The best way to conference proceedings by Francis Academic Press

Web of Proceedings - Francis Academic Press
Web of Proceedings - Francis Academic Press

Diagnosis of DC Bias in Power Transformers with Vibration test Method

Download as PDF

DOI: 10.25236/iccem.2021.034

Author(s)

Hao Cao, Sheng Hu, Xiaozhong Wu, Zhiqiang Xu

Corresponding Author

Hao Cao

Abstract

This paper deals with a vibration test method to diagnose DC bias in power transformers. Vibration variation process of a 500kV autotransformer is tested under the influence of DC bias. Transformer vibrations in normal and DC-biased conditions are compared. Three features are proposed and is validated by sensitivity analysis. The Principal Component Analysis method is employed for feature de-correlation and dimensionality reduction. The least square support vector machine algorithm is used and verified successful in DC bias recognition. The suggested diagnostic method could be useful in targeted DC bias control and improving the safe operation level of power transformers.

Keywords

Power transformer, DC bias, feature extraction, pattern recognition