Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Unbalanced SVM Transformer Fault Detection Method Based on Simplification and Reduction

A transformer fault and detection method technology, applied in the direction of instruments, measuring electronics, measuring devices, etc., can solve the problems that the improvement of the classification effect is not obvious, and the overall spatial structure information is not taken into account.

Active Publication Date: 2017-10-03
STATE GRID CORP OF CHINA +2
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, since these upsampling algorithms only consider selecting a partial subset of the minority class to generate artificial samples to achieve balance, and do not take into account the overall spatial structure information, sometimes the improvement of the classification effect is not obvious

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Unbalanced SVM Transformer Fault Detection Method Based on Simplification and Reduction
  • Unbalanced SVM Transformer Fault Detection Method Based on Simplification and Reduction
  • Unbalanced SVM Transformer Fault Detection Method Based on Simplification and Reduction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] A method for detecting a fault in a transformer based on streamlined sets and reduced unbalanced SVMs, the method comprising the steps of:

[0035] 1) Use the fault feature extraction method based on the GARCH model to obtain the transformer feature vector set, and then determine the boundary samples for the minority class samples. The minority class samples are fault samples, and obtain the minority class boundary sample set ;

[0036] (2) Random selection , for collection base, , Set to 1, use the reduced set algorithm to get , repeat the operation times, where: is the number of samples in the majority class, is the number of minority class samples, so we get An artificial minority class sample, guaranteed to be at least once ;

[0037](3) After combining the artificial minority samples generated in step (2) with the original minority samples, they are combined with the original majority sample set as the training samples of the SVM classifier, ...

Embodiment 2

[0041] In the described unbalanced SVM transformer fault detection method based on streamlined sets, the SVM algorithm involved and its unbalanced data classification performance analysis are as follows:

[0042] The SVM algorithm is a learning method with a strict theoretical basis. Its basic idea is to realize classification by establishing an optimal linear hyperplane, and attribute the solution of the hyperplane to solving a convex programming problem. At the same time, the Mercer kernel expansion theorem is used to upgrade the sample space to the Hilbert feature space, so that the linear learning method can be used in the feature space to solve the nonlinear classification problem of the sample space.

[0043] Taking two-class classification as an example, suppose the training sample set is , Represents the training samples, the kernel function is . In order to prevent SVM from overfitting the training samples, it is necessary to increase the soft interval, constr...

Embodiment 3

[0061] According to the simplified unbalanced SVM transformer fault detection method based on embodiment 1, the fault feature extraction method based on the GARCH model described in step (1) includes the following steps: using a vibration acceleration sensor to collect Then the vibration signal is processed by a low-pass filter with a cutoff frequency of 1500 Hz to remove high-frequency noise information and obtain a noise-reduced vibration signal; the obtained noise-reduced vibration signal is processed in sections according to time series, And use the generalized autoregressive conditional heteroscedastic model GARCH to model the segmented time series, and then select the model parameters after modeling as fault characteristics to analyze the state of the transformer. The specific method is as follows:

[0062] for vibration signals Truncate processing, get truncated signal And use GARCH (1,1) to model the truncated time series, the model is as follows:

[0063] (1), ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

Disclosed is a transformer fault detecting method based on a simplified set unbalanced SVM. The method comprises (1) obtaining a characteristic vector set through a fault characteristic extracting method based on GARCH models; (2) performing determination of boundary samples on minority-class samples to obtain a minority-class boundary sample set S, wherein the minority-class samples are fault samples; randomly selecting N[x]={2, , ISI}, wherein ISI is the cardinal number of S, a[i]=1, and i=1, N[x], and setting N[z] to be 1, utilizing a simplified set solution algorithm to obtain Z[1] and repeating the operation for N[L]-N[M] times, wherein N[L] is the number of majority samples, N[M] is the number of minority samples, and accordingly, N[L]-N[M] is the number of artificial minority samples, and guaranteeing N[z]=ISI for at least one once; (3) combining the artificial minority samples obtained in the step (2) with original minority samples to serve as the training samples of an SVM classifier and lastly to obtain an SVM decision model; (4) inputting newly-obtained transformer characteristic vectors into the decision model for judgment. The transformer fault detecting method based on the simplified set unbalanced SVM is applied to transformer fault detection.

Description

Technical field: [0001] The invention relates to a fault detection method for an unbalanced SVM transformer based on simplified set reduction. Background technique: [0002] Transformer, as an important hub equipment in the power system, is not only expensive but also very important. Ensuring its safe and reliable operation is the key to the safe operation of the entire power grid. However, the internal structure of the transformer is very complicated, and various problems may occur in the manufacturing and installation process, such as short circuit, deformation, displacement or loosening of the transformer winding, changes in the compression state of the iron core, and loosening of the foot bolts. These problems will cause the transformer to fail for a long time. run. Therefore, it becomes very necessary to detect its running state. However, since the causes of transformer failure are many and random, it is very difficult to detect them. In recent years, scholars have o...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G01R31/00
Inventor 刘福荣陶新民孙福军田伟张凯
Owner STATE GRID CORP OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com