摘要
小波神经网络分为松散型和紧致型。前者以小波分析作为神经网络的前置处理,为其提供输入特征向量。后者采用小波函数和尺度函数形成神经元,达到小波和神经网络的直接融合。刀具故障在线监测松散型小波神经网络,用声传感器在线采集刀具的状态数据,用小波分析处理数据并提取特征量以确定网络的输入向量;用BP神经网络训练数据,调整权值和故障输出,建立刀具故障在线监测小波神经网络模型。
Wavelet Neural-Networks is fallen into relax-type and close-type. In the former, wavelet analysis is used as prepositive control and supply eigenvector for NN. Nerve cell is formed with wavelet function and scale function to integrate wavelet and NN in the latter. In relax-type wavelet NN to inspect on-line tools faults, state data of tool is collected by sound-sensor, the data is processed with wavelet analysis, and picks up eigenvalues to confirm input-vectors of NN. Those data is exercised BP NN, and adjust the value and output of faults. Finally a model of wavelet NN to inspect on-line tools faults is established.
出处
《兵工自动化》
2003年第5期45-46,共2页
Ordnance Industry Automation
