Transformer fault diagnosis method based on improved whale optimization algorithm to optimize support vector machine

An improved whale algorithm optimized support vector machine (SVM) for fault diagnosis of oil-immersed transformers is proposed. Firstly, based on the dissolved gas analysis (DGA) in oil, with no coding ratio method is utilized to extract oil can express the state of the transformer in 9 dimensional...

Full description

Saved in:
Bibliographic Details
Published in:Energy reports Vol. 7; pp. 856 - 866
Main Authors: Fan, Qingchuan, Yu, Fei, Xuan, Min
Format: Journal Article
Language:English
Published: Elsevier Ltd 01.11.2021
Elsevier
Subjects:
Dissolved gas analysis
Electric transformer
Fault diagnosis
Support vector machine
Whale optimization algorithm
Electric transformer
Fault diagnosis
Dissolved gas analysis
Support vector machine
Whale optimization algorithm
ISSN:2352-4847, 2352-4847
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An improved whale algorithm optimized support vector machine (SVM) for fault diagnosis of oil-immersed transformers is proposed. Firstly, based on the dissolved gas analysis (DGA) in oil, with no coding ratio method is utilized to extract oil can express the state of the transformer in 9 dimensional vector, will be the vector normalization and USES kernel principal component analysis (KPCA) for feature extraction, and reduce the dimension of feature vector, to avoid the overlapping between the information at the same time to speed up the speed of 5 s; Secondly, the selection, crossover and mutation operations of genetic algorithm (GA) were introduced into whale optimization algorithm (WOA), and the penalty factor and kernel parameters of SVM were further iterated. Finally, the DGA data of 627 cases of oil-immersed transformers were collected and input into GWOA-SVM, GA-SVM, PSO-SVM and WOA-SVM models, and the diagnosis results were 94.05%, 78.56%, 82.28% and 86.37%, respectively. At the same time, the stability test experiment was carried out with 10% fault data, and the diagnostic accuracy of GWOA-SVM model only decreased by about 5%. The results show that the model established in this paper has the characteristics of high diagnostic accuracy and strong stability.
ISSN:2352-4847
2352-4847
DOI:10.1016/j.egyr.2021.09.188