Application of Data Mining Technology Based on RVM for Power Transformer Fault Diagnosis

Abstract

One of the most challenging problems in real-time operation of power system is the state monitoring and fault diagnosis for large-scale transformer. Fast and accurate techniques are imperative to achieve on-line state assessment. To counter the shortcoming of common machine learning methods, a novel machine learning technique, i.e. ‘relevance vector machine’ (RVM), for on-line state assessment is presented in this paper. The proposed method is tested and compared with ‘support vector machine’ (SVM) classifier. It is demonstrated that the RVM classifier can yield a decision function that is much sparser than the SVM classifier while providing higher classification accuracy. Consequently, the RVM classifier greatly reduces the computational complexity, making it more suitable for real-time implementation.