of the power distribution network;
s3, constructing a training set and a testing set according to the mismatching degree and the zero sequence current polarity of each line of the power distribution network, and training a stack self-encoder through the training set;
and S4, inputting a test set into the trained stack self-encoder, and selecting the power distribution network line with the single-phase earth fault.
The invention has the beneficial effects that: the physical phenomenon that after a single-phase earth fault of a power distribution network line occurs, the zero sequence current waveform of the power distribution network line can change is utilized, the mismatching degree characteristic and the zero sequence current polarity characteristic of each line reflected by zero sequence current are taken as starting points, the mismatching degree characteristic and the zero sequence current polarity characteristic of each line are analyzed through a stack self-encoder with machine learning capacity, finally, the fault line is identified, the accuracy is high, the influence of the high-resistance earth fault phenomenon is avoided, and the power distribution network line single-phase earth fault identification method has good applicability, reliability and robustness.
Further, in step S2, the method for calculating the degree of mismatch of each line of the power distribution network according to the zero sequence current of each line of the power distribution network includes the following sub-steps:
a1, calculating the equivalent distance of current quantity among all lines of the power distribution network according to the zero sequence current of all the lines of the power distribution network;
and A2, calculating the mismatching degree of each line of the power distribution network according to the equivalent distance of the current amount between each line of the power distribution network.
Further, the calculation expression of the equivalent distance of the current quantities between the lines of the power distribution network in the step a1 includes the following equation:
H(I，J)＝max(h(i，j)，h(j，i))
h(i，j)＝||ia-jb||
h(j，i)＝||ja-ib||
wherein, H (I, J) is the equivalent distance of the current amount between the line I and the line J, max () is the maximum value operation, I is the zero sequence current of the line I, J is the zero sequence current of the line J, H (I, J) is the equivalent distance of the one-way current amount from the line I to the line J, H (J, I) is the equivalent distance of the one-way current amount from the line J to the line I, | | | | is the euclidean distance operation, I | | |aIs the maximum value of zero sequence current I of the line IbIs the minimum value, j, of the zero sequence current I of the line IaIs the maximum value of zero sequence current J of line JbIs the minimum value of the zero sequence current J of the line J.
Further, the calculation expression of the mismatch of each line of the power distribution network in step a2 is as follows:
wherein HIAnd M is the total line value of the power distribution network.
The beneficial effects of the above further scheme are: after a single-phase grounding fault occurs on a certain line of the power distribution network, zero-sequence current of the line is different from other lines of the power distribution network, the unmatched characteristics among the lines are extracted by specifically comparing Euclidean distance models of zero-sequence current values of each line and the other lines, compared with other physical characteristics, the unmatched characteristics are more relevant to fault events and serve as input characteristics of follow-up machine learning, the training effect can be effectively improved, compared with the prior art, the unmatched degree is only related to the current size difference of different lines, whether grounding equivalent resistance is close to 0 or not does not need to be judged, and therefore the influence of a high-resistance grounding phenomenon is avoided.
Further, in step S2, the method for calculating the polarity of the zero sequence current of each line of the power distribution network according to the zero sequence current of each line of the power distribution network includes the following sub-steps:
b1, calculating a zero sequence current average value of each line of the power distribution network according to the zero sequence current of each line of the power distribution network;
and B2, recording the zero sequence current polarity of the power distribution network line with the zero sequence current average value more than or equal to 0 as positive, and recording the zero sequence current polarity of the power distribution network line with the zero sequence current average value less than 0 as negative.
Further, the zero sequence current average value calculation expression of each line of the power distribution network in step B1 is as follows:
wherein I (T) is the value of sampling points of the zero sequence current I of the line I at the time T, T is the sampling duration, N is the total number of the sampling points in the sampling duration T,
and the zero sequence current average value of the line I is obtained.
The beneficial effects of the above further scheme are: the zero sequence current is a dynamically changing value, the measured zero sequence current is obtained in engineering, the real situation of the zero sequence current can be obtained better by keeping a certain sampling time to carry out multi-point sampling than single-point sampling, and in the extraction process of the polarity characteristics of the zero sequence current, the zero sequence current is more reasonable and accurate by taking the positive and negative of the zero sequence current average value as the basis.
Further, the stack self-encoder comprises L self-encoders; the self-encoder is used for carrying out unsupervised learning on characteristics of each line of the power distribution network and comprises an encoder, a hidden layer and a decoder, wherein the input end of the hidden layer is in communication connection with the output end of the encoder, and the input end of the decoder is in communication connection with the output end of the hidden layer.
Further, the step S3 includes the following sub-steps:
s31, constructing a training set and a testing set according to the mismatching degree and the zero sequence current polarity of each line of the power distribution network;
s32, sequentially training and testing 1 st to L self-encoders through a training set to finish the independent training test of L self-encoders of the stack self-encoder;
s33, L self-encoders which are independently trained and completed by the combined stack self-encoder;
and S34, training the L self-coders after combination through a training set to realize network fine adjustment of the stack self-coder and enable the stack self-coder to complete training.
The beneficial effects of the above further scheme are: compared with the conventional training method in the field of machine learning, the method can avoid the phenomenon that parameter oscillation is conducted to another stage when the first stage does not converge, can further save training time, and can also avoid the situation that the stack self-encoder falls into a local optimal solution, so that the stack self-encoder can reach a global optimal solution at a higher speed, and the accuracy of fault line selection is improved.
Further, the step S33 includes the following sub-steps:
c1, shielding the communication connection among the encoder, the hidden layer and the decoder of each of the L self-encoders;
c2, using the input end of the encoder of the 1 st self-encoder as the input end of the stack self-encoder, and the output end of the decoder as the output end of the stack self-encoder;
c3, using the output sequence of the encoder of the 1 st self-encoder as the input sequence of the hidden layer;
c4, taking the output sequence of the hidden layer of the mth self-encoder as the input sequence of the hidden layer of the (m + 1) th self-encoder, wherein m is more than or equal to 1 and less than or equal to L-1;
c5, using the output sequence of the hidden layer of the L-th self-encoder as the input sequence of the decoder;
and C6, taking the output sequence of the decoder of the nth self-encoder as the input sequence of the decoder of the (n-1) th self-encoder, wherein n is more than or equal to 2 and less than or equal to L.
The beneficial effects of the above further scheme are: the designed self-encoder combination mode shields redundant encoders in the combination process, and a similar unstacking and stacking mode is adopted in the stacking process of the hidden layers of the self-encoders and the decoders, so that compared with the simple front-back stacking of the self-encoders, the stability and the continuity of signal propagation can be improved.
Drawings
FIG. 1 is a schematic flow chart of a single-phase earth fault line selection method for a power distribution network;
fig. 2 is a comparison graph of the zero sequence currents of 3 normal lines and 1 fault line in the power distribution network detected by the embodiment of the present invention,
wherein, L1, L3 and L4 are zero sequence currents of three normal lines, and L2 is a zero sequence current of a fault line;
FIG. 3 is a schematic diagram of the combination of two self-encoders of a stacked self-encoder according to an embodiment of the present invention,
where x is the encoder of the 1 st auto-encoder, h1Hidden layer for 1 st self-encoder, h2Is as follows2 hidden layers of the self-encoder, r1Decoder for 1 st self-encoder, r2The decoder of the 2 nd self-encoder.
Detailed Description
The following description of the embodiments of the present invention is provided to facilitate the understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and it will be apparent to those skilled in the art that various changes may be made without departing from the spirit and scope of the invention as defined and defined in the appended claims, and all matters produced by the invention using the inventive concept are protected.
As shown in fig. 1, a method for selecting a single-phase earth fault line of a power distribution network includes the following steps:
and S1, measuring the zero sequence current of each line of the power distribution network.
As shown in fig. 2, in the power distribution network detected in this embodiment, the zero sequence current of the faulty line and the zero sequence current of the normal line have a great difference. The sources of this difference are: the zero sequence current of the sound circuit flows to the circuit from the bus, the polarities of the currents are the same, the zero sequence current of the fault circuit flows to the bus from the circuit, and the current is the sum of the zero sequence currents of all the sound circuits and the current of the arc suppression coil. Therefore, the zero sequence current of the fault line is not matched with the zero sequence current of the normal line, and the zero sequence current of the fault line also presents the characteristic of opposite polarity. The characteristic of the zero sequence current lays a theoretical foundation for the fault line selection method provided by the invention. In order to improve the identification accuracy, further scheme design is needed.
And S2, calculating the mismatching degree and the zero sequence current polarity of each line of the power distribution network according to the zero sequence current of each line of the power distribution network.
In step S2, the method for calculating the degree of mismatch of each line of the power distribution network according to the zero sequence current of each line of the power distribution network includes the following sub-steps:
and A1, calculating the equivalent distance of the current amount among the lines of the power distribution network according to the zero sequence current of each line of the power distribution network. The calculation expression of the equivalent distance of the current amount among the lines of the power distribution network comprises the following equation:
H(I，J)＝max(h(i，j)，h(j，i))
h(i，j)＝||ia-jb||
h(j，i)＝||ja-ib||
wherein, H (I, J) is the equivalent distance of the current amount between the line I and the line J, max () is the maximum value operation, I is the zero sequence current of the line I, J is the zero sequence current of the line J, H (I, J) is the equivalent distance of the one-way current amount from the line I to the line J, H (J, I) is the equivalent distance of the one-way current amount from the line J to the line I, | | | | is the euclidean distance operation, I | | |aIs the maximum value of zero sequence current I of the line IbIs the minimum value, j, of the zero sequence current I of the line IaIs the maximum value of zero sequence current J of line JbIs the minimum value of the zero sequence current J of the line J.
And A2, calculating the mismatching degree of each line of the power distribution network according to the equivalent distance of the current amount between each line of the power distribution network. The calculation expression of the mismatching degree of each line of the power distribution network is as follows:
wherein HIAnd M is the total line value of the power distribution network.
After a single-phase grounding fault occurs on a certain line of the power distribution network, zero-sequence current of the line is different from other lines of the power distribution network, the unmatched characteristics among the lines are extracted by specifically comparing Euclidean distance models of zero-sequence current values of each line and the other lines, compared with other physical characteristics, the unmatched characteristics are more relevant to fault events and serve as input characteristics of follow-up machine learning, the training effect can be effectively improved, compared with the prior art, the unmatched degree is only related to the current size difference of different lines, whether grounding equivalent resistance is close to 0 or not does not need to be judged, and therefore the influence of a high-resistance grounding phenomenon is avoided.
In step S2, the method for calculating the polarity of the zero sequence current of each line of the power distribution network according to the zero sequence current of each line of the power distribution network includes the following sub-steps:
b1, calculating a zero sequence current average value of each line of the power distribution network according to the zero sequence current of each line of the power distribution network, wherein the calculation expression is as follows:
wherein I (T) is the value of sampling points of the zero sequence current I of the line I at the time T, T is the sampling duration, N is the total number of the sampling points in the sampling duration T,
and the zero sequence current average value of the line I is obtained.
And B2, recording the zero sequence current polarity of the power distribution network line with the zero sequence current average value more than or equal to 0 as positive, and recording the zero sequence current polarity of the power distribution network line with the zero sequence current average value less than 0 as negative.
The zero sequence current is a dynamically changing value, the measured zero sequence current is obtained in engineering, the real situation of the zero sequence current can be obtained better by keeping a certain sampling time to carry out multi-point sampling than single-point sampling, and in the extraction process of the polarity characteristics of the zero sequence current, the zero sequence current is more reasonable and accurate by taking the positive and negative of the zero sequence current average value as the basis.
S3, constructing a training set and a testing set according to the mismatching degree and the zero sequence current polarity of each line of the power distribution network, and training a stack self-encoder through the training set;
the stack self-encoder comprises L self-encoders; the self-encoder is used for carrying out unsupervised learning on characteristics of each line of the power distribution network and comprises an encoder, a hidden layer and a decoder, wherein the input end of the hidden layer is in communication connection with the output end of the encoder, and the input end of the decoder is in communication connection with the output end of the hidden layer.
Step S3 includes the following substeps:
and S31, constructing a training set and a testing set according to the mismatching degree and the zero sequence current polarity of each line of the power distribution network.
And S32, sequentially training and testing the 1 st to L self-encoders through the training set to finish the independent training test of the L self-encoders of the stack self-encoder.
S33, and L self-encoders finished by the independent training of the joint stack self-encoder.
Step S33 includes the following substeps:
c1, shielding the communication connection among the encoder, the hidden layer and the decoder of each of the L self-encoders;
c2, using the input end of the encoder of the 1 st self-encoder as the input end of the stack self-encoder, and the output end of the decoder as the output end of the stack self-encoder;
c3, using the output sequence of the encoder of the 1 st self-encoder as the input sequence of the hidden layer;
c4, taking the output sequence of the hidden layer of the mth self-encoder as the input sequence of the hidden layer of the (m + 1) th self-encoder, wherein m is more than or equal to 1 and less than or equal to L-1;
c5, using the output sequence of the hidden layer of the L-th self-encoder as the input sequence of the decoder;
and C6, taking the output sequence of the decoder of the nth self-encoder as the input sequence of the decoder of the (n-1) th self-encoder, wherein n is more than or equal to 2 and less than or equal to L.
The stacked self-encoder of the present embodiment has 2 self-encoders. The first self-encoder, the hidden layer size is 100, the weight regularization coefficient is 0.004, the sparse regularization term coefficient is 4, and the coefficient ratio is 0.15; in the second self-encoder, the hidden layer size is 50, the weight regularization coefficient is 0.002, the sparse regularization coefficient is 4, and the sparse ratio is 0.1. The combination is shown in fig. 3.
The designed self-encoder combination mode shields redundant encoders in the combination process, and a similar unstacking and stacking mode is adopted in the stacking process of the hidden layers of the self-encoders and the decoders, so that compared with the simple front-back stacking of the self-encoders, the stability and the continuity of signal propagation can be improved.
And S34, training the L self-coders after combination through a training set to realize network fine adjustment of the stack self-coder and enable the stack self-coder to complete training.
Compared with the conventional training method in the field of machine learning, the method can avoid the phenomenon that parameter oscillation is conducted to another stage when the first stage does not converge, can further save training time, and can also avoid the situation that the stack self-encoder falls into a local optimal solution, so that the stack self-encoder can reach a global optimal solution at a higher speed, and the accuracy of fault line selection is improved.
And S4, inputting a test set into the trained stack self-encoder, and selecting the power distribution network line with the single-phase earth fault.
By the method of the embodiment, the power distribution network of the embodiment is detected, and the conditions that 4 lines have different fault conditions at the fault initial angle of 0 °, 30 °, 45 °, 60 °, 90 °, 120 °, the transition resistance of 10 Ω, 100 Ω, 200 Ω, 250 Ω, 500 Ω and 500 Ω are sequentially tested, the fault positions are 10%, 20%, 50%, 60% and 90% of the length from the bus, the total number of 720 test samples, 4 × 6 × 6 × 5, and the fault line selection success rate is 100% under different ground fault types and high-resistance grounding conditions are tested.
In order to simulate the effect in the outdoor environment, white gaussian noise is applied to the zero sequence current of each line in the laboratory test environment, as shown in table 1, the accuracy can reach over 98% under three different noise interferences, and the applicability, reliability and robustness are high.
TABLE 1 accuracy of embodiments of the present invention under white Gaussian noise interference
In summary, by utilizing the physical phenomenon that after a single-phase earth fault of a power distribution network line occurs, the zero sequence current waveform of the power distribution network line can change, the mismatch characteristic and the polarity characteristic of the zero sequence current of each line reflected by the zero sequence current are taken as starting points, and the mismatch characteristic and the polarity characteristic of the zero sequence current of each line are analyzed through a stack self-encoder decoder with machine learning capability, so that the fault line is finally identified, the accuracy is high, the influence of the high-resistance earth fault is avoided, and the fault line identification method has good applicability, reliability and robustness.
The principle and the implementation mode of the invention are explained by applying specific embodiments in the invention, and the description of the embodiments is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Those skilled in the art can make various other specific changes and combinations based on the teachings of the present invention without departing from the spirit of the invention, and these changes and combinations are within the scope of the invention.
Claims (9)
1. A single-phase earth fault line selection method for a power distribution network is characterized by comprising the following steps:
s1, measuring zero sequence current of each line of the power distribution network;
s2, calculating the mismatching degree and the zero sequence current polarity of each line of the power distribution network according to the zero sequence current of each line of the power distribution network;
s3, constructing a training set and a testing set according to the mismatching degree and the zero sequence current polarity of each line of the power distribution network, and training a stack self-encoder through the training set;
and S4, inputting a test set into the trained stack self-encoder, and selecting the power distribution network line with the single-phase earth fault.
2. The method for selecting the single-phase earth fault of the power distribution network according to claim 1, wherein in the step S2, the method for calculating the mismatch of the lines of the power distribution network according to the zero sequence current of the lines of the power distribution network comprises the following sub-steps:
a1, calculating the equivalent distance of current quantity among all lines of the power distribution network according to the zero sequence current of all the lines of the power distribution network;
and A2, calculating the mismatching degree of each line of the power distribution network according to the equivalent distance of the current amount between each line of the power distribution network.
3. The method for selecting the single-phase earth fault line of the power distribution network according to claim 2, wherein the calculation expression of the equivalent distance of the current quantities between the lines of the power distribution network in the step a1 comprises the following equation:
H(I，J)＝max(h(i，j)，h(j，i))
h(i，j)＝||ia-jb||
h(j，i)＝||ja-ib||
wherein, H (I, J) is the equivalent distance of the current amount between the line I and the line J, max () is the maximum value operation, I is the zero sequence current of the line I, J is the zero sequence current of the line J, H (I, J) is the equivalent distance of the one-way current amount from the line I to the line J, H (J, I) is the equivalent distance of the one-way current amount from the line J to the line I, | | | | is the euclidean distance operation, I | | |aIs the maximum value of zero sequence current I of the line IbIs the minimum value, j, of the zero sequence current I of the line IaIs the maximum value of zero sequence current J of line JbIs the minimum value of the zero sequence current J of the line J.
4. The method for selecting the single-phase earth fault of the power distribution network according to claim 3, wherein the computational expression of the degree of mismatch of the lines of the power distribution network in the step A2 is as follows:
wherein HIAnd M is the total line value of the power distribution network.
5. The method for selecting the single-phase earth fault of the power distribution network according to claim 4, wherein in the step S2, the method for calculating the polarity of the zero sequence current of each line of the power distribution network according to the zero sequence current of each line of the power distribution network comprises the following sub-steps:
b1, calculating a zero sequence current average value of each line of the power distribution network according to the zero sequence current of each line of the power distribution network;
and B2, recording the zero sequence current polarity of the power distribution network line with the zero sequence current average value more than or equal to 0 as positive, and recording the zero sequence current polarity of the power distribution network line with the zero sequence current average value less than 0 as negative.
6. The method for selecting the single-phase earth fault line of the power distribution network according to claim 5, wherein the zero sequence current average value calculation expression of each line of the power distribution network in the step B1 is as follows:
wherein I (T) is the value of sampling points of the zero sequence current I of the line I at the time T, T is the sampling duration, N is the total number of the sampling points in the sampling duration T,
and the zero sequence current average value of the line I is obtained.
7. The single-phase ground fault line selection method for the power distribution network according to claim 1, wherein the stack self-encoder comprises L self-encoders; the self-encoder is used for carrying out unsupervised learning on characteristics of each line of the power distribution network and comprises an encoder, a hidden layer and a decoder, wherein the input end of the hidden layer is in communication connection with the output end of the encoder, and the input end of the decoder is in communication connection with the output end of the hidden layer.
8. The single-phase ground fault line selection method for the power distribution network according to claim 7, wherein the step S3 comprises the following sub-steps:
s31, constructing a training set and a testing set according to the mismatching degree and the zero sequence current polarity of each line of the power distribution network;
s32, sequentially training and testing 1 st to L self-encoders through a training set to finish the independent training test of L self-encoders of the stack self-encoder;
s33, L self-encoders which are independently trained and completed by the combined stack self-encoder;
and S34, training the L self-coders after combination through a training set to realize network fine adjustment of the stack self-coder and enable the stack self-coder to complete training.
9. The single-phase ground fault line selection method for the power distribution network according to claim 8, wherein the step S33 comprises the following sub-steps:
c1, shielding the communication connection among the encoder, the hidden layer and the decoder of each of the L self-encoders;
c2, using the input end of the encoder of the 1 st self-encoder as the input end of the stack self-encoder, and the output end of the decoder as the output end of the stack self-encoder;
c3, using the output sequence of the encoder of the 1 st self-encoder as the input sequence of the hidden layer;
c4, taking the output sequence of the hidden layer of the mth self-encoder as the input sequence of the hidden layer of the (m + 1) th self-encoder, wherein m is more than or equal to 1 and less than or equal to L-1;
c5, using the output sequence of the hidden layer of the L-th self-encoder as the input sequence of the decoder;
and C6, taking the output sequence of the decoder of the nth self-encoder as the input sequence of the decoder of the (n-1) th self-encoder, wherein n is more than or equal to 2 and less than or equal to L.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110881486.4A CN113625107A (en) | 2021-08-02 | 2021-08-02 | Power distribution network single-phase earth fault line selection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110881486.4A CN113625107A (en) | 2021-08-02 | 2021-08-02 | Power distribution network single-phase earth fault line selection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113625107A true CN113625107A (en) | 2021-11-09 |
Family
ID=78382291
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110881486.4A Pending CN113625107A (en) | 2021-08-02 | 2021-08-02 | Power distribution network single-phase earth fault line selection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113625107A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114113909A (en) * | 2021-12-03 | 2022-03-01 | 国网四川省电力公司营销服务中心 | Power distribution network single-phase earth fault line selection method and system |
| CN114910742A (en) * | 2022-05-05 | 2022-08-16 | 湖南腾河智慧能源科技有限公司 | Single-phase fault grounding monitoring method and system, electronic equipment and storage medium |
| CN115047296A (en) * | 2022-08-15 | 2022-09-13 | 四川轻化工大学 | Power distribution network fault section positioning method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103439634A (en) * | 2013-09-02 | 2013-12-11 | 北京四方继保自动化股份有限公司 | Method for fault line selection by means of transient state polarity characteristics of zero-sequence current at grounding moment |
| CN103941163A (en) * | 2014-05-12 | 2014-07-23 | 福州大学 | Resonant earthed system fault line selection method utilizing fuzzy K-means clustering |
| US20160041216A1 (en) * | 2013-03-29 | 2016-02-11 | Beijing Inhand Networks Technology Co., Ltd. | Method and system for detecting and locating single-phase ground fault on low current grounded power-distribution network |
| CN107329041A (en) * | 2017-06-20 | 2017-11-07 | 上海电力学院 | A kind of distribution network fault line selection method based on B-spline bistable state denoising |
| 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 |
| CN110286279A (en) * | 2019-06-05 | 2019-09-27 | 武汉大学 | Based on extreme random forest and the sparse Fault Diagnosis of Power Electronic Circuits method from encryption algorithm of stacking-type |
| CN112180210A (en) * | 2020-09-24 | 2021-01-05 | 华中科技大学 | Power distribution network single-phase earth fault line selection method and system |
| CN112305374A (en) * | 2020-10-22 | 2021-02-02 | 西安工程大学 | Power distribution network single-phase earth fault line selection method |
-
2021
- 2021-08-02 CN CN202110881486.4A patent/CN113625107A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160041216A1 (en) * | 2013-03-29 | 2016-02-11 | Beijing Inhand Networks Technology Co., Ltd. | Method and system for detecting and locating single-phase ground fault on low current grounded power-distribution network |
| CN103439634A (en) * | 2013-09-02 | 2013-12-11 | 北京四方继保自动化股份有限公司 | Method for fault line selection by means of transient state polarity characteristics of zero-sequence current at grounding moment |
| CN103941163A (en) * | 2014-05-12 | 2014-07-23 | 福州大学 | Resonant earthed system fault line selection method utilizing fuzzy K-means clustering |
| CN107329041A (en) * | 2017-06-20 | 2017-11-07 | 上海电力学院 | A kind of distribution network fault line selection method based on B-spline bistable state denoising |
| 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 |
| CN110286279A (en) * | 2019-06-05 | 2019-09-27 | 武汉大学 | Based on extreme random forest and the sparse Fault Diagnosis of Power Electronic Circuits method from encryption algorithm of stacking-type |
| CN112180210A (en) * | 2020-09-24 | 2021-01-05 | 华中科技大学 | Power distribution network single-phase earth fault line selection method and system |
| CN112305374A (en) * | 2020-10-22 | 2021-02-02 | 西安工程大学 | Power distribution network single-phase earth fault line selection method |
Non-Patent Citations (2)
| Title |
|---|
| 谭颖婕: "基于堆栈自动编码器的配电网单相接地故障定位方法研究", 《中国优秀硕士学位论文全文数据库》 * |
| 魏鑫等: "基于Hausdorff距离的配电网单相接地故障选线方法", 《电力系统及其自动化学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114113909A (en) * | 2021-12-03 | 2022-03-01 | 国网四川省电力公司营销服务中心 | Power distribution network single-phase earth fault line selection method and system |
| CN114910742A (en) * | 2022-05-05 | 2022-08-16 | 湖南腾河智慧能源科技有限公司 | Single-phase fault grounding monitoring method and system, electronic equipment and storage medium |
| CN115047296A (en) * | 2022-08-15 | 2022-09-13 | 四川轻化工大学 | Power distribution network fault section positioning method |
| CN115047296B (en) * | 2022-08-15 | 2022-10-25 | 四川轻化工大学 | Power distribution network fault section positioning method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113625107A (en) | Power distribution network single-phase earth fault line selection method | |
| US10338127B2 (en) | Method and system for identifying full parameters of element by fault recording and fault locating method | |
| CN112180210B (en) | Power distribution network single-phase earth fault line selection method and system | |
| Huang et al. | Faulty feeder detection by adjusting the compensation degree of arc‐suppression coil for distribution network | |
| CN110208650A (en) | The more criterion fault-line selecting methods of small current neutral grounding system based on radial basis function network | |
| CN103543376A (en) | Radial basis function neutral network method used for fault line selection of small current grounding system | |
| CN113675850A (en) | Power grid information rapid and accurate sensing method based on nonlinear robust estimation | |
| CN109541395B (en) | Distribution network section positioning method based on Hausdorff under characteristic frequency band | |
| Yuan et al. | Study on fault line detection methods based on multi‐feature fusion in distribution systems | |
| CN115587329A (en) | Power distribution network fault classification method and system based on convolutional neural network extraction features | |
| CN111159841A (en) | Power distribution network short-circuit current calculation method based on data fusion | |
| CN112255493A (en) | Multi-criterion comprehensive voting power distribution network fault line selection method | |
| CN115980514B (en) | Complex power distribution network fault positioning method based on multi-terminal traveling wave frequency matrix | |
| CN103942416B (en) | Voltage estimation method based on weighted node spanning tree | |
| Naidu et al. | Traveling wave based adaptive auto-reclosing and fault location for three-terminal mixed lines | |
| CN112363009B (en) | Single-ended fault location method and system for same-tower line ground fault | |
| CN112710924B (en) | Single-phase earth fault line selection method based on steady-state variable quantity before and after fault | |
| CN111781468B (en) | Asynchronous fault distance measurement method and system for T-shaped high-voltage transmission line | |
| CN115078901A (en) | Power distribution network single-phase earth fault identification method and identification device | |
| CN112505490A (en) | Power distribution network single-phase earth fault line selection method based on mutual difference absolute value sum | |
| Serna-Suárez et al. | Impact of static load models on the power distribution fault location problem | |
| Eldin et al. | High impedance fault detection in EHV series compensated lines using the wavelet transform | |
| CN111796165A (en) | Power distribution network fault positioning method based on self-adaptive Fourier transform | |
| CN112285421A (en) | System side harmonic impedance estimation method based on minimum norm criterion | |
| CN116148599B (en) | Kurtosis and skewness coefficient-based high-resistance ground fault diagnosis protection method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211109 |