Steel Flatness Pattern Recognition Using RBF and Wavelet Packet Analysis
Pattern recognition for flatness is one of the difficult techniques in °atness control system of high precision. A novel pattern recognition method based on wavelet packet analysis (WPA) and radial basis function (RBF) network is presented in this paper. The wavelet packet de-noising method is used to filter the noise of the measure data availably. The pattern recognition model for °atness based on radial basis function network can avoid considering the complicated function relationship between measure data and °atness. This method can classify the complex °atness correctly.
Min Huang Li Peng
College of Communication and Control Engineering Southern Yangtze University Wuxi, China 214122
国际会议
南宁
英文
2007-07-20(万方平台首次上网日期,不代表论文的发表时间)