CN114564559A - Method, device, equipment and storage medium for detecting power failure type based on traveling wave - Google Patents
Method, device, equipment and storage medium for detecting power failure type based on traveling wave Download PDFInfo
- Publication number
- CN114564559A CN114564559A CN202210031603.2A CN202210031603A CN114564559A CN 114564559 A CN114564559 A CN 114564559A CN 202210031603 A CN202210031603 A CN 202210031603A CN 114564559 A CN114564559 A CN 114564559A
- Authority
- CN
- China
- Prior art keywords
- traveling wave
- features
- principal component
- fusion
- representing
- 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 37
- 230000004927 fusion Effects 0.000 claims abstract description 73
- 238000001514 detection method Methods 0.000 claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims description 58
- 238000012545 processing Methods 0.000 claims description 24
- 230000009467 reduction Effects 0.000 claims description 21
- 238000000513 principal component analysis Methods 0.000 claims description 15
- 238000011176 pooling Methods 0.000 claims description 14
- 230000002087 whitening effect Effects 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 9
- 238000000354 decomposition reaction Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 230000008439 repair process Effects 0.000 abstract description 5
- 238000004458 analytical method Methods 0.000 abstract description 4
- 239000012634 fragment Substances 0.000 abstract description 3
- 230000000875 corresponding effect Effects 0.000 description 22
- 238000005070 sampling Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 238000012952 Resampling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000002759 z-score normalization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
-
- 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
- G01R31/085—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution lines, e.g. overhead
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a method, a device, equipment and a storage medium for detecting power failure types based on traveling waves. Acquiring traveling wave data acquired by a traveling wave acquisition device, extracting global features representing global information from the traveling wave data, intercepting a plurality of segments with preset lengths from the traveling wave data, extracting correlation features representing the correlation of the plurality of segments, fusing the global features and the correlation features to obtain fusion features, inputting the fusion features into a pre-trained clustering code book for retrieval, and determining the type of the power failure. The overall characteristic representing the overall information of the traveling wave data and the local characteristic representing the correlation of the multi-section fragments are combined for analysis, so that the detection accuracy of the traveling wave data with low signal-to-noise ratio can be improved, and the detection accuracy is improved. When a power line fails, the fault type is determined quickly and accurately, reference is provided for power line repair, and economic loss is reduced.
Description
Technical Field
The embodiment of the invention relates to a power failure detection technology, in particular to a method, a device, equipment and a storage medium for detecting power failure types based on traveling waves.
Background
With the rapid development of power systems, the power transmission line develops a cable-overhead line hybrid power transmission line on the basis of the original cable and overhead power transmission line, and the application is more and more extensive. The ultra-high voltage cable-overhead line hybrid line can span large water channels and straits and can directly supply power to centers of large cities and industrial areas. Meanwhile, due to the limitation of urban space and planning, the cable-overhead hybrid line is more and more widely applied in cities. The cable-overhead line hybrid line is also applied to low-current power transmission systems with neutral points not directly grounded, such as railway signal power supply systems.
Once the power grid fails, a large amount of alarm information generated by each level of automatic devices can rapidly rush into a power system dispatching center, so that a dispatcher cannot rapidly judge the failure reason in a short time, serious potential safety hazards and huge economic loss can be brought to the normal operation of the power grid, and the timely determination of the failure reason has great significance for the safe and economic operation of the power grid.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for detecting a power fault type based on a traveling wave, which are used for quickly and accurately determining the fault type, providing reference for power first-aid repair and reducing economic loss.
In a first aspect, an embodiment of the present invention provides a method for detecting a power failure type based on a traveling wave, including:
acquiring traveling wave data acquired by a traveling wave acquisition device;
extracting global features representing global information from the traveling wave data;
intercepting a plurality of segments with preset lengths from the traveling wave data;
extracting correlation characteristics representing the relevance of a plurality of segments;
fusing the global features and the associated features to obtain fused features;
and inputting the fusion characteristics into a pre-trained clustering code book for retrieval, and determining the type of the power failure.
Optionally, extracting global features representing global information from the traveling wave data includes:
extracting a plurality of characteristics with different scales from the traveling wave data;
and fusing the characteristics of various different scales to obtain the global characteristics.
Optionally, extracting relevant features characterizing the relevance of the segments includes:
inputting a plurality of segments into a preset encoder of a Transformer model for processing to obtain a characteristic matrix;
performing pooling treatment on the characteristic matrix to obtain a pooled matrix;
mapping the pooling matrix to a correlation characteristic characterizing a correlation of segments.
Optionally, inputting the fusion features into a pre-trained clustering codebook for retrieval, and determining the power failure type, including:
performing principal component analysis on the fusion features to find principal component features of principal components for representing the fusion features;
whitening the principal component characteristics to obtain dimension reduction characteristics of effective information for representing the fusion characteristics;
calculating the similarity between each principal component of the dimensionality reduction features and N cluster centers in a clustering code book to obtain N similarity values corresponding to each principal component;
determining a power failure type based on the similarity value.
Optionally, performing principal component analysis on the fusion feature to find principal component features of principal components used for characterizing the fusion feature, including:
calculating covariance matrixes of all elements in the fusion characteristics;
singular value decomposition is carried out on the covariance matrix to obtain a plurality of characteristic components of the covariance matrix;
selecting the first k characteristic components as principal components;
and projecting the value of each element in the fusion feature to the principal component to obtain a principal component feature of the principal component for representing the fusion feature.
Optionally, the whitening processing is performed on the principal component feature to obtain a dimension reduction feature of the effective information for characterizing the fusion feature, including:
and carrying out standard deviation normalization processing on the data of each of the k principal components of the principal component characteristics to obtain dimension reduction characteristics of effective information for representing the principal component characteristics.
Optionally, determining the power failure type based on the similarity value includes:
taking the maximum value of the N similarity values corresponding to each principal component to obtain k maximum similarity values corresponding to k principal components;
taking the maximum value of the k maximum similarity values as a target similarity value;
and taking the fault type corresponding to the cluster center in the cluster codebook corresponding to the target similarity value as a power fault type.
In a second aspect, an embodiment of the present invention further provides a device for detecting a power failure type based on a traveling wave, including:
the traveling wave data acquisition module is used for acquiring traveling wave data acquired by the traveling wave acquisition device;
the global feature extraction module is used for extracting global features representing global information from the traveling wave data;
the section intercepting module is used for intercepting a plurality of sections with preset lengths from the traveling wave data;
the relevant feature extraction module is used for extracting relevant features representing the relevance of the segments;
the feature fusion module is used for fusing the global features and the associated features to obtain fusion features;
and the fault type determining module is used for inputting the fusion characteristics into a pre-trained clustering code book for retrieval and determining the power fault type.
In a third aspect, an embodiment of the present invention further provides a computer device, including:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a traveling wave based power failure type detection method as provided by the first aspect of the invention.
In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the traveling wave-based power fault type detection method according to the first aspect of the present invention.
The method for detecting the type of the power fault based on the traveling wave provided by the embodiment of the invention comprises the following steps: acquiring traveling wave data acquired by a traveling wave acquisition device, extracting global features representing global information from the traveling wave data, intercepting a plurality of segments with preset lengths from the traveling wave data, extracting correlation features representing the correlation of the plurality of segments, fusing the global features and the correlation features to obtain fusion features, inputting the fusion features into a pre-trained clustering code book for retrieval, and determining the type of the power failure. The overall characteristic representing the overall information of the traveling wave data and the local characteristic representing the correlation of the multi-section fragments are combined for analysis, so that the detection accuracy of the traveling wave data with low signal-to-noise ratio can be improved, and the detection accuracy is improved. When a power line fails, the fault type is determined quickly and accurately, reference is provided for power line repair, and economic loss is reduced.
Drawings
Fig. 1 is a flowchart of a method for detecting a power failure type based on a traveling wave according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for detecting a power failure type based on a traveling wave according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a traveling wave-based power failure type detection apparatus according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computer device according to a fourth embodiment of the present invention,
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a method for detecting a power failure type based on a traveling wave according to an embodiment of the present invention, where this embodiment is applicable to power failure detection, and the method can be executed by a device for detecting a power failure type based on a traveling wave according to an embodiment of the present invention, where the device can be implemented by software and/or hardware, and is generally configured in a computer device, as shown in fig. 1, where the method specifically includes the following steps:
and S101, acquiring traveling wave data acquired by the traveling wave acquisition device.
In the embodiment of the invention, the traveling wave acquisition device can be intelligent electric power monitoring equipment, and the traveling wave acquisition device is distributed at each electric tower and is directly or indirectly connected with the power transmission line for acquiring traveling wave data. Specifically, when the power transmission line sends a fault such as a short circuit to ground or a lightning strike, the change of the line current and voltage (i.e., traveling wave data) caused by the fault is collected by the traveling wave collection device. The embodiment of the invention adopts a distributed installation mode, divides a long line into a plurality of short lines, shortens the monitoring distance, solves the problem of signal attenuation, improves the monitoring sensitivity and can realize fault positioning in high-resistance grounding. In addition, each row of wave acquisition devices can be mutually standby, and the stability of the system is improved.
It should be noted that, in the embodiment of the present invention, a network model used for extracting the global feature is not limited as long as the global feature representing the global information can be obtained.
In some embodiments of the present invention, the raw data may be further preprocessed to obtain traveling wave data. For example, the original data is first resampled, and in an exemplary embodiment of the present invention, the resampling sampling rate is 200Hz, that is, the original electrocardiographic signal is sampled 200 times in 1 second, and 200 sampling points are acquired in 1 second, so that the resampled data is obtained. And receiving, namely filtering and denoising the data after resampling, and removing noise in the original data to obtain filtered data with required frequency. Further, in order to accelerate the speed of solving the optimal solution by gradient descent in the subsequent network model processing process, the filtered data may be standardized, that is, the data is scaled to fall into a specific interval. In the embodiment of the present invention, the specific method for implementing the standardization is not limited. Illustratively, in one embodiment, z-score normalization may be employed to process the filtered signal into a "zero mean, one variance" normalized signal.
And S102, extracting global features representing global information from the traveling wave data.
In the embodiment of the invention, the collected traveling wave data can be input into a pre-trained global feature extraction network for processing, and the global features for representing the global information are extracted. The global characteristics refer to that all characteristics of the traveling wave data are used for representing the traveling wave data, and the traveling wave data are integral attributes of the traveling wave data. Common global features include peaks, valleys, zeros, inflection points, etc. The global feature is a low-level visual feature, so the global feature has the characteristics of good invariance, simple calculation, visual representation and the like.
And S103, intercepting a plurality of segments with preset lengths from the traveling wave data.
In the embodiment of the invention, after the traveling wave is obtained, a plurality of segments with preset lengths are intercepted from traveling wave data. Illustratively, in embodiments of the present invention, the length of the truncated segment may be 0.5 seconds.
And S104, extracting the correlation characteristics representing the correlation of the multiple segments.
In the embodiment of the invention, the relevance mining is carried out on the intercepted multiple segments, and the relevance characteristics for representing the relevance of the multiple segments are extracted from the multiple segments.
And S105, fusing the global features and the associated features to obtain fused features.
In the embodiment of the invention, the obtained global features and the associated features are subjected to fusion processing to obtain fusion features. Illustratively, the global feature and the associated feature may be concatenated (concatenated), increasing feature dimensions, and thus improving detection accuracy.
And S106, inputting the fusion characteristics into a pre-trained clustering code book for retrieval, and determining the type of the power failure.
In the embodiment of the invention, a large number of traveling wave data samples are adopted in advance to carry out cluster coding to obtain a cluster code book (PQ Codebook). And then, inputting the fusion characteristics into a pre-trained clustering code book for retrieval, wherein the type of the power failure is determined.
Specifically, the obtained fusion characteristics of the traveling wave data are input into a clustering code book for retrieval, the similarity between the fusion characteristics and each cluster center in the clustering code book is calculated, and the power fault type is determined based on the similarity. The greater the similarity with a certain cluster center, the greater the probability that there is a fault type corresponding to the cluster center.
The method for detecting the type of the power fault based on the traveling wave provided by the embodiment of the invention comprises the following steps: acquiring traveling wave data acquired by a traveling wave acquisition device, extracting global features representing global information from the traveling wave data, intercepting a plurality of segments with preset lengths from the traveling wave data, extracting correlation features representing the correlation of the plurality of segments, fusing the global features and the correlation features to obtain fusion features, inputting the fusion features into a pre-trained clustering code book for retrieval, and determining the type of the power failure. The overall characteristic representing the overall information of the traveling wave data and the local characteristic representing the correlation of the multi-section fragments are combined for analysis, so that the detection accuracy of the traveling wave data with low signal-to-noise ratio can be improved, and the detection accuracy is improved. When the power line breaks down, the fault type is determined quickly and accurately, reference is provided for power line repair, and economic loss is reduced.
Example two
Fig. 2 is a flowchart of a method for detecting a power failure type based on a traveling wave according to a second embodiment of the present invention, which is detailed based on the first embodiment, and describes in detail the detailed processes of the steps in the first embodiment, as shown in fig. 2, the method includes:
s201, acquiring traveling wave data acquired by the traveling wave acquisition device.
As described in the foregoing embodiment, in the embodiment of the present invention, the traveling wave acquisition device is installed in a distributed installation manner, and traveling wave data acquired by the traveling wave acquisition device is acquired.
S202, extracting a plurality of features with different scales from the traveling wave data.
The small-scale features have higher resolution and contain more position and detail information, but have lower semanteme and more noise due to less convolution. Features with large scale have stronger semantic information, but the resolution is very low, and the perception capability of details is poor. In the embodiment of the invention, the characteristics of various scales are extracted from the traveling wave data, and the detail information and the semantic information are considered.
And S203, fusing the characteristics of various different scales to obtain the global characteristics.
And performing feature fusion on the obtained features with different scales to obtain a fused global feature. The fused global features give consideration to detailed information and semantic information, and the accuracy of power failure type detection is improved.
In an embodiment of the present invention, the feature pyramid network or the U-like network may be used to implement the above feature extraction and fusion processes of multiple scales, which is not limited herein.
And S204, intercepting a plurality of segments with preset lengths from the traveling wave data.
In the embodiment of the invention, the first sampling point of the traveling wave data is taken as a reference point, and one sampling point is selected as a target point every other sampling point with a preset number. Then, taking the target point as a center, and cutting a segment with a preset length.
Illustratively, in the embodiment of the present invention, starting from the first sampling point, one sampling point is selected as the target point every 100 sampling points, i.e., the 100 th, 200 th and 300 th 300 … th sampling points are selected as the target points. In the embodiment of the present invention, the length of the intercepted segment may be 0.5 second, and the segment with the preset length is 100 sampling points centered on the target point according to the sampling rate of 200 Hz.
S205, inputting the multiple segments into a preset Transformer model encoder for processing to obtain a feature matrix.
In the embodiment of the invention, the correlation of the multi-segment segments is mined through an encoder of a Transformer model, so that the local characteristics for representing the correlation of the multi-segment segments are obtained.
The encoder of the Transformer model comprises M coding blocks which are stacked in sequence, wherein M is a positive integer greater than or equal to 2. Exemplarily, M is 6 in the embodiment of the present invention. The processing procedure of the encoder of the Transformer model is as follows:
a plurality of segments (each segment comprises 100 sampling points) are input into a first encoding block as an input matrix of the first encoding block to be processed. And taking the output matrix of the previous coding block as the input matrix of the next coding block, and so on until the output matrix of the last coding block is obtained as the feature matrix.
Wherein each coding block processes the input matrix based on a multi-head attention mechanism. Specifically, the input matrix is processed based on a multi-head attention mechanism to obtain an attention matrix. And adding the attention matrix and the input matrix of the coding block to obtain a first fusion matrix. The first normalization layer performs layer normalization processing on the first fusion matrix to obtain a normalization matrix. And carrying out nonlinear transformation on the normalized matrix to obtain a transformation matrix. And adding the transformation matrix and the normalization matrix to obtain a second fusion matrix. And carrying out layer normalization processing on the second fusion matrix to obtain an output matrix of the coding block.
S206, performing pooling processing on the feature matrix to obtain a pooled matrix.
And inputting the characteristic matrix output by the encoder of the Transformer model into a pooling layer for processing, and performing global average pooling processing on the characteristic matrix by the pooling layer to obtain a pooled matrix.
And S207, mapping the pooling matrix into correlation characteristics representing the correlation of the multi-segment segments.
In an embodiment of the invention, the pooling matrix is input into a fully-connected layer, which maps the pooling matrix to correlation features characterizing the correlation of the multi-segment segments.
And S208, fusing the global features and the associated features to obtain fused features.
In the embodiment of the invention, the global feature and the associated feature are subjected to feature splicing in a dimension space to obtain a fusion feature, wherein the dimension of the fusion feature is the sum of the dimensions of the global feature and the associated feature.
And S209, performing principal component analysis on the fusion characteristics to find principal component characteristics of principal components for characterizing the fusion characteristics.
Principal Component Analysis (PCA) is a common data Analysis method, is commonly used for dimensionality reduction of high-dimensional data, and can be used for extracting main characteristic components of the data. The main idea of PCA is to map n-dimensional features onto k-dimensions, which are completely new orthogonal features, also called principal components, and k-dimensional features reconstructed on the basis of the original n-dimensional features.
In the embodiment of the invention, principal component analysis is carried out on the fusion characteristics, principal component characteristics of principal components used for representing the fusion characteristics are found, and the dimensionality of the principal component characteristics is smaller than that of the fusion characteristics.
Illustratively, in the embodiment of the present invention, the principal component analysis of the fusion features includes the following sub-steps:
1. the covariance matrix of all elements in the fused feature is calculated.
For high dimensional data, we use covariance to represent the degree of dispersion of the data. Covariance may represent the correlation of two variables. In order to make two variables represent as much original information as possible, we want no linear correlation between them, because correlation means that two variables are not completely independent, and there must be information represented repeatedly.
The covariance matrix is used to represent the correlation of each element in the fused feature with any other element. The covariance matrix is calculated as follows:
where m is the total number of elements in the fused feature, xiFor the ith element in the fused feature, (x)i)TIs xiThe transposing of (1).
2. And carrying out singular value decomposition on the covariance matrix to obtain a plurality of characteristic components of the covariance matrix.
After the covariance matrix is solved, singular value decomposition is carried out on the covariance matrix, and each column in the obtained U vector is the characteristic component of the covariance matrix. Specifically, the decomposition process is as follows:
where U is an orthogonal matrix, U1…unAre characteristic components.
3. For selecting the first k feature components as principal components.
Specifically, k feature components (u) to be ranked first1…uk) As the principal component.
4. And projecting the value of each element in the fusion feature to the principal component to obtain a principal component feature of the principal component for representing the fusion feature.
Specifically, the values of the elements in the fusion feature are projected to the principal components, so that the dimensionality reduction of the fusion feature is realized, and the principal component feature of the principal component for representing the fusion feature is obtained.
And S210, whitening the principal component characteristics to obtain dimension reduction characteristics of effective information for representing the fusion characteristics.
Since adjacent elements in the traveling wave data are correlated, that is, elements in the principal component features are correlated, some information in the principal component features is redundant, and whitening is to reduce redundancy of the input. Specifically, in the embodiment of the present invention, whitening (whitening) processing is performed on the principal component feature. Whitening of data must satisfy two conditions: firstly, the correlation among different characteristics is minimum and is close to 0; the second is that the variances of all features are equal (not necessarily 1).
As described above, after PCA processing, principal component features are obtained, which include k principal components, and the principal components are independent and uncorrelated. The first condition for whitening is satisfied, and therefore, it is only necessary to divide each principal component in the principal component feature by the standard deviation to obtain a variance of 1 for each principal component, that is, equal variances. Therefore, the data of each of the k principal components of the local feature is normalized by the standard deviation to obtain the dimension reduction feature of the effective information for representing the fusion feature.
S211, calculating the similarity between each principal component of the dimensionality reduction features and N cluster centers in a cluster code book to obtain N similarity values corresponding to each principal component.
In the embodiment of the invention, a large number of traveling wave data samples are adopted in advance for cluster coding, and a cluster code book based on negative samples is obtained.
Illustratively, a large number of traveling wave data samples are obtained first, and global features for characterizing global information are extracted from each traveling wave data sample. Then, intercepting a plurality of segments with preset lengths from the traveling wave data samples, extracting correlation characteristics representing the correlation of the plurality of segments, then fusing the global characteristics and the correlation characteristics to obtain fusion characteristics, and finally performing cluster coding on the fusion characteristics of a large number of traveling wave data samples to obtain a cluster code book. In the embodiment of the present invention, the clustering algorithm may adopt a K-MEANS clustering algorithm or other clustering algorithms, and the embodiment of the present invention is not limited herein. Clustering a large number of samples through clustering codes to obtain a clustering code book comprising a plurality of fault categories, wherein the samples of each fault category have similar characteristics and are represented by clusters. Specifically, the obtaining process of the dimension reduction feature of the sample may refer to the obtaining process of the dimension reduction feature of the chest wave data, and the embodiment of the present invention is not described herein again.
Illustratively, the dimensionality reduction features of the sample include 8 principal components (i.e., k-8), each of which is represented by 128 dimensions. Assuming that the total number of samples is 50000, we get a matrix of 50k × 8 × 128. Then, 50000 samples are clustered on the 8 principal components, and assuming that the number of cluster clusters is 256, 256 cluster centers (i.e., N is 256) are obtained, i.e., a 8 × 256 × 128 cluster codebook is obtained. This clustering codebook shows that the sample has 256 cluster centers in 8 principal components, each cluster center being represented by 128 dimensions.
And calculating the similarity between each principal component of the dimension reduction feature and 256 cluster centers in the clustering codebook, namely obtaining 256 similarity values for each principal component. Specifically, the similarity may be a constine similarity, and the greater the similarity with a certain cluster center, the greater the probability that there is a fault type corresponding to the cluster center is.
And S212, determining the type of the power failure based on the similarity value.
For example, in the embodiment of the present invention, a maximum value of the N similarity values corresponding to each principal component is taken to obtain k maximum similarity values corresponding to k principal components. Illustratively, for each principal component, the maximum value of 256 similarity values is taken, and thus, 8 maximum similarity values corresponding to 8 principal components are obtained. In the embodiment of the invention, the maximum value is selected
Then, the maximum value of the k maximum similarity values is taken as the target similarity value. Illustratively, the maximum value is selected from the 8 maximum similarity values as the target similarity value. In the embodiment of the present invention, the selection manner of the maximum value and the minimum value is not limited, and for example, a sorting method or a sliding window method may be adopted.
And taking the fault type corresponding to the cluster center in the cluster codebook corresponding to the target similarity value as the power fault type.
According to the method for detecting the power failure type based on the traveling wave, provided by the embodiment of the invention, the global characteristics for representing the global information of the traveling wave data and the local characteristics for representing the correlation of the multi-segment segments are combined for analysis, so that the detection accuracy of the traveling wave data with low signal-to-noise ratio can be improved, and the detection accuracy is improved. When the power line breaks down, the fault type is determined quickly and accurately, reference is provided for power line repair, and economic loss is reduced. And performing principal component analysis and whitening processing on the fusion features to realize the dimensionality reduction of the fusion features, and on the premise of retaining the effective information of the fusion features, reducing the workload of subsequent retrieval in a clustering code book, saving computing resources and improving the retrieval efficiency.
EXAMPLE III
Fig. 3 is a schematic structural diagram of a traveling wave-based power failure type detection apparatus according to a third embodiment of the present invention, as shown in fig. 3, the apparatus includes:
a traveling wave data acquisition module 301, configured to acquire traveling wave data acquired by a traveling wave acquisition device;
a global feature extraction module 302, configured to extract a global feature representing global information from the traveling wave data;
a segment intercepting module 303, configured to intercept multiple segments of segments with preset lengths from the traveling wave data;
an associated feature extraction module 304, configured to extract an associated feature that characterizes a correlation of the segments;
a feature fusion module 305, configured to fuse the global feature and the associated feature to obtain a fusion feature;
and a fault type determining module 306, configured to input the fusion feature into a pre-trained cluster codebook for retrieval, and determine a power fault type.
In some embodiments of the present invention, the global feature extraction module 302 comprises:
the multi-feature extraction submodule is used for extracting features of various different scales from the traveling wave data;
and the feature fusion submodule is used for fusing the features of various different scales to obtain the global features.
In some embodiments of the present invention, the associated feature extraction module 304 comprises:
the coding submodule is used for inputting the segments into a coder of a preset Transformer model for processing to obtain a characteristic matrix;
the pooling submodule is used for pooling the characteristic matrix to obtain a pooled matrix;
a mapping sub-module for mapping the pooling matrix to a correlation characteristic characterizing a correlation of segments.
In some embodiments of the present invention, the fault type determination module 306 comprises:
the principal component analysis submodule is used for carrying out principal component analysis on the fusion characteristic and finding out principal component characteristics of principal components for representing the fusion characteristic;
the whitening processing submodule is used for whitening the principal component characteristics to obtain dimension reduction characteristics of effective information for representing the fusion characteristics;
the similarity operator module is used for calculating the similarity between each principal component of the dimensionality reduction features and N cluster centers in a clustering code book to obtain N similarity values corresponding to each principal component;
and the fault type determination submodule is used for determining the power fault type based on the similarity value.
In some embodiments of the invention, the principal component analysis submodule comprises:
the covariance matrix calculation unit is used for calculating covariance matrixes of all elements in the fusion characteristics;
the singular value decomposition unit is used for carrying out singular value decomposition on the covariance matrix to obtain a plurality of characteristic components of the covariance matrix;
a principal component determination unit for selecting the top k feature components as principal components;
and the projection unit is used for projecting the values of all the elements in the fusion feature to the principal component to obtain a principal component feature of the principal component for representing the fusion feature.
In some embodiments of the invention, the whitening processing sub-module comprises:
and the normalization processing unit is used for carrying out standard deviation normalization processing on the data of each of the k principal components of the principal component characteristics to obtain the dimension reduction characteristics of the effective information for representing the principal component characteristics.
In some embodiments of the invention, the fault type determination submodule comprises:
the first selecting unit is used for selecting the maximum value of the N similarity values corresponding to each principal component to obtain k maximum similarity values corresponding to k principal components;
the second selection unit is used for selecting the maximum value of the k maximum similarity values as a target similarity value;
and the fault type determining unit is used for taking the fault type corresponding to the cluster center in the cluster code book corresponding to the target similarity value as the power fault type.
The power failure type detection device based on the traveling wave can execute the power failure type detection method based on the traveling wave provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Example four
A fourth embodiment of the present invention provides a computer device, and fig. 4 is a schematic structural diagram of the computer device provided in the fourth embodiment of the present invention, as shown in fig. 4, the computer device includes:
a processor 401, a memory 402, a communication module 403, an input device 404, and an output device 405; the number of the processors 401 in the computer device may be one or more, and one processor 401 is taken as an example in fig. 4; the processor 401, the memory 402, the communication module 403, the input device 404 and the output device 405 in the computer apparatus may be connected by a bus or other means, and fig. 4 illustrates an example of connection by a bus. The processor 401, memory 402, communication module 403, input device 404, and output device 405 described above may be integrated on a computer device.
The memory 402, which is a computer-readable storage medium, may be used to store software programs, computer-executable programs, and modules, such as the modules corresponding to the traveling wave-based power failure type detection method in the above embodiments. The processor 401 executes various functional applications and data processing of the computer device by executing software programs, instructions and modules stored in the memory 402, that is, implements the traveling wave-based power failure type detection method described above.
The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the microcomputer, and the like. Further, the memory 402 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 402 may further include memory located remotely from the processor 401, which may be connected to a computer device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
And a communication module 403, configured to establish a connection with an external device (e.g., an intelligent terminal), and implement data interaction with the external device. The input device 404 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function controls of the computer apparatus.
The computer device provided by this embodiment may execute the method for detecting a power failure type based on a traveling wave provided by any of the above embodiments of the present invention, and has corresponding functions and beneficial effects.
EXAMPLE five
An embodiment of the present invention provides a storage medium containing computer-executable instructions, where the storage medium stores a computer program, and the computer program, when executed by a processor, implements a traveling wave-based power failure type detection method according to any of the above embodiments of the present invention, where the method includes:
acquiring traveling wave data acquired by a traveling wave acquisition device;
extracting global features representing global information from the traveling wave data;
intercepting a plurality of segments with preset lengths from the traveling wave data;
extracting correlation characteristics representing the relevance of a plurality of segments;
fusing the global features and the associated features to obtain fused features;
and inputting the fusion characteristics into a pre-trained clustering code book for retrieval, and determining the type of the power failure.
It should be noted that, as for the apparatus, the computer device and the storage medium embodiment, since they are basically similar to the method embodiment, the description is relatively simple, and in relation to the description, reference may be made to part of the description of the method embodiment.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, and the computer software product may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions to enable a computer device (which may be a robot, a personal computer, a server, or a network device) to execute the traveling wave-based power failure type detection method according to any embodiment of the present invention.
It should be noted that, in the above apparatus, each module, sub-module, and unit included in the apparatus are merely divided according to functional logic, but are not limited to the above division, as long as the corresponding function can be implemented; in addition, the specific names of the functional modules are only for convenience of distinguishing from each other and are not used for limiting the protection scope of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by suitable instruction execution devices. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description of the specification, reference to the description of "one embodiment," "some embodiments," "an example," "a specific example," or "some examples" or the like means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method for detecting the type of a power fault based on traveling waves is characterized by comprising the following steps:
acquiring traveling wave data acquired by a traveling wave acquisition device;
extracting global features representing global information from the traveling wave data;
intercepting a plurality of segments with preset lengths from the traveling wave data;
extracting correlation characteristics representing the relevance of a plurality of segments;
fusing the global features and the associated features to obtain fused features;
and inputting the fusion characteristics into a pre-trained clustering code book for retrieval, and determining the type of the power failure.
2. The traveling wave-based power fault type detection method of claim 1, wherein extracting global features characterizing global information from the traveling wave data comprises:
extracting a plurality of characteristics with different scales from the traveling wave data;
and fusing the characteristics of various different scales to obtain the global characteristics.
3. The traveling wave-based power fault type detection method according to claim 1, wherein extracting correlation features characterizing correlation of a plurality of segments comprises:
inputting a plurality of segments into a preset encoder of a Transformer model for processing to obtain a characteristic matrix;
performing pooling treatment on the characteristic matrix to obtain a pooled matrix;
mapping the pooling matrix to a correlation characteristic characterizing a correlation of segments.
4. The traveling-wave-based power failure type detection method according to any one of claims 1 to 3, wherein the step of inputting the fusion features into a pre-trained cluster code book for retrieval to determine the power failure type comprises:
performing principal component analysis on the fusion features to find principal component features of principal components for representing the fusion features;
whitening the principal component characteristics to obtain dimension reduction characteristics of effective information for representing the fusion characteristics;
calculating the similarity between each principal component of the dimensionality reduction features and N cluster centers in a clustering code book to obtain N similarity values corresponding to each principal component;
determining a power failure type based on the similarity value.
5. The traveling wave-based power fault type detection method according to claim 4, wherein performing principal component analysis on the fused feature to find principal component features for characterizing principal components of the fused feature comprises:
calculating covariance matrixes of all elements in the fusion characteristics;
singular value decomposition is carried out on the covariance matrix to obtain a plurality of characteristic components of the covariance matrix;
selecting the first k characteristic components as principal components;
and projecting the value of each element in the fusion feature to the principal component to obtain a principal component feature of the principal component for representing the fusion feature.
6. The traveling wave-based power fault type detection method according to claim 4, wherein the whitening processing is performed on the principal component features to obtain dimension reduction features used for representing effective information of the fusion features, and the method comprises the following steps:
and carrying out standard deviation normalization processing on the data of each of the k principal components of the principal component characteristics to obtain dimension reduction characteristics of effective information for representing the principal component characteristics.
7. The traveling-wave based power fault type detection method of claim 4, wherein determining a power fault type based on the similarity value comprises:
taking the maximum value of the N similarity values corresponding to each principal component to obtain k maximum similarity values corresponding to k principal components;
taking the maximum value of the k maximum similarity values as a target similarity value;
and taking the fault type corresponding to the cluster center in the cluster code book corresponding to the target similarity value as a power fault type.
8. A traveling wave based power failure type detection apparatus, comprising:
the traveling wave data acquisition module is used for acquiring traveling wave data acquired by the traveling wave acquisition device;
the global feature extraction module is used for extracting global features representing global information from the traveling wave data;
the section intercepting module is used for intercepting a plurality of sections with preset lengths from the traveling wave data;
the relevant feature extraction module is used for extracting relevant features representing the relevance of the segments;
the feature fusion module is used for fusing the global features and the associated features to obtain fusion features;
and the fault type determining module is used for inputting the fusion characteristics into a pre-trained clustering code book for retrieval, and determining the power fault type.
9. A computer device, comprising:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the traveling wave based power fault type detection method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a travelling wave based power failure type detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210031603.2A CN114564559A (en) | 2022-01-12 | 2022-01-12 | Method, device, equipment and storage medium for detecting power failure type based on traveling wave |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210031603.2A CN114564559A (en) | 2022-01-12 | 2022-01-12 | Method, device, equipment and storage medium for detecting power failure type based on traveling wave |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114564559A true CN114564559A (en) | 2022-05-31 |
Family
ID=81711254
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210031603.2A Pending CN114564559A (en) | 2022-01-12 | 2022-01-12 | Method, device, equipment and storage medium for detecting power failure type based on traveling wave |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114564559A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117149846A (en) * | 2023-08-16 | 2023-12-01 | 湖北中恒电测科技有限公司 | Power data analysis method and system based on data fusion |
| CN118228772B (en) * | 2024-05-24 | 2024-07-16 | 昆明理工大学 | Distance measurement method, framework, system and medium for actually measured traveling wave of power transmission line |
-
2022
- 2022-01-12 CN CN202210031603.2A patent/CN114564559A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117149846A (en) * | 2023-08-16 | 2023-12-01 | 湖北中恒电测科技有限公司 | Power data analysis method and system based on data fusion |
| CN117149846B (en) * | 2023-08-16 | 2024-05-24 | 上海永天科技股份有限公司 | Power data analysis method and system based on data fusion |
| CN118228772B (en) * | 2024-05-24 | 2024-07-16 | 昆明理工大学 | Distance measurement method, framework, system and medium for actually measured traveling wave of power transmission line |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112016602B (en) | Method, equipment and storage medium for analyzing correlation between power grid fault cause and state quantity | |
| CN115098705B (en) | Network security event analysis method and system based on knowledge graph reasoning | |
| CN116148679B (en) | Battery health state prediction method and related device | |
| CN110232499A (en) | A kind of power distribution network information physical side method for prewarning risk and system | |
| CN114024829B (en) | Fault maintenance method, device, equipment and storage medium for power communication network | |
| CN116614177B (en) | Optical fiber state multidimensional parameter monitoring system | |
| CN114281864A (en) | Correlation analysis method for power network alarm information | |
| CN116610983B (en) | Abnormality analysis method and system for air purification control system | |
| CN107784392A (en) | A kind of the defects of transmission line of electricity based on machine learning Forecasting Methodology and device | |
| CN112464439A (en) | Three-layer data mining-based power system fault type analysis method and device | |
| CN114564559A (en) | Method, device, equipment and storage medium for detecting power failure type based on traveling wave | |
| CN112528113A (en) | Terminal user dividing method and system based on power supply reliability multi-dimensional big data | |
| CN116192668A (en) | Space information network performance prediction method based on virtualized resource evaluation | |
| CN111401218A (en) | Smart city monitoring method and system | |
| CN117592595A (en) | Method and device for building and predicting load prediction model of power distribution network | |
| CN117117780A (en) | Circuit breaker anti-blocking method and system based on secondary information fusion of transformer substation | |
| CN117495422A (en) | Cost management system and method based on power communication network construction | |
| CN113128006A (en) | High-voltage power distribution method and device based on neural network | |
| CN114297453A (en) | Alarm prediction method, alarm prediction device, electronic equipment and storage medium | |
| CN113051006B (en) | Auxiliary configuration method and system based on application service and relation vectorization | |
| CN116126931B (en) | Power consumption data mining method, device and system for power distribution area and storage medium | |
| CN115168597A (en) | Regulation and control system operation event rule analysis method and device, medium and equipment | |
| CN113569904B (en) | Bus wiring type identification method, system, storage medium and computing device | |
| CN116821718A (en) | Non-invasive industrial load identification method and system based on Mean-shift clustering and twin network | |
| CN114338187B (en) | Terminal safety detection method and device based on decision tree |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |