In order to improve the accuracy of transformer fault diagnosis results

a transformer fault diagnosis method based on PCA-ISO-SVM is proposed. The principal component analysis(PCA) method is used to reduce the dimensionality of transformer fault diagnosis feature quantities

and the original 20 dimensional feature quantities are reduced to 7 based on the cumulative contribution rate. By utilizing Tent mapping

dynamic adjustment

and Cauchy mutation strategy

the basic snake optimizer(SO) algorithm is improved to obtain an ISO algorithm with better optimization performance. The ISO algorithm is used to optimize the parameters of SVM

and a transformer fault diagnosis model based on PCA-ISO-SVM is established. Using actual transformer fault data for case analysis and comparing with other models

the results show that the PCA-ISO-SVM model only experienced one misdiagnosis

with a diagnostic accuracy of 98.33%

which is significantly higher than other models. The method is applied to fault diagnosis in a power company

the diagnostic results of the PCA-ISO- SVM model are consistent with the results of the transformer hanging core inspection

verifying the effectiveness and practicality of the proposed transformer fault diagnosis method.