Crack Identification of Drawing Parts Based on Loccal Wave Demomposition and Neural Network

This paper relates to local wave decomposition and back-propagation (BP) neural network.With local wave method, an arbitrary acoustic emission signal can be decomposed efficiently and accurately into a set of intrinsic mode functions (IMFs) and a residual trend. The energy feature parameters extracted from IMFs were employed as the input parameters of the neural network to identify the acoustic emission signals of drawing parts.The experimental results showed this method was effective for crack identification of drawing parts.