Monitoring and fault diagnosing system design for power transformer based on temperature field model and DGA feature extraction

Based on support vector machine, a new fault diagnosis model for power transformer combining on-line extracting dissolved gases analysis (DGA) data with its three-dimensional temperature field information is proposed, which can realize fusion of the multivariate fault characteristic information. The finite element method is applied into establishing a three-dimensional temperature field for power transformer. Some issues about applying support vector machine (SVM) into power transformer fault diagnosis are further analyzed. In order to realize its state on-line monitoring and the measuring data remote communication, an embedded on-line monitoring system composed of ARM and general packet radio service is designed for the power transformer in this paper. Simulation experimental results show that the proposed fault diagnosis model has an excellent performance on training speed and correct ratio, and the developed embedded system can effectively monitor power transformerpsilas state during its running. Therefore, it will change the existing maintenance and repair pattern for power transformer and realizes accurate fault diagnosis or forecasts.