The ICA-SVM based operation state identification for oil immersed distribution transformers

Distribution transformers are one of the most important equipment in the modern distribution systems. They directly affect the stability and security of distribution power grid. Dissolved gas analysis (DGA) is a common approach used for operation state identification of distribution transformers. This paper proposes an operation state identification method by using the DGA data measured from the distribution transformers. The proposed method is based on the integration of independent component analysis and support vector machine (ICA-SVM). Firstly, the ICA is performed on the DGA data to extract feature vectors; Then, the feature vectors are served as input of SVM to identify the transformer operation states; Finally, the on-site monitoring DGA data from 110 kV distribution transformers are applied to verify the effectiveness of the proposed method. The experimental results show that the proposed ICA-SVM method can recognize the operation states of distribution transformers effectively.