CN112578248A - Partial discharge positioning error correction method and system - Google Patents
Partial discharge positioning error correction method and system Download PDFInfo
- Publication number
- CN112578248A CN112578248A CN202011569407.8A CN202011569407A CN112578248A CN 112578248 A CN112578248 A CN 112578248A CN 202011569407 A CN202011569407 A CN 202011569407A CN 112578248 A CN112578248 A CN 112578248A
- Authority
- CN
- China
- Prior art keywords
- neural network
- delay value
- partial discharge
- network model
- time delay
- 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
- 238000012937 correction Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000005070 sampling Methods 0.000 claims abstract description 53
- 238000013528 artificial neural network Methods 0.000 claims abstract description 45
- 239000002245 particle Substances 0.000 claims abstract description 26
- 238000005457 optimization Methods 0.000 claims description 78
- 238000003062 neural network model Methods 0.000 claims description 68
- 238000005259 measurement Methods 0.000 claims description 44
- 238000012549 training Methods 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 12
- 230000000694 effects Effects 0.000 abstract description 10
- 238000004364 calculation method Methods 0.000 description 3
- 238000009413 insulation Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 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
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Physiology (AREA)
- Feedback Control In General (AREA)
Abstract
The application discloses partial discharge positioning error correction method and system, the accuracy of partial discharge source positioning is improved without improving hardware, only a common sensor is needed for sampling, a better positioning effect can be obtained, and on the basis of using a deep neural network, a weight and a threshold of the deep neural network are optimized by combining a particle swarm algorithm, and the positioning accuracy is further optimized.
Description
Technical Field
The application relates to the technical field of partial discharge detection, in particular to a partial discharge positioning error correction method and system.
Background
Partial Discharge (PD) is an early sign of insulation degradation of high voltage electrical equipment, and therefore PD detection can be an effective means of equipment insulation state assessment. Over time, partial discharges may lead to failure of the entire insulation system, and therefore there is a great need for accurate positioning of the partial discharges. The existing method mainly uses a very high frequency method for local discharge positioning, and the method for improving the positioning accuracy focuses on improving hardware and improving the TDOA algorithm, but the method for improving the hardware has higher requirements on the hardware, and the positioning accuracy effect of the TDOA algorithm is poor.
Disclosure of Invention
The application provides a partial discharge positioning error correction method and a partial discharge positioning error correction system, which are used for obtaining a better partial discharge source positioning effect under the existing hardware condition.
In view of the above, a first aspect of the present application provides a partial discharge positioning error correction method, including:
dividing a GIS structure into a plurality of equally spaced regions, and extracting region sampling points;
collecting partial discharge signals of sampling points, and calculating a measurement time delay value of a partial discharge source signal;
calculating a theoretical time delay value of a local discharge source signal according to the spatial position of the sampling point;
establishing an error optimization neural network model of a measurement delay value and a theoretical delay value of a sampling point partial discharge source signal;
training the error optimization neural network model, and optimizing parameters of the error optimization neural network model by adopting a particle swarm algorithm to obtain a trained error optimization neural network model;
collecting local discharge source signals at any position of a GIS structure, calculating a measurement delay value, and inputting the measurement delay value into a trained error optimization neural network model to obtain a corrected delay value;
and correcting the positioning error of the local discharge source according to the corrected time delay value.
Optionally, dividing the GIS structure into a plurality of equally spaced regions, and extracting region sampling points, including:
the GIS structure is divided into a plurality of equally spaced areas, and a plurality of sampling points are uniformly extracted in each area.
Optionally, the collecting a partial discharge signal of a sampling point, and calculating a measurement delay value of a partial discharge source signal includes:
and acquiring the partial discharge signal of the sampling point through an oscilloscope, and calculating the measurement time delay value of the partial discharge source signal.
Optionally, the calculation formula of the theoretical time delay value of the partial discharge source signal is as follows:
wherein (x)i,yi,zi) Is the coordinate of sensor i, (x)k,yk,zk) As coordinates of the sensor k, τikFor the time difference between sensor i and sensor k receiving the partial discharge ultrahigh frequency signal, (x, y, z)) And upsilon is the propagation speed of the ultrahigh frequency signal in the medium and is the coordinate of the partial discharge source.
Optionally, establishing an error optimization neural network model of a measurement delay value and a theoretical delay value of a local discharge source signal at a sampling point, including:
and normalizing the measured delay value and the theoretical delay value, taking the measured delay value as input, taking the theoretical delay value as output, and constructing an error optimization neural network model.
Optionally, training the error-optimized neural network model, and optimizing parameters of the error-optimized neural network model by using a particle swarm optimization algorithm to obtain the trained error-optimized neural network model, including:
setting the number of hidden layers and the value range and the initial value of the number of nodes of the error optimization neural network model;
optimizing the error optimization neural network model by adopting a particle swarm algorithm to obtain the optimal weight and threshold under the current hidden layer number and node number neural network structure;
training and testing input data under the current neural network structure to obtain and store a test result;
changing the number of hidden layers and the number of nodes in a value range, and if the number of hidden layers and the number of nodes do not meet the end condition, returning to the step of optimizing the error optimization neural network model by adopting a particle swarm algorithm until the end condition is met;
and selecting the error optimization neural network model with the highest prediction precision from the stored test results as the trained error optimization neural network model.
Optionally, the value range of the number of hidden layers and the number of nodes is determined according to a preset empirical formula, where the preset empirical formula is:
wherein L is the number of hidden layers and the number of nodes, m and n are the number of nodes of an input layer and an output layer respectively, and a is a positive number of 0-10.
Optionally, the fitness function for optimizing the error optimization neural network model by using the particle swarm optimization is as follows:
wherein, yjIs a theoretical time delay value of a partial discharge source, y'jIs the output value of the neural network.
A second aspect of the present application provides a partial discharge positioning error correction system, including:
the structure dividing unit is used for dividing the GIS structure into a plurality of equally spaced regions and extracting region sampling points;
the time delay measuring unit is used for collecting the partial discharge signals of the sampling points and calculating the measuring time delay value of the partial discharge source signals;
the theoretical time delay calculating unit is used for calculating a theoretical time delay value of the local discharge source signal according to the spatial position of the sampling point;
the modeling unit is used for establishing an error optimization neural network model of a measurement delay value and a theoretical delay value of a sampling point local discharge source signal;
the network training unit is used for training the error optimization neural network model and optimizing parameters of the error optimization neural network model by adopting a particle swarm algorithm to obtain a trained error optimization neural network model;
the time delay correction unit is used for collecting local discharge source signals at any position of the GIS structure, calculating a measurement time delay value, and inputting the measurement time delay value into a trained error optimization neural network model to obtain a corrected time delay value;
and the error correction unit is used for correcting the positioning error of the local discharge source according to the corrected time delay value.
Optionally, the structure dividing unit is specifically configured to:
the GIS structure is divided into a plurality of equally spaced areas, and a plurality of sampling points are uniformly extracted in each area.
According to the technical scheme, the embodiment of the application has the following advantages:
the application provides a partial discharge positioning error correction method, which comprises the following steps: dividing a GIS structure into a plurality of equally spaced regions, and extracting region sampling points; collecting partial discharge signals of sampling points, and calculating a measurement time delay value of a partial discharge source signal; calculating a theoretical time delay value of a local discharge source signal according to the spatial position of the sampling point; establishing an error optimization neural network model of a measurement delay value and a theoretical delay value of a sampling point partial discharge source signal; training the error optimization neural network model, and optimizing parameters of the error optimization neural network model by adopting a particle swarm algorithm to obtain a trained error optimization neural network model; collecting local discharge source signals at any position of a GIS structure, calculating a measurement delay value, and inputting the measurement delay value into a trained error optimization neural network model to obtain a corrected delay value; and correcting the positioning error of the local discharge source according to the corrected time delay value.
The partial discharge positioning error correction method provided by the application does not need to improve hardware to improve the positioning precision of a partial discharge source, can obtain a better positioning effect only by sampling a common sensor, and further optimizes the positioning precision by combining the weight and the threshold of a particle swarm optimization deep neural network on the basis of using the deep neural network.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other related drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flowchart of a partial discharge positioning error correction method provided in an embodiment of the present application;
FIG. 2 is a flow chart of deep neural network training provided in an embodiment of the present application;
fig. 3 is a schematic structural diagram of a partial discharge positioning error correction system provided in an embodiment of the present application.
Detailed Description
In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Example 1
For easy understanding, referring to fig. 1 and fig. 2, the present application provides an embodiment of a partial discharge positioning error correction method, including:
It should be noted that, the selection of the sample partial discharge source plays a decisive role in the correction of the partial discharge positioning error and the final positioning, so it is necessary to ensure that the sampling points are uniformly distributed at each position of the GIS as much as possible. The sampling points are uniformly selected in each area.
And 102, collecting partial discharge signals of sampling points, and calculating a measurement time delay value of a partial discharge source signal.
It should be noted that the measurement delay value of the partial discharge signal may be directly obtained by an oscilloscope or other experimental apparatus, but due to the hardware condition factor, a certain delay error may exist, so in the embodiment of the present application, the influence of the hardware factor on the positioning result is reduced by using a delay error correction manner, so as to improve the positioning accuracy.
And 103, calculating a theoretical time delay value of the local discharge source signal according to the spatial position of the sampling point.
It should be noted that the theoretical time delay value of the partial discharge signal may be calculated by a theoretical formula, and the calculation formula of the theoretical time delay value of the partial discharge source signal in the embodiment of the present application is as follows:
wherein (x)i,yi,zi) Is the coordinate of sensor i, (x)k,yk,zk) As coordinates of the sensor k, τikThe time difference between the sensor i and the sensor k receiving the partial discharge ultrahigh frequency signal is shown, the (x, y, z) is the coordinate of a partial discharge source, and the upsilon is the propagation speed of the ultrahigh frequency signal in the medium.
And step 104, establishing an error optimization neural network model of the measurement delay value and the theoretical delay value of the local discharge source signal of the sampling point.
It should be noted that, in the embodiment of the present application, a deep neural network is adopted to compensate a delay error, normalization processing is performed on a measurement delay value and theoretical delay value data, the measurement delay value is used as an input of the deep neural network, the theoretical delay value is used as an output of the deep neural network, and an error optimization neural network model of the measurement delay value and the theoretical delay value of a partial discharge source signal is constructed.
And 105, training the error optimization neural network model, and optimizing parameters of the error optimization neural network model by adopting a particle swarm optimization algorithm to obtain the trained error optimization neural network model.
It should be noted that, in the embodiment of the present application, the error optimization neural network that is initially established needs to be trained, and the structure of the deep neural network is initially determined, that is, the number of hidden layers of the deep neural network and the value range and the initial value of the number of nodes, and the selection of the number of hidden layers and the number of nodes of the deep neural network are initially determinedThe method plays a crucial role in correcting the partial discharge delay error, but at present, no specific formula is available to specifically calculate the number of layers and the number of nodes of the hidden layer, so that a way of determining the number of layers and the number of nodes of the hidden layer is provided in the embodiment of the application, and a value range is determined through an empirical formula, where the empirical formula is:
wherein L is the number of hidden layers and the number of nodes, m and n are the number of nodes of an input layer and an output layer respectively, and a is a positive number of 0-10. In the embodiment of the present application, the number of nodes of the input layer and the output layer is 4, and the range of the number of hidden layer layers is [3, 13 ]]Under the condition of ensuring the network precision and efficiency, the value range can be expanded to [2, 15 ]]. And then optimizing the deep neural network by using a particle swarm algorithm to obtain the optimal weight and threshold under the neural network structure of the current hidden layer number and node number, training and testing the input data under the current structure to obtain a corresponding test result and storing the test result, then changing the hidden layer number and the node number in the value range, if the end condition is not met, namely all elements in the value range of the hidden layer number and the node number are not traversed in sequence, repeatedly returning to execute the step of optimizing the deep neural network by using the particle swarm algorithm until the end condition is met, and selecting the deep network model with the highest prediction precision from the stored result as the trained error optimization neural network model after the end.
In one embodiment, the fitness function used for optimizing the deep neural network using the particle swarm optimization may be expressed by a mean square error of theoretical output values and actual output values of the deep neural network, that is:
wherein, yjIs a theoretical time delay value of a partial discharge source, y'jIs the output value of the neural network.
And 106, collecting local discharge source signals at any position of the GIS structure, calculating a measurement delay value, and inputting the measurement delay value into a trained error optimization neural network model to obtain a corrected delay value.
It should be noted that, after the final error optimization neural network model is determined, partial discharge signals of other arbitrary positions of the GIS are collected and measured to obtain a measurement delay value, and the measurement delay value is input into the neural network to obtain a corrected delay value.
And step 107, correcting the positioning error of the local discharge source according to the corrected time delay value.
It should be noted that, after the time delay correction, the corrected time delay value is used to correct the positioning error of the local discharge source.
It should be noted that when the positioning accuracy is severely reduced due to the change of the surrounding environment, a good error correction effect can be obtained again by repeatedly updating and optimizing the neural network, so as to ensure the positioning effect.
The partial discharge positioning error correction method provided by the embodiment of the application does not need to improve hardware to improve the positioning precision of the partial discharge source, can obtain a better positioning effect only by sampling with a common sensor, and further optimizes the weight and the threshold of the deep neural network by combining the particle swarm optimization on the basis of using the deep neural network, so that the positioning precision is further optimized.
Example 2
For ease of understanding, referring to fig. 3, an embodiment of a partial discharge positioning error correction system is provided in the present application, including:
the structure dividing unit is used for dividing the GIS structure into a plurality of equally spaced regions and extracting region sampling points;
the time delay measuring unit is used for collecting the partial discharge signals of the sampling points and calculating the measuring time delay value of the partial discharge source signals;
the theoretical time delay calculating unit is used for calculating a theoretical time delay value of the local discharge source signal according to the spatial position of the sampling point;
the modeling unit is used for establishing an error optimization neural network model of a measurement delay value and a theoretical delay value of a sampling point local discharge source signal;
the network training unit is used for training the error optimization neural network model and optimizing parameters of the error optimization neural network model by adopting a particle swarm algorithm to obtain a trained error optimization neural network model;
the time delay correction unit is used for collecting local discharge source signals at any position of the GIS structure, calculating a measurement time delay value, and inputting the measurement time delay value into a trained error optimization neural network model to obtain a corrected time delay value;
and the error correction unit is used for correcting the positioning error of the local discharge source according to the corrected time delay value.
The structure dividing unit is specifically configured to:
the GIS structure is divided into a plurality of equally spaced areas, and a plurality of sampling points are uniformly extracted in each area.
It should be noted that, in the embodiment of the present application, the GIS structure is divided into a plurality of equally spaced regions, and region sampling points are extracted. The selection of the sample partial discharge source plays a decisive role in correcting the partial discharge positioning error and finally positioning, so that the sampling points are required to be uniformly distributed at each position of the GIS as much as possible. The sampling points are uniformly selected in each area.
And collecting the partial discharge signals of the sampling points, and calculating the measurement time delay value of the partial discharge source signal.
The measurement time delay value of the partial discharge signal can be directly obtained through an oscilloscope and other experimental instruments, but due to the hardware condition factors, a certain time delay error exists, so that the influence of the hardware factors on the positioning result is reduced by using a time delay error correction mode in the embodiment of the application, and the positioning precision is improved.
And calculating the theoretical time delay value of the local discharge source signal according to the spatial position of the sampling point.
The theoretical time delay value of the partial discharge signal can be calculated through a theoretical formula, and the calculation formula of the theoretical time delay value of the partial discharge source signal in the embodiment of the application is as follows:
wherein (x)i,yi,zi) Is the coordinate of sensor i, (x)k,yk,zk) As coordinates of the sensor k, τikThe time difference between the sensor i and the sensor k receiving the partial discharge ultrahigh frequency signal is shown, the (x, y, z) is the coordinate of a partial discharge source, and the upsilon is the propagation speed of the ultrahigh frequency signal in the medium.
And establishing an error optimization neural network model of the measurement delay value and the theoretical delay value of the local discharge source signal of the sampling point.
In the embodiment of the application, a deep neural network is adopted to compensate the time delay error, the measured time delay value and the theoretical time delay value data are subjected to normalization processing, the measured time delay value is used as the input of the deep neural network, the theoretical time delay value is used as the output of the deep neural network, and an error optimization neural network model of the measured time delay value and the theoretical time delay value of the partial discharge source signal is constructed.
And training the error optimization neural network model, and optimizing parameters of the error optimization neural network model by adopting a particle swarm optimization algorithm to obtain the trained error optimization neural network model.
In the embodiment of the application, the preliminarily established error optimization neural network needs to be trained, the structure of the deep neural network is preliminarily determined, that is, the number of hidden layers of the deep neural network and the value range and the initial value of the node number are preliminarily determined, the selection of the number of hidden layers and the node number of the deep neural network plays a crucial role in correcting the local discharge delay error, but no specific formula is available at present for specifically calculating the number of hidden layers and the node numberThe number of the hidden layers is determined according to an empirical formula, and the empirical formula is as follows:
wherein L is the number of hidden layers and the number of nodes, m and n are the number of nodes of an input layer and an output layer respectively, and a is a positive number of 0-10. In the embodiment of the present application, the number of nodes of the input layer and the output layer is 4, and the range of the number of hidden layer layers is [3, 13 ]]Under the condition of ensuring the network precision and efficiency, the value range can be expanded to [2, 15 ]]. And then optimizing the deep neural network by using a particle swarm algorithm to obtain the optimal weight and threshold under the neural network structure of the current hidden layer number and node number, training and testing the input data under the current structure to obtain a corresponding test result and storing the test result, then changing the hidden layer number and the node number in the value range, if the end condition is not met, namely all elements in the value range of the hidden layer number and the node number are not traversed in sequence, repeatedly returning to execute the step of optimizing the deep neural network by using the particle swarm algorithm until the end condition is met, and selecting the deep network model with the highest prediction precision from the stored result as the trained error optimization neural network model after the end.
In one embodiment, the fitness function used for optimizing the deep neural network using the particle swarm optimization may be expressed by a mean square error of theoretical output values and actual output values of the deep neural network, that is:
wherein, yjIs a theoretical time delay value of a partial discharge source, y'jIs the output value of the neural network.
Collecting local discharge source signals of any position of a GIS structure, calculating a measurement delay value, and inputting the measurement delay value into a trained error optimization neural network model to obtain a corrected delay value.
And after the final error optimization neural network model is determined, acquiring partial discharge signals of any other positions of the GIS, measuring to obtain a measurement time delay value, and inputting the measurement time delay value into the neural network to obtain a corrected time delay value.
And correcting the positioning error of the local discharge source according to the corrected time delay value.
And after time delay correction, correcting the positioning error of the local discharge source by adopting the corrected time delay value.
When the surrounding environment changes to cause the positioning accuracy to be severely reduced, a good error correction effect can be obtained again by repeatedly updating and optimizing the neural network so as to ensure the positioning effect.
The partial discharge positioning error correction method provided by the embodiment of the application does not need to improve hardware to improve the positioning precision of the partial discharge source, can obtain a better positioning effect only by sampling with a common sensor, and further optimizes the weight and the threshold of the deep neural network by combining the particle swarm optimization on the basis of using the deep neural network, so that the positioning precision is further optimized.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (10)
1. A partial discharge positioning error correction method, comprising:
dividing a GIS structure into a plurality of equally spaced regions, and extracting region sampling points;
collecting partial discharge signals of sampling points, and calculating a measurement time delay value of a partial discharge source signal;
calculating a theoretical time delay value of a local discharge source signal according to the spatial position of the sampling point;
establishing an error optimization neural network model of a measurement delay value and a theoretical delay value of a sampling point partial discharge source signal;
training the error optimization neural network model, and optimizing parameters of the error optimization neural network model by adopting a particle swarm algorithm to obtain a trained error optimization neural network model;
collecting local discharge source signals at any position of a GIS structure, calculating a measurement delay value, and inputting the measurement delay value into a trained error optimization neural network model to obtain a corrected delay value;
and correcting the positioning error of the local discharge source according to the corrected time delay value.
2. The partial discharge positioning error correction method of claim 1, wherein dividing the GIS structure into a plurality of equally spaced regions and extracting region sampling points comprises:
the GIS structure is divided into a plurality of equally spaced areas, and a plurality of sampling points are uniformly extracted in each area.
3. The method for correcting the partial discharge positioning error according to claim 2, wherein the step of collecting the partial discharge signal of the sampling point and calculating the measurement time delay value of the partial discharge source signal comprises the steps of:
and acquiring the partial discharge signal of the sampling point through an oscilloscope, and calculating the measurement time delay value of the partial discharge source signal.
4. The partial discharge positioning error correction method of claim 3, wherein the theoretical time delay value of the partial discharge source signal is calculated by the formula:
wherein (x)i,yi,zi) Is the coordinate of sensor i, (x)k,yk,zk) As coordinates of the sensor k, τikThe time difference between the sensor i and the sensor k receiving the partial discharge ultrahigh frequency signal is shown, the (x, y, z) is the coordinate of a partial discharge source, and the upsilon is the propagation speed of the ultrahigh frequency signal in the medium.
5. The method for correcting the partial discharge positioning error according to claim 4, wherein establishing an error optimization neural network model of a measurement delay value and a theoretical delay value of a sampling point partial discharge source signal comprises:
and normalizing the measured delay value and the theoretical delay value, taking the measured delay value as input, taking the theoretical delay value as output, and constructing an error optimization neural network model.
6. The partial discharge positioning error correction method of claim 5, wherein training the error optimization neural network model, and optimizing parameters of the error optimization neural network model by using a particle swarm optimization algorithm to obtain the trained error optimization neural network model comprises:
setting the number of hidden layers and the value range and the initial value of the number of nodes of the error optimization neural network model;
optimizing the error optimization neural network model by adopting a particle swarm algorithm to obtain the optimal weight and threshold under the current hidden layer number and node number neural network structure;
training and testing input data under the current neural network structure to obtain and store a test result;
changing the number of hidden layers and the number of nodes in a value range, and if the number of hidden layers and the number of nodes do not meet the end condition, returning to the step of optimizing the error optimization neural network model by adopting a particle swarm algorithm until the end condition is met;
and selecting the error optimization neural network model with the highest prediction precision from the stored test results as the trained error optimization neural network model.
7. The method of claim 6, wherein the range of values of the number of hidden layers and the number of nodes is determined according to a preset empirical formula, the preset empirical formula being:
wherein L is the number of hidden layers and the number of nodes, m and n are the number of nodes of an input layer and an output layer respectively, and a is a positive number of 0-10.
8. The partial discharge positioning error correction method of claim 7, wherein the fitness function for optimizing the error optimization neural network model by using the particle swarm optimization is as follows:
wherein, yjIs a theoretical time delay value of a partial discharge source, y'jIs the output value of the neural network.
9. A partial discharge positioning error correction system, comprising:
the structure dividing unit is used for dividing the GIS structure into a plurality of equally spaced regions and extracting region sampling points;
the time delay measuring unit is used for collecting the partial discharge signals of the sampling points and calculating the measuring time delay value of the partial discharge source signals;
the theoretical time delay calculating unit is used for calculating a theoretical time delay value of the local discharge source signal according to the spatial position of the sampling point;
the modeling unit is used for establishing an error optimization neural network model of a measurement delay value and a theoretical delay value of a sampling point local discharge source signal;
the network training unit is used for training the error optimization neural network model and optimizing parameters of the error optimization neural network model by adopting a particle swarm algorithm to obtain a trained error optimization neural network model;
the time delay correction unit is used for collecting local discharge source signals at any position of the GIS structure, calculating a measurement time delay value, and inputting the measurement time delay value into a trained error optimization neural network model to obtain a corrected time delay value;
and the error correction unit is used for correcting the positioning error of the local discharge source according to the corrected time delay value.
10. The partial discharge positioning error correction system according to claim 9, wherein the structure dividing unit is specifically configured to:
the GIS structure is divided into a plurality of equally spaced areas, and a plurality of sampling points are uniformly extracted in each area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011569407.8A CN112578248A (en) | 2020-12-26 | 2020-12-26 | Partial discharge positioning error correction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011569407.8A CN112578248A (en) | 2020-12-26 | 2020-12-26 | Partial discharge positioning error correction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112578248A true CN112578248A (en) | 2021-03-30 |
Family
ID=75139869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011569407.8A Pending CN112578248A (en) | 2020-12-26 | 2020-12-26 | Partial discharge positioning error correction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112578248A (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102903104A (en) * | 2012-09-13 | 2013-01-30 | 杭州电子科技大学 | Subtractive clustering based rapid image segmentation method |
| CN105911497A (en) * | 2016-06-15 | 2016-08-31 | 国网北京市电力公司 | Correction method and device for partial discharge positioning error |
| CN106503790A (en) * | 2015-08-28 | 2017-03-15 | 余学飞 | A kind of Pressure wire temperature compensation of Modified particle swarm optimization neutral net |
| CN106503689A (en) * | 2016-11-14 | 2017-03-15 | 哈尔滨理工大学 | Neutral net local discharge signal mode identification method based on particle cluster algorithm |
| CN107271868A (en) * | 2017-06-29 | 2017-10-20 | 国家电网公司 | A kind of shelf depreciation time-delay calculation error compensating method based on multiple neural network |
| CN107422233A (en) * | 2017-06-21 | 2017-12-01 | 国家电网公司 | A kind of shelf depreciation time-delay calculation error compensating method and shelf depreciation alignment system |
| CN107703418A (en) * | 2017-08-30 | 2018-02-16 | 上海交通大学 | Shelf depreciation location error compensation method based on more radial base neural nets |
| CN107966638A (en) * | 2017-12-29 | 2018-04-27 | 国网北京市电力公司 | Method and apparatus, storage medium and the processor of correction error |
| CN110069310A (en) * | 2019-04-23 | 2019-07-30 | 北京小米移动软件有限公司 | Switch the method, apparatus and storage medium of desktop wallpaper |
| CN110385720A (en) * | 2019-07-26 | 2019-10-29 | 南京航空航天大学 | A kind of robot localization error compensating method based on deep neural network |
-
2020
- 2020-12-26 CN CN202011569407.8A patent/CN112578248A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102903104A (en) * | 2012-09-13 | 2013-01-30 | 杭州电子科技大学 | Subtractive clustering based rapid image segmentation method |
| CN106503790A (en) * | 2015-08-28 | 2017-03-15 | 余学飞 | A kind of Pressure wire temperature compensation of Modified particle swarm optimization neutral net |
| CN105911497A (en) * | 2016-06-15 | 2016-08-31 | 国网北京市电力公司 | Correction method and device for partial discharge positioning error |
| CN106503689A (en) * | 2016-11-14 | 2017-03-15 | 哈尔滨理工大学 | Neutral net local discharge signal mode identification method based on particle cluster algorithm |
| CN107422233A (en) * | 2017-06-21 | 2017-12-01 | 国家电网公司 | A kind of shelf depreciation time-delay calculation error compensating method and shelf depreciation alignment system |
| CN107271868A (en) * | 2017-06-29 | 2017-10-20 | 国家电网公司 | A kind of shelf depreciation time-delay calculation error compensating method based on multiple neural network |
| CN107703418A (en) * | 2017-08-30 | 2018-02-16 | 上海交通大学 | Shelf depreciation location error compensation method based on more radial base neural nets |
| CN107966638A (en) * | 2017-12-29 | 2018-04-27 | 国网北京市电力公司 | Method and apparatus, storage medium and the processor of correction error |
| CN110069310A (en) * | 2019-04-23 | 2019-07-30 | 北京小米移动软件有限公司 | Switch the method, apparatus and storage medium of desktop wallpaper |
| CN110385720A (en) * | 2019-07-26 | 2019-10-29 | 南京航空航天大学 | A kind of robot localization error compensating method based on deep neural network |
Non-Patent Citations (2)
| Title |
|---|
| 李公法 等: "《人工智能与计算智能及其应用》", 30 June 2020, 华中科技大学出版社, pages: 152 * |
| 王丽 著: "《智能算法的改进与案例分析》", 31 January 2019, 北京邮电大学出版社, pages: 17 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10096121B2 (en) | Human-shape image segmentation method | |
| CN109932627A (en) | A kind of localization method and system of GIS pressure test shelf depreciation | |
| CN104507050B (en) | Probabilistic type finger print matching method in a kind of WiFi indoor positionings | |
| CN108051779B (en) | TDOA-oriented positioning node optimization method | |
| CN110401978B (en) | Indoor positioning method based on neural network and particle filter multi-source fusion | |
| KR101268761B1 (en) | Method for deciding defect type of partial discharge and apparatus thereof | |
| CN112990330B (en) | User energy abnormal data detection method and device | |
| CN105491661A (en) | Improved Kalman filtering algorithm-based indoor positioning system and method | |
| CN110907810B (en) | Analog circuit single fault diagnosis method based on particle swarm algorithm | |
| CN114584230B (en) | Predictive channel modeling method based on countermeasure network and long-term and short-term memory network | |
| CN112560981A (en) | Training method, apparatus, device, program and storage medium for generating countermeasure model | |
| CN117590172A (en) | Partial discharge acoustic-electric combined positioning method, device and equipment applied to transformer | |
| CN113640712B (en) | Prediction method for vertical component of vertical induction magnetic field of ship | |
| CN106019098A (en) | Partial discharge positioning method, device and system | |
| CN111194000A (en) | Distance measurement method and system based on Bluetooth fusion hybrid filtering and neural network | |
| Qin et al. | A wireless sensor network location algorithm based on insufficient fingerprint information | |
| CN113449254A (en) | Method for analyzing monitoring stability of arbitrary net type deformation and method for determining position of monitoring point | |
| CN113189513A (en) | Ripple-based redundant power supply current sharing state identification method | |
| CN117289097A (en) | Power equipment partial discharge detection method, model training method, device and equipment | |
| CN112528566A (en) | Real-time air quality data calibration method and system based on AdaBoost training model | |
| CN112578248A (en) | Partial discharge positioning error correction method and system | |
| CN107132500A (en) | A kind of synchronous phasor measurement unit on-line calibration method and apparatus | |
| CN103942416B (en) | Voltage estimation method based on weighted node spanning tree | |
| CN111695501B (en) | Equipment soft fault detection method based on operating system kernel calling data | |
| CN108519579B (en) | WiFi fingerprint positioning method for analyzing optimal AP based on interval overlapping degree |
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: 20210330 |