CN112001246A - Partial discharge type identification method and device based on singular value decomposition - Google Patents
Partial discharge type identification method and device based on singular value decomposition Download PDFInfo
- Publication number
- CN112001246A CN112001246A CN202010698430.0A CN202010698430A CN112001246A CN 112001246 A CN112001246 A CN 112001246A CN 202010698430 A CN202010698430 A CN 202010698430A CN 112001246 A CN112001246 A CN 112001246A
- Authority
- CN
- China
- Prior art keywords
- partial discharge
- singular value
- neural network
- value decomposition
- signal
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- 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
- G01R31/1227—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 of components, parts or materials
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Evolutionary Biology (AREA)
- Signal Processing (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Testing Relating To Insulation (AREA)
Abstract
The invention discloses a partial discharge type identification method and a partial discharge type identification device based on singular value decomposition, wherein the method comprises the following steps: s1, acquiring known partial discharge types, and establishing a partial discharge type library; s2, performing wavelet packet denoising pretreatment on the local discharge signal, intercepting a signal in a denoised local discharge signal generation region to construct a Hankel matrix, and then performing singular value decomposition; s3, selecting energy percentages of proper number of singular values as characteristic quantities T, and dividing each type of partial discharge characteristic quantities into training samples TtrainAnd test sample Ttest(ii) a S4 reaction of TtrainTraining to obtain an optimal BP neural network as the input of the BP neural network optimized by the genetic algorithm; s5 reaction of TtestAnd obtaining the type of the partial discharge signal as the input of the trained BP neural network. The invention passes through the intercepting bureauThe partial discharge generation area is subjected to singular value decomposition, a small number of singular values can be used for representing waveform information of partial discharge, the data volume of characteristic quantity is greatly reduced, and the discharge type can be identified more quickly and better.
Description
Technical Field
The invention relates to the technical field of electric power, in particular to a partial discharge type identification method and device based on singular value decomposition.
Background
Partial discharge refers to a discharge phenomenon that occurs only in a partial region in an insulator without forming a penetrating discharge channel. The mechanism and location of the partial discharge are different, and the damage degree to the equipment insulation is different. Through the extraction and the analysis to PD signal characteristic quantity and carry out mode identification work, can in time and accurately master cable insulation fault's type and characteristic, help the maintainer rationally to formulate the maintenance plan, guarantee the reliability and the security of power supply, prevent the emergence of power consumption accident.
The characteristic extraction of the PD signal determines the method and the effect of PD signal pattern recognition, and the current PD signal characteristic extraction mainly comprises a statistical characteristic method and a time domain analysis method. The statistical feature method is a method for extracting features of various statistical distribution graphs of PD signals, and the main method comprises the following steps: fractal dimension method, gray scale image method, Weibull distribution method, PRPD image spectrum analysis method, PRPS image spectrum analysis method,the statistical characteristic parameters include: skewness SkAbruptness KuDischarge quantity factor Q, cross correlation coefficient CC, scale parameter alpha and shape parameter beta of Weibull distribution and the like. However, the parameters applied to feature extraction are numerous, and no effective criterion exists for effectively selecting representative feature quantities. The time domain analysis method is to extract features according to waveform features of time domain pulses of PD signals or corresponding transforms (Gabor transform, Wigner distribution, FFT transform, wavelet transform), but how to find representative feature quantities by this method has not been solved yet.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a partial discharge type identification method and device based on singular value decomposition, which can represent the waveform information of partial discharge by using a small number of singular values, greatly reduce the data volume of characteristic quantity and can identify the type of the partial discharge more quickly and better.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a partial discharge type identification method based on singular value decomposition comprises the following steps:
s1, acquiring known partial discharge types, and establishing a partial discharge type library;
s2, carrying out denoising pretreatment on the partial discharge signal, intercepting a signal in a denoised partial discharge signal generation region to construct a Hankel matrix, and then carrying out singular value decomposition;
s3, selecting energy percentages of proper number of singular values as characteristic quantities T, and dividing each type of partial discharge characteristic quantities into training samples TtrainAnd test sample Ttest；
S4, mixing TtrainTraining to obtain an optimal BP neural network as the input of the BP neural network optimized by the genetic algorithm;
s5, mixing TtestAnd obtaining the type of the partial discharge signal as the input of the trained BP neural network.
The invention also provides a partial discharge type identification device based on singular value decomposition, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the processor realizes the steps of the identification method when executing the program.
The invention also provides a computer-readable storage medium, in which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the above-mentioned identification method.
Compared with the prior art, the invention has the beneficial effects that:
1. by intercepting the partial discharge occurrence area to carry out singular value decomposition, a small amount of singular values can represent waveform information of partial discharge, the data volume of characteristic quantity is greatly reduced, and the discharge type can be identified more quickly and better;
2. the invention utilizes the comparison of similarity NCC to iterate and find out the number p of singular values which effectively represent partial discharge signals;
3. the method utilizes the proportional invariance of the singular value and adopts the singular value percentage as the characteristic quantity, thereby well avoiding the prediction failure caused by the unpredictability of the singular value.
Drawings
Fig. 1 is a flowchart of a partial discharge type identification method based on singular value decomposition according to the present invention.
Fig. 2 is a schematic diagram of four partial discharge time-domain waveforms.
FIG. 3 is a diagram of the cable termination discharge waveform before and after de-noising.
FIG. 4 shows reconstructions Y (a) and
and (4) waveform.
Fig. 5 is a typical three-layer BP network topology.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
The embodiment discloses a partial discharge type identification method based on singular value decomposition, which is used for identifying the partial discharge type of a cable, and as shown in fig. 1, the method comprises the following steps:
s1, acquiring known partial discharge types, and establishing a partial discharge type library;
known partial discharge types are four partial discharge signals, namely a cable body PD, a cable termination PD, a corona discharge in a switchgear cabinet and a surface discharge, which can be seen as a small amount of noise interference, as shown in fig. 2. According to the types of the partial discharge data, 1000,0100,0010,0001 is used for marking, and the partial discharge data are classified into the same database, so that a partial discharge type library is established.
S2, carrying out denoising pretreatment on the partial discharge signal, intercepting the signal in the denoised partial discharge signal generation region to construct a Hankel matrix, and then carrying out singular value decomposition:
s3, selecting energy percentages of proper number of singular values as characteristic quantities T, and dividing each type of partial discharge characteristic quantities into training samples TtrainAnd test sample Ttest：
S4, mixing TtrainTraining to obtain an optimal BP neural network as the input of the BP neural network optimized by the genetic algorithm;
s5, mixing TtestAnd obtaining the type of the partial discharge signal as the input of the trained BP neural network.
Further, the step S2 includes the following steps:
s21, denoising a partial discharge signal y (i) (1, 2., N) by adopting a wavelet packet method, dividing the partial discharge signal into a series of detail components and approximate components, obtaining a y (i) partial discharge threshold THR by using a matlab packaged threshold function ddencmp, substituting the THR into a packaged wavelet packet decomposition denoising function wpdencmp function to decompose and denoise the signal, and obtaining the denoised partial discharge signal
As can be seen from fig. 3, after the cable termination sample is decomposed by the wavelet packet, the noise is obviously removed.
The calling format of the function is as follows:
THR＝ddencmp(’den’,’wp’,’y(i)’)；
wherein y (i) represents a partial discharge signal, den represents denoising, wp represents selection of a wavelet packet, THR represents a returned threshold value,'s' represents denoising with a soft threshold value, 8 represents the number of decomposition layers, 'db8' represents a wavelet name, 'sure' represents an entropy standard, 1 represents that threshold quantization processing is not performed,
representing the denoised partial discharge signal.
S22, intercepting
Region of occurrence [ l1,l2]Of (2) a signal
And establishing a Hankel matrix H. The Hankel matrix is constructed as follows:
wherein: l1<n'<l2，m'＝l2+l1N', if l2+l1Is an even number and is provided with a plurality of groups,
if l is2+l1Is an odd number of the components,
s23, performing singular value decomposition on H, wherein the singular value decomposition is as follows:
wherein: u, V are expressed as m × m and n × n orthogonal matrices, respectively, and D ═ diag (λ1,λ2,...,λr)(r＝min(m,n))，uiAnd viRepresenting m, n dimensional column vectors.
Further, the step S3 includes the following steps:
s31, when r is 1, reconstructing singular values using the formula of S23 to obtain a matrix H*Taking the first row and the n' th column in H ·, a partial discharge signal y (a) (a ═ l) is formed1,l1+1,...,l2) Then, similarity NCC evaluation was performed, and z is made NCC. The formula NCC is:
s32, where r is r +1, the process continues to S31 to determine the similarity NCCr。
S33, if z<NCCrIf z is NCCrReturning to S32 for execution, if z>NCCrAnd recording the current r, and enabling p to be r and quitting the operation.
Through operation, when p is 8, the maximum value is obtained by exiting the operation z, as shown in fig. 4, the first 8 singular values reconstruct y (a) and
the waveforms are highly coincident, which shows that the first 8 singular values can well describe the waveform characteristics of the original PD signal.
And S34, calculating the energy percentage of the previous p singular values as the characteristic quantity T.
Order to
λ of t-th singular valuetThe singular value energy percentage of is λt/λ∑Therefore, the feature vector T is:
T＝[λ1/λ∑,λ2/λ∑,...,λt/λ∑]
s35, dividing each type of partial discharge characteristic quantity into training samples TtrainAnd test sample Ttest。
Further, the step S4 includes the following steps:
s41, determining a BP neural network topology, as shown in fig. 5, where there are 4 sample outputs and 16 input parameters, and the BP neural network topology may be determined as follows: 16 × 23 × 4, initializing the weight and threshold of the BP neural network.
And S42, binary coding the weight and the threshold of the initialized BP neural network by adopting a genetic algorithm, initializing a population, wherein the population individual number NIND is 40, the channel GGAP is 0.95, the cross probability px is 0.7, the variation probability pm is 0.01, and the maximum genetic algebra MAXGEN is 20.
S43, selecting, crossing and mutating population individuals, taking error rate err between the prediction classification and the actual classification as a fitness function, and taking T as a fitness functiontrainAnd performing fitness evaluation as the input of the BP neural network, wherein the smaller the err is, the higher the possibility of the ERr as a genetic operator of the next generation is, and further, the weight and the threshold of the BP neural network are continuously optimized.
And S44, judging whether the maximum genetic algebra MAXGEN is reached, if so, quitting the operation, obtaining the optimal weight and threshold value, obtaining the optimal BP neural network, and if not, continuing optimizing the optimal weight and threshold value of the BP neural network.
In this embodiment, the obtained 4 kinds of partial discharge signals of 50 groups are used as samples, and the total number of the partial discharge signals is 200 groups, and 30 groups of data of each type are used as training samples, and the total number of the partial discharge signals is 120 groups, and 20 groups of the partial discharge signals are test samples, and the total number of the partial discharge signals is 80 groups. The average recognition rate of the four partial discharge signals obtained by the method reaches 97.5%, and as shown in table 1, the method proves that a small number of singular values can represent waveform information of partial discharge by intercepting a partial discharge occurrence region to carry out singular value decomposition, so that the data volume of characteristic quantity is greatly reduced, and the discharge type can be recognized more quickly and better.
TABLE 1 GA-BP neural network recognition Effect
In addition, all or part of the flow in the method according to the embodiments of the present invention may be implemented by a computer program, which may be stored in a computer readable storage medium and used by a processor to implement the steps of the embodiments of the method. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), etc.
The above embodiments are only for illustrating the technical concept and features of the present invention, and the purpose thereof is to enable those skilled in the art to understand the contents of the present invention and implement the present invention accordingly, and not to limit the protection scope of the present invention accordingly. All equivalent changes or modifications made in accordance with the spirit of the present disclosure are intended to be covered by the scope of the present disclosure.
Claims (7)
1. A partial discharge type identification method based on singular value decomposition is characterized in that: the identification method comprises the following steps:
s1, acquiring known partial discharge types, and establishing a partial discharge type library;
s2, carrying out denoising pretreatment on the partial discharge signal, intercepting a signal in a denoised partial discharge signal generation region to construct a Hankel matrix, and then carrying out singular value decomposition;
s3, selecting energy percentages of proper number of singular values as characteristic quantities T, and dividing each type of partial discharge characteristic quantities into training samples TtrainAnd test sample Ttest；
S4, mixing TtrainTraining to obtain an optimal BP neural network as the input of the BP neural network optimized by the genetic algorithm;
s5, mixing TtestAnd obtaining the type of the partial discharge signal as the input of the trained BP neural network.
2. The partial discharge type identification method based on singular value decomposition according to claim 1, wherein: in step S1, g types of known partial discharge types are obtained, binary numbers are used to mark the types of the known partial discharge types, and the partial discharge data are classified and placed in the same database to establish a partial discharge type library.
3. The partial discharge type identification method based on singular value decomposition according to claim 1, wherein: the step S2 includes the steps of:
s21, denoising a partial discharge signal y (i) (1, 2., N) by adopting a wavelet packet method, dividing the partial discharge signal into a series of detail components and approximate components, obtaining a y (i) partial discharge threshold THR by using a matlab packaged threshold function ddencmp, substituting the THR into a packaged wavelet packet decomposition denoising function wpdencmp function to decompose and denoise the signal, and obtaining the denoised partial discharge signal
S22, intercepting
Region of occurrence [ l1,l2]Of (2) a signal
Establishing a Hankel matrix H, wherein the Hankel matrix is constructed as follows:
wherein: l1<n'<l2，m'＝l2+l1N', if l2+l1Is an even number and is provided with a plurality of groups,
if l is2+l1Is an odd number of the components,
s23, performing singular value decomposition on H, wherein the singular value decomposition is as follows:
wherein: u, V are expressed as m × m and n × n orthogonal matrices, respectively, and D ═ diag (λ1,λ2,...,λr)(r＝min(m,n))，uiAnd viRepresenting m, n dimensional column vectors.
4. The partial discharge type identification method based on singular value decomposition according to claim 1, wherein: the step S3 includes the steps of:
s31, when r is 1, reconstructing singular values using the formula of S23 to obtain a matrix H*Taking the first row and the n' th column in H ·, a partial discharge signal y (a) (a ═ l) is formed1,l1+1,...,l2) Then, similarity NCC evaluation is performed, let z ═ NCC, the formula of NCC:
s32, where r is r +1, the process continues to S31 to determine the similarity NCCr；
S33, if z<NCCrIf z is NCCrReturning to S32 for execution, e.gFruit Z>NCCrRecording current r, and enabling p to be r, and quitting the operation;
s34, calculating the energy percentage of the previous p singular values as a characteristic quantity T:
order to
λ of t-th singular valuetThe singular value energy percentage of is λt/λ∑Therefore, the feature vector T is:
T＝[λ1/λ∑,λ2/λ∑,...,λt/λ∑]
s35, dividing each type of partial discharge characteristic quantity into training samples TtrainAnd test sample Ttest。
5. The partial discharge type identification method based on singular value decomposition according to claim 1, wherein: the step S4 includes the steps of:
s41, determining a topological structure of the BP neural network, and initializing a weight and a threshold of the BP neural network;
s42, binary coding is carried out on the weight and the threshold of the initialized BP neural network by a genetic algorithm, the population is initialized, the population individual number NIND, the generation channel GGAP, the cross probability px, the variation probability pm and the maximum genetic algebra MAXGEN are obtained;
s43, selecting, crossing and mutating population individuals, taking error rate err between the prediction classification and the actual classification as a fitness function, and taking T as a fitness functiontrainAs the input of the BP neural network, fitness evaluation is carried out, the smaller err is, the higher the possibility that the ERR is used as a genetic operator of the next generation is, and then the weight and the threshold of the BP neural network are continuously optimized;
and S44, judging whether the maximum genetic algebra MAXGEN is reached, if so, quitting the operation, obtaining the optimal weight and threshold value, obtaining the optimal BP neural network, and if not, continuing optimizing the optimal weight and threshold value of the BP neural network.
6. A partial discharge type identification device based on singular value decomposition, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the identification method according to any one of claims 1 to 5 when executing the program.
7. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the identification method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010698430.0A CN112001246A (en) | 2020-07-20 | 2020-07-20 | Partial discharge type identification method and device based on singular value decomposition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010698430.0A CN112001246A (en) | 2020-07-20 | 2020-07-20 | Partial discharge type identification method and device based on singular value decomposition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112001246A true CN112001246A (en) | 2020-11-27 |
Family
ID=73468230
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010698430.0A Pending CN112001246A (en) | 2020-07-20 | 2020-07-20 | Partial discharge type identification method and device based on singular value decomposition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112001246A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112578240A (en) * | 2020-12-04 | 2021-03-30 | 国网北京市电力公司 | Method and device for confirming discharge type, storage medium and electronic device |
| CN112686182A (en) * | 2021-01-04 | 2021-04-20 | 华北电力大学（保定） | Partial discharge mode identification method and terminal equipment |
| CN114065814A (en) * | 2021-11-16 | 2022-02-18 | 中国南方电网有限责任公司超高压输电公司广州局 | Method and device for identifying defect types of GIL partial discharge |
| CN117092206A (en) * | 2023-08-09 | 2023-11-21 | 国网四川省电力公司电力科学研究院 | Defect detection method for cable lead sealing area, computer equipment and storage medium |
| CN117630679A (en) * | 2023-11-30 | 2024-03-01 | 湖北工业大学 | Battery fault diagnosis method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101984463A (en) * | 2010-11-02 | 2011-03-09 | 中兴通讯股份有限公司 | Method and device for synthesizing panoramic image |
| CN106154344A (en) * | 2016-08-01 | 2016-11-23 | 湖南文理学院 | A kind of Magnetotelluric signal denoising method based on combined filter |
| CN106599777A (en) * | 2016-11-02 | 2017-04-26 | 华南理工大学 | Cable partial discharge signal identification method based on energy percentage |
| CN107037327A (en) * | 2016-10-09 | 2017-08-11 | 中国电力科学研究院 | Partial discharges fault judges feature extracting method and decision method |
| CN109829412A (en) * | 2019-01-24 | 2019-05-31 | 三峡大学 | The Partial Discharge Pattern Recognition Method of fractal characteristic is decomposed based on dynamic mode |
| CN109975665A (en) * | 2019-03-22 | 2019-07-05 | 华南理工大学 | A kind of local discharge signal in electric power equipment electric discharge type recognition methods |
-
2020
- 2020-07-20 CN CN202010698430.0A patent/CN112001246A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101984463A (en) * | 2010-11-02 | 2011-03-09 | 中兴通讯股份有限公司 | Method and device for synthesizing panoramic image |
| CN106154344A (en) * | 2016-08-01 | 2016-11-23 | 湖南文理学院 | A kind of Magnetotelluric signal denoising method based on combined filter |
| CN107037327A (en) * | 2016-10-09 | 2017-08-11 | 中国电力科学研究院 | Partial discharges fault judges feature extracting method and decision method |
| CN106599777A (en) * | 2016-11-02 | 2017-04-26 | 华南理工大学 | Cable partial discharge signal identification method based on energy percentage |
| CN109829412A (en) * | 2019-01-24 | 2019-05-31 | 三峡大学 | The Partial Discharge Pattern Recognition Method of fractal characteristic is decomposed based on dynamic mode |
| CN109975665A (en) * | 2019-03-22 | 2019-07-05 | 华南理工大学 | A kind of local discharge signal in electric power equipment electric discharge type recognition methods |
Non-Patent Citations (2)
| Title |
|---|
| 唐炬 等: "基于Hankel矩阵的复小波-奇异值分解法提取局部放电特征信息", 《中国电机工程学报》 * |
| 谢敏 等: "一种基于短时奇异值分解的局部放电白噪声抑制方法"", 《中国电机工程学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112578240A (en) * | 2020-12-04 | 2021-03-30 | 国网北京市电力公司 | Method and device for confirming discharge type, storage medium and electronic device |
| CN112686182A (en) * | 2021-01-04 | 2021-04-20 | 华北电力大学（保定） | Partial discharge mode identification method and terminal equipment |
| CN112686182B (en) * | 2021-01-04 | 2023-12-26 | 华北电力大学（保定） | Partial discharge mode identification method and terminal equipment |
| CN114065814A (en) * | 2021-11-16 | 2022-02-18 | 中国南方电网有限责任公司超高压输电公司广州局 | Method and device for identifying defect types of GIL partial discharge |
| CN117092206A (en) * | 2023-08-09 | 2023-11-21 | 国网四川省电力公司电力科学研究院 | Defect detection method for cable lead sealing area, computer equipment and storage medium |
| CN117630679A (en) * | 2023-11-30 | 2024-03-01 | 湖北工业大学 | Battery fault diagnosis method and system |
| CN117630679B (en) * | 2023-11-30 | 2024-06-07 | 湖北工业大学 | Battery fault diagnosis method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112001246A (en) | Partial discharge type identification method and device based on singular value decomposition | |
| Wang et al. | Optimized feature extraction for learning-based image steganalysis | |
| CN109975665B (en) | Power equipment partial discharge signal discharge type identification method | |
| KR101664913B1 (en) | Method and system for determining a quality measure for an image using multi-level decomposition of images | |
| CN110361778B (en) | Seismic data reconstruction method based on generation countermeasure network | |
| CN109711283B (en) | Occlusion expression recognition method combining double dictionaries and error matrix | |
| CN107194378B (en) | Face recognition method and device based on mixed dictionary learning | |
| CN109389171B (en) | Medical image classification method based on multi-granularity convolution noise reduction automatic encoder technology | |
| CN106845516B (en) | Footprint image recognition method based on multi-sample joint representation | |
| Ringrose | Bootstrapping and correspondence analysis in archaeology | |
| Bhateja et al. | Fast SSIM index for color images employing reduced-reference evaluation | |
| Yang et al. | Wavelet-based EEG preprocessing for biometric applications | |
| CN115546558A (en) | Electrical equipment insulation fault state classification method and device and storage medium | |
| CN114067915A (en) | scRNA-seq data dimension reduction method based on deep antithetical variational self-encoder | |
| CN115792729A (en) | Transformer composite fault diagnosis method, device, equipment and storage medium | |
| CN114067286A (en) | High-order camera vehicle weight recognition method based on serialized deformable attention mechanism | |
| CN106803105B (en) | Image classification method based on sparse representation dictionary learning | |
| CN115839847A (en) | Bearing fault diagnosis method and system based on WDCNN-LSTM | |
| Wu et al. | Reduced-reference image quality assessment with local binary structural pattern | |
| CN114118177B (en) | Bit plane noise reduction method and system based on singular spectrum analysis and storage medium | |
| CN111275109A (en) | Power equipment state data characteristic optimization method and system based on self-encoder | |
| CN116028858A (en) | Pre-labeled self-supervision neural network learning heart beat classification method and system | |
| CN115496993A (en) | Target detection method, device and equipment based on frequency domain fusion and storage medium | |
| CN114862803A (en) | Industrial image anomaly detection method based on fast Fourier convolution | |
| Zhang et al. | Nature scene statistics approach based on ICA for no-reference image quality assessment |
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 |
Application publication date: 20201127 |
|
| RJ01 | Rejection of invention patent application after publication |