摘要
提出小波神经网络对微孔钻削进行实时监测的方法,利用扭矩信号的小波包分解,以分解后的各能量向量作为神经网络的输入,对系统进行训练,利用Matlab和LabView软件建立微孔钻削在线监测软件系统。试验结果表明小波神经网络精度高、收敛速度快,采用小波神经网络对提高微孔钻削在线监测的准确性是有效的。
A micro-role drilling monitoring method was brought forward based on BP neural network, whose input signals were the energy eigenvectors of torque signal by using wavelet packet transform. Furthermore, a kind of micro-role drilling on-line monitoring software system has been constructed by using Matlab software and LabView software. Experiments validated that the rate of checking out micro-drills breakage was very high by using wavelet neural networks, with the characters of high-precision, quick-convergence compared with BP neural network, and the monitoring system has good practical value.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2007年第2期176-178,182,共4页
Transactions of the Chinese Society for Agricultural Machinery
基金
吉林省科技发展计划项目(项目编号:20010574)
关键词
微孔钻削
微钻头
小波神经网络
在线监测
Micro-role drilling, Micro-drill, Wavelet neural networks, On-line monitoring
