CN117310361A - Power distribution network fault patrol positioning method based on intelligent perception and equipment image - Google Patents
Power distribution network fault patrol positioning method based on intelligent perception and equipment image Download PDFInfo
- Publication number
- CN117310361A CN117310361A CN202310858088.XA CN202310858088A CN117310361A CN 117310361 A CN117310361 A CN 117310361A CN 202310858088 A CN202310858088 A CN 202310858088A CN 117310361 A CN117310361 A CN 117310361A
- Authority
- CN
- China
- Prior art keywords
- fault
- distribution network
- power distribution
- equipment image
- sparrow
- 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
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000008447 perception Effects 0.000 title claims abstract description 14
- 241000287127 Passeridae Species 0.000 claims abstract description 36
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 10
- 230000006978 adaptation Effects 0.000 claims description 9
- 210000002569 neuron Anatomy 0.000 claims description 8
- 238000010845 search algorithm Methods 0.000 claims description 8
- 210000004027 cell Anatomy 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 5
- 101100499229 Mus musculus Dhrsx gene Proteins 0.000 claims description 4
- 230000003213 activating effect Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims 1
- 230000007704 transition Effects 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 4
- 238000002347 injection Methods 0.000 description 4
- 239000007924 injection Substances 0.000 description 4
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 2
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
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
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Locating Faults (AREA)
Abstract
The invention provides a power distribution network fault patrol positioning method based on intelligent perception and equipment image, and belongs to the technical field of fault positioning. Firstly, constructing a T topology distribution network model containing branches, and collecting fault data; preprocessing the collected data to generate a fault data set; constructing an LSTM model, setting initial parameters of an LSTM network and a sparrow algorithm, calculating optimal parameters of the LSTM network through the sparrow algorithm, establishing a corresponding network, constructing a final model to realize fault location, and analyzing and evaluating a location result. The positioning scheme of the invention has higher precision, is less influenced by fault types, has stronger noise resistance and transition resistance and is easy to realize.
Description
Technical Field
The invention relates to a power distribution network fault patrol positioning method based on intelligent perception and equipment image, and belongs to the technical field of fault positioning.
Background
With the development of modern technology and the expansion of power systems, network topology is more complex, and the power stability needs to be improved. In order to ensure the stable operation of the power distribution network and maintain the residential electricity safety, firstly, the occurrence of faults of a power system is reduced, secondly, the position of a fault point can be found quickly and accurately after the faults occur as far as possible, and the faults are timely removed. Therefore, in order to improve the power supply stability, safety and reliability, reduce economic loss and reduce the workload of workers, the invention provides a power distribution network fault patrol positioning scheme based on intelligent perception and equipment image.
At present, various fault positioning schemes of the power distribution network are proposed at home and abroad, and the fault positioning schemes can be divided into active positioning and passive positioning. The active mode can be divided into a medium resistance method and an injection method; passive methods can be classified into impedance methods, traveling wave methods, and artificial intelligence methods. Active fault localization is based on injection methods, which require injection of a large fault current signal to identify the returned signal. The invention provides a non-ground fault positioning method, which considers unbalanced capacitor injection and solves the problem of single-phase ground. The impedance method is simple in principle, is easily influenced by the transition resistance, and an improved positioning scheme is provided for the problem, so that the influence of the transition resistance is reduced, and the overall positioning accuracy is still to be improved. The traveling wave method comprises a single-end traveling wave method and a double-end traveling wave method, the acquired information of the two traveling wave methods is different, and the accuracy of the double-end traveling wave method is higher than that of the single-end traveling wave method in general. The invention provides a novel double-end traveling wave ranging method, which does not require time synchronization, can identify different wave heads, and has more complex fault mechanism and calculation. Along with the rapid rise of intelligent technology and neural networks, a positioning method based on artificial intelligence is provided by a plurality of inventions, and the invention utilizes the artificial neural network to establish the corresponding relation between fault positioning and fault characteristic values, so that different fault types can be identified and positioned, and the positioning precision is higher.
At present, the following problems need to be solved in power distribution network positioning: 1) Network topology and operation mode are complex and are difficult to describe by functional relation. 2) There are fewer measuring devices and there are difficulties in acquiring data. 3) The existing traditional positioning scheme depends on the characteristics of the traditional positioning scheme, and there is room for improvement.
Disclosure of Invention
The invention aims to provide a power distribution network fault patrol positioning method based on intelligent perception and equipment image, which has higher positioning scheme precision, stronger noise resistance and transition resistance and is easy to realize.
The invention aims to achieve the aim, and the aim is achieved by the following technical scheme:
step 1: constructing a T-shaped topological power distribution network model containing branches in PSCAD or EMTDC, and collecting fault data;
step 2: preprocessing the acquired data, and obtaining a line mode voltage component through phase mode conversion;
step 3: identifying a traveling wave head through a wavelet mode maximum value, carrying out normalization processing on line mode voltage data to generate a fault data set, and dividing the fault data set into a training set and a testing set;
step 4: constructing an LSTM network to perform fault distance fitting, and determining an optimal initial structure and parameters of the LSTM network through a sparrow search algorithm, wherein the parameters comprise the number of hidden neurons, the learning rate and the iteration times;
step 5: and inputting the test set into the constructed LSTM network to calculate the fault distance, and determining the fault position according to the fault distance.
Preferably, the specific way of determining the optimal initial structure and parameters of the LSTM network by using the sparrow search algorithm is as follows:
setting initial parameters of a sparrow algorithm, including the size and the maximum iteration number of the sparrow population;
randomly generating an initial sparrow populationWherein->Representing the number of LSTM hidden layer neurons, +.>Representing learning rate->Representing the iteration times, the sparrow population is +.>The initialization parameters of the (1) are applied to an LSTM model, and a corresponding network structure is established;
setting initial positions of discoverers and joiners in the sparrow population, and determining an fitness function; solving the new fitness value to obtain an optimal solution with the minimum current fitness;
continuously updating the positions of the discoverer and the joiner, and calculating a new fitness value; if the adaptation degree becomes larger after the sparrow updates the position, the original state is kept unchanged, otherwise, the sparrow population is changed according to the updated adaptation degree value, and the latest adaptation degree value is solved, so that the optimal position is obtained;
and judging whether the iteration times and the fitness of the algorithm reach the threshold value, if so, distributing the optimal parameters to the LSTM network, and if not, returning to the previous step for re-iteration.
Preferably, the fitness function is an average absolute percentage error or a mean square error.
Preferably, the threshold is 0.004.
Preferably, the calculating the fault distanceThe specific formula is as follows:
，
，
wherein,activating a function for sigmoid->Output at time t; />For outputting the weight of the gate, +.>For outputting the bias of the gate +.>Memory cell at time t +.>For the output value of the hidden layer at time t-1, < >>Is the input value at time t.
Preferably, the memory unit at the time tThe specific formula is as follows:
，
wherein,for the overall output of the forgetting gate, +.>Memory cell at time t-1 +.>In order for the information to be stored,is a candidate value vector.
Preferably, the information to be storedAnd candidate value vector +.>The specific formula is as follows:
wherein,and->The weights of the input gate and the candidate value vector are respectively; />And->The bias of the input gate and candidate vector, respectively.
Preferably, the overall output of the forgetting gateThe specific formula is as follows:
wherein W is the weight of the forgetting gate,is the bias of the forgetting gate.
The invention has the advantages that: the positioning scheme of the invention has higher precision, is less influenced by fault types, has stronger noise resistance and transition resistance and is easy to realize.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention.
Fig. 1 is a schematic diagram of a T topology distribution network model of the present invention.
Fig. 2 is a schematic diagram of the principle of the small-section circuit of the present invention.
FIG. 3 is a schematic diagram of a failure network of the present invention.
Fig. 4 is a schematic diagram of an equivalent normal network according to the present invention.
Fig. 5 is a schematic diagram of an equivalent fault network of the present invention.
Fig. 6 is a schematic diagram of attenuation law of traveling wave heads at different distances according to the present invention.
FIG. 7 is a schematic diagram of an LSTM cell of the present invention.
FIG. 8 is a schematic flow chart of the SSA algorithm of the invention.
Fig. 9 is a schematic overall flow chart of the present invention.
FIG. 10 is a comparison of the distances between faults according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
1: firstly, a T-shaped topological power distribution network model containing branches is built in PSCAD/EMTDC, and fault data are collected.
Distribution network topological structure and data acquisition:
first, in order to deeply study the fault characteristics of the power distribution network, a topology model of the power distribution network needs to be established. Because the distribution network has the characteristic of multi-branch operation, the system is more complex, and therefore, PSCAD/EMTP is utilized to construct a T-shaped topological line with branch lines for research.
The study is carried out by using a T topology distribution network model shown in fig. 1, the length of a main line of a line is 50km, and a line branch is arranged at the midpoint of the line, and the length of the line branch is 25km. In order to fully simulate the fault condition in the actual engineering, a plurality of groups of fault points are arranged on each branch circuit of the power distribution network model of fig. 1, the fault interval is set to be 0.5km, and 147 fault positions are all arranged. Meanwhile, 4 different fault resistors are also arranged, and the resistance values of the fault resistors are respectively 0.01Ω, 1Ω, 10Ω and 15Ω. The time of occurrence of the fault also affects the fault traveling wave, so that it is necessary to study the fault phase angle when the fault occurs. The failure time is equally divided into 6 times, and is set to 0 °, 60 °, 120 °, 180 °, 240 °, 300 °, respectively.
In combination with the above, 24 fault parameters (including 4 fault resistances and 6 fault phase angles) are respectively set at each fault point. The fault data collected by the present model is 147×24=3528 set.
2: preprocessing the acquired data, and obtaining a line mode voltage component through phase mode conversion.
3: and identifying the traveling wave head through the wavelet mode maximum value, and carrying out normalization processing on the linear mode voltage data to generate a fault data set.
4: and constructing an LSTM network to perform fault distance fitting, and determining an optimal initial structure and parameters of the LSTM network through a sparrow search algorithm, wherein the parameters comprise the number of hidden neurons, the learning rate and the iteration times. The LSTM network model can fit the fault distance well, but the problem that parameters such as the number of hidden layer neurons, the learning rate and the iteration number are difficult to determine in the network is also faced in the network training process. And a sparrow algorithm with higher precision and higher convergence speed than the particle swarm algorithm is selected for parameter optimization. Sparrow search algorithms divide sparrow behavior into three categories, discoverers, enrollees and scouts. Fig. 8 shows a detailed flow of the sparrow search algorithm.
The specific mode for determining the optimal initial structure and parameters of the LSTM network by the sparrow search algorithm is as follows:
setting initial parameters of a sparrow algorithm, including the size and the maximum iteration number of the sparrow population;
randomly generating an initial sparrow populationWherein->Representing the number of LSTM hidden layer neurons, +.>Representing learning rate->Representing the iteration times, the sparrow population is +.>The initialization parameters of the (1) are applied to an LSTM model, and a corresponding network structure is established;
setting initial positions of discoverers and joiners in the sparrow population, and determining an fitness function; solving the new fitness value to obtain an optimal solution with the minimum current fitness;
continuously updating the positions of the discoverer and the joiner, and calculating a new fitness value; if the adaptation degree becomes larger after the sparrow updates the position, the original state is kept unchanged, otherwise, the sparrow population is changed according to the updated adaptation degree value, and the latest adaptation degree value is solved, so that the optimal position is obtained;
and judging whether the iteration times and the fitness of the algorithm reach the threshold value, if so, distributing the optimal parameters to the LSTM network, and if not, returning to the previous step for re-iteration.
5: and inputting the test set into the constructed LSTM network to calculate the fault distance, and determining the fault position according to the fault distance.
Example 2
The LSTM network can realize continuous data transmission and selective storage data through the gating device, and the unit structure diagram is as follows:
the LSTM has three gating units, namely a forget gate, an input gate and an output gate. The forget gate may selectively store information. The formula is as follows:
，
wherein,activating a function for sigmoid; w is the weight of the forgetting gate; />The output value of the hidden layer at the previous moment; />The input value is the current moment; />Bias for forgetting the door; finally use->As the overall output of the forget gate.
The function of the input gate is to receive and modify parameters, which transfer the information of the previous moment and the information of the current moment to the memory unit and determine which parameters are to be updated. The formula is as follows:
，
wherein,information to be stored; />And->The weights of the input gate and the candidate value vector are respectively; />For candidate value vector, ++>And->The bias of the input gate and candidate vector, respectively.
After the LSTM reads the information at the current time, the cell state must be updated to complete the slaveTo the point ofThe memory unit updates the formula:
，
the output gate is used for controlling the information output at the current moment, and can autonomously select the information to be output. The formula is as follows:
，
wherein,output at time t; />For outputting the weight of the gate, +.>To output the gate bias.
Example 3
Analysis of traveling wave propagation attenuation characteristics of power distribution network
The actual power line is a lossy line, and attenuation of the traveling wave occurs during operation, mainly due to attenuation of the amplitude of the wire resistance and conductance.
Analysis of FIG. 2 can yield KCL and KVL equations as:
(1)
solving the differential equation of formula (1) results in:
(2)
(3)
(4)
(5)
wherein,is an attenuation coefficient representing the amplitude attenuation characteristic of traveling wave propagation; />Is a phase constant used to characterize the phase change during traveling wave propagation. Attenuation coefficient->Frequency of reception->It increases with increasing frequency and varies non-linearly. As can be seen from the formulas (2) and (3), the amplitude of the traveling wave passing through the lossy conducting wire decays exponentially, and the larger the traveling wave propagation distance is, the smaller the amplitude of the traveling wave is.
The circuit (fig. 3) at the time of power system failure can be equivalent to a network (fig. 4) operating normally and a failed network (fig. 5) by the superposition theorem. When a fault occurs, it is equivalent to adding a step excitation source at the point of the fault. The step voltage traveling wave amplitude of the input power grid is obtained as follows:
(6)
wherein Uf is a fault excitation source, zf is a fault resistor, Z1 is the wave impedance of the line, and the voltage U is affected by the voltage and the fault resistor when a fault occurs. When the frequency is more than 0, the frequency characteristic of the step signal is as follows:
(7)
attenuation coefficientFrequency of reception->The higher the frequency, the greater the attenuation coefficient and the faster the attenuation. The attenuated amplitude of the different frequencies can be identified at the fault measurement:
(8)
wherein,representative frequency is +.>Initial traveling wave amplitude at that time.
From equation (8), it can be seen that the higher the frequency, the more severely the traveling wave decays, and therefore the longer the traveling wave travels along the line, the smoother the wave head becomes. The traveling wave heads at different fault points are as follows:
as can be seen from fig. 6, the longer the fault distance, the more pronounced the attenuation of the traveling wave head, and the smaller the traveling wave steepness and traveling wave amplitude.
According to the analysis, the fault data (3528 groups of samples) collected by the model are divided into a training set and a testing set according to the proportion of 5:1, and the corresponding label is the fault distance of the line.
After the initial parameters of the LSTM network are determined, in order to ensure that the parameters of the LSTM network model are more accurate, the parameters are quickly determined to reduce the training difficulty of the model and prevent the model from being in local optimum, the parameters of the LSTM network are optimized by utilizing a sparrow algorithm, and the neuron number of the three hidden layers, the learning rate of the model and the iteration number are taken as the parameters to be optimized, namely. Then setting the corresponding parameter range as [30,200 ]]、[30,200]、[30,200]、[0.001,0.02]、[100,900]。
Next, random initialization setting is performed on the positions, fitness values and convergence curves of the sparrow population, the producer proportion of the sparrow population is set to 0.2, the optimizing dimension is set to 5, the population number is set to 20, the iteration number is set to 70, and the safety threshold of the population is set to 0.6. On the basis, the average absolute percentage error MAPE is set as an optimal fitness function of the model, and the threshold value of the average absolute percentage error MAPE is determined to be 0.004 through multiple experimental comparison analysis. And finally, obtaining the optimal parameters of the LSTM model by using SSA optimization, and realizing accurate fault positioning.
In summary, the general scheme of the present invention is shown in FIG. 9.
And randomly extracting 100 groups of samples to perform fault location, and generating a comparison graph of a true value and a predicted value of the fault distance as shown in fig. 10.
Finally, it should be noted that: the foregoing description is only a preferred embodiment of the present invention, and the present invention is not limited thereto, but it is to be understood that modifications and equivalents of some of the technical features described in the foregoing embodiments may be made by those skilled in the art, although the present invention has been described in detail with reference to the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The utility model provides a distribution network fault inspection position method based on intelligent perception and equipment image, which is characterized by comprising the following steps:
step 1: constructing a T-shaped topological power distribution network model containing branches in PSCAD or EMTDC, and collecting fault data;
step 2: preprocessing the acquired data, and obtaining a line mode voltage component through phase mode conversion;
step 3: identifying a traveling wave head through a wavelet mode maximum value, carrying out normalization processing on line mode voltage data to generate a fault data set, and dividing the fault data set into a training set and a testing set;
step 4: constructing an LSTM network to perform fault distance fitting, and determining an optimal initial structure and parameters of the LSTM network through a sparrow search algorithm, wherein the parameters comprise the number of hidden neurons, the learning rate and the iteration times;
step 5: and inputting the test set into the constructed LSTM network to calculate the fault distance, and determining the fault position according to the fault distance.
2. The power distribution network fault patrol positioning method based on intelligent perception and equipment image according to claim 1, wherein the method for determining the optimal initial structure and parameters of the LSTM network by using the sparrow search algorithm is as follows:
setting initial parameters of a sparrow algorithm, including the size and the maximum iteration number of the sparrow population;
randomly generating an initial sparrow populationWherein->Representing the number of LSTM hidden layer neurons, +.>Representing learning rate->Representing the iteration times, the sparrow population is +.>The initialization parameters of the (1) are applied to an LSTM model, and a corresponding network structure is established;
setting initial positions of discoverers and joiners in the sparrow population, and determining an fitness function; solving the new fitness value to obtain an optimal solution with the minimum current fitness;
continuously updating the positions of the discoverer and the joiner, and calculating a new fitness value; if the adaptation degree becomes larger after the sparrow updates the position, the original state is kept unchanged, otherwise, the sparrow population is changed according to the updated adaptation degree value, and the latest adaptation degree value is solved, so that the optimal position is obtained;
and judging whether the iteration times and the fitness of the algorithm reach the threshold value, if so, distributing the optimal parameters to the LSTM network, and if not, returning to the previous step for re-iteration.
3. The intelligent perception and equipment image-based power distribution network fault patrol positioning method according to claim 2, wherein the fitness function is an average absolute percentage error or a mean square error.
4. The power distribution network fault patrol positioning method based on intelligent perception and equipment image according to claim 2, wherein the threshold value is 0.004.
5. The distribution network fault patrol positioning method based on intelligent perception and equipment image according to claim 2, wherein the calculated fault distanceThe specific formula is as follows:
，
，
wherein,activating a function for sigmoid->Output at time t; />For outputting the weight of the gate, +.>For outputting the bias of the gate +.>Memory cell at time t +.>For the output value of the hidden layer at time t-1, < >>Is the input value at time t.
6. The method for locating faults in a power distribution network based on intelligent sensing and equipment image as claimed in claim 5, wherein the memory unit at time tThe specific formula is as follows:
，
wherein,for the overall output of the forgetting gate, +.>Memory cell at time t-1 +.>For information to be stored +.>Is a candidate value vector.
7. The method for locating faults in a power distribution network based on intelligent sensing and equipment image as claimed in claim 7, wherein the information to be stored isAnd candidate value vector +.>The specific formula is as follows:
wherein,and->The weights of the input gate and the candidate value vector are respectively; />And->The bias of the input gate and candidate vector, respectively.
8. According to claim 6The distribution network fault patrol positioning method based on intelligent perception and equipment image is characterized in that the total output of the forgetting doorThe specific formula is as follows:
wherein W is the weight of the forgetting gate,is the bias of the forgetting gate.
9. A power distribution network fault patrol positioning device based on intelligent perception and equipment image, comprising a processor and a memory storing program instructions, wherein the processor is configured to execute the power distribution network fault patrol positioning method based on intelligent perception and equipment image according to any one of claims 1 to 8 when running the program instructions.
10. A storage medium storing program instructions which, when executed, perform the power distribution network fault locating method based on intelligent awareness and equipment image according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310858088.XA CN117310361A (en) | 2023-07-13 | 2023-07-13 | Power distribution network fault patrol positioning method based on intelligent perception and equipment image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310858088.XA CN117310361A (en) | 2023-07-13 | 2023-07-13 | Power distribution network fault patrol positioning method based on intelligent perception and equipment image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117310361A true CN117310361A (en) | 2023-12-29 |
Family
ID=89280068
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310858088.XA Pending CN117310361A (en) | 2023-07-13 | 2023-07-13 | Power distribution network fault patrol positioning method based on intelligent perception and equipment image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117310361A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117668471A (en) * | 2024-02-02 | 2024-03-08 | 国网辽宁省电力有限公司抚顺供电公司 | Tree line discharge fault identification method based on fault traveling wave current characteristics |
-
2023
- 2023-07-13 CN CN202310858088.XA patent/CN117310361A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117668471A (en) * | 2024-02-02 | 2024-03-08 | 国网辽宁省电力有限公司抚顺供电公司 | Tree line discharge fault identification method based on fault traveling wave current characteristics |
| CN117668471B (en) * | 2024-02-02 | 2024-04-05 | 国网辽宁省电力有限公司抚顺供电公司 | Tree line discharge fault identification method based on fault traveling wave current characteristics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108038300B (en) | Optical fiber state evaluation method based on improved membership function combined with neural network | |
| CN109459669B (en) | 10kV power distribution network single-phase earth fault section positioning method | |
| CN103728535A (en) | Extra-high-voltage direct-current transmission line fault location method based on wavelet transformation transient state energy spectrum | |
| WO2022021726A1 (en) | Pmu-based power system state estimation performance evaluation method | |
| CN105260786B (en) | A kind of simulation credibility of electric propulsion system assessment models comprehensive optimization method | |
| CN117310361A (en) | Power distribution network fault patrol positioning method based on intelligent perception and equipment image | |
| LU500551B1 (en) | Virtual load dominant parameter identification method based on incremental learning | |
| CN115618732B (en) | Nuclear reactor digital twin key parameter autonomous optimization data inversion method | |
| Xu et al. | Oversmoothing relief graph convolutional network-based fault diagnosis method with application to the rectifier of high-speed trains | |
| CN108491931A (en) | A method of non-destructive testing precision is improved based on machine learning | |
| CN112883522A (en) | Micro-grid dynamic equivalent modeling method based on GRU (generalized regression Unit) recurrent neural network | |
| CN115128395A (en) | Voltage sag source positioning method and device | |
| CN105353270B (en) | Consider the grid-connected fault-tolerant localization method of power quality disturbance of distributed generation resource | |
| CN114942140A (en) | Rolling bearing fault diagnosis method based on multi-input parallel graph convolution neural network | |
| CN110222098A (en) | Electric power high amount of traffic abnormality detection based on flow data clustering algorithm | |
| CN114330486A (en) | Power system bad data identification method based on improved Wasserstein GAN | |
| Li et al. | A recognition method of plate shape defect based on RBF-BP neural network optimized by genetic algorithm | |
| CN117350170A (en) | Nonlinear oscillation analysis method based on KOOPHAN deep neural network | |
| CN117407675A (en) | Lightning arrester leakage current prediction method based on multi-variable reconstruction combined dynamic weight | |
| Hassani et al. | Design of a cost-effective deep convolutional neural network–based scheme for diagnosing faults in smart grids | |
| Li et al. | Predicting Software Quality by Optimized BP Network Based on PSO. | |
| CN116470504A (en) | Self-healing control system of distributed power distribution network | |
| CN113836818B (en) | Ocean current motion prediction algorithm based on BP neural network prediction model | |
| CN115603320A (en) | Real-time sensing and online evaluation method and system based on wide area measurement data | |
| Shi et al. | A fault location method for distribution system based on one-dimensional convolutional neural network |
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 |