Research on electrical life Trace prediction of contact based on SG-BP algorithm

Abstract In order to prevent electrical fire in low-voltage distribution system, it is a very effective measure predict the life of various appliances running on line. At present, difficult residual AC contactors. How establish reasonable and reliable prediction model for contactors key test characteristic research this paper, Savitzky-Golay convolution smoothing algorithm combined with BP neural network algorithm, remaining contact established by taking arcing time, energy opening phase angle as input parameters model. The results show that average relative error less than 4%, maximum 10%. Combined SG-BP has 3% 8%, which improves accuracy life, lays good foundation making safe judgment better use trace data..