A novel quantitative spectral analysis method based on parallel BP neural network for dissolved gas in transformer oil

The dissolved gas in transformer oil, which could represents the transformer faults, can be analyzed by spectroscopy. Since the BP neural network method involves a large amount of data matrices operation which leads to much computation and that single computer can not meet the requirements of real-time analysis. In order to improve the situation, this paper proposes a quantitative spectral analysis of dissolved gas in transformer oil based on parallel BP neural network. This paper designs parallel BP neural network model and builds independently the Hadoop clustering computing platform to implement the parallel model. The cluster computing system is a parallel and distributed computer system with high performance which can easily handle large data sets and improve the computing speed. We use a dissolved gas spectral data-set of a real transformer oil in the experiments to evaluate this approach. The parallel BP neural network model is performed on the Hadoop clustering computing platform for component prediction. The experimental results verify that the proposed model can predict the component concentrations of the dissolved gas in transformer oil correctly and has high effectiveness.