This paper presents a back propagation neural network (BPNN) pattern recognition algorithm based on distributed framework. In this paper, four typical partial discharge(PD) models of artificial oil-paper insulation defects are designed. PD signal characteristic parameters are extracted to form PD spectrum statistical feature library under pulsating DC voltage. This algorithm is applied to pattern recognition of four typical PD models. This algorithm is dissimilar from the traditional BPNN algorithm. The distributed BPNN algorithm can greatly reduce the training speed of the model, so as to achieve the purpose of rapid identification of fault modes. Compared with the traditional BPNN algorithm, the calculation speed of the algorithm is faster than that of the traditional algorithm.

BPNN;partial discharge;distributed framework