论文部分内容阅读
以振动筛下横梁作为损伤结构识别的研究对象,对其进行实验室加载试验。利用声发射检测技术采集与部件疲劳损伤相关的声发射波形信号,并应用“小波包-能量”法提取其信号特征量,将其作为神经网络的输入向量。在MATLAB 6.5的环境下进行了神经网络识别计算,结果表明,将小波包-神经网络用于声发射信号处理及零件早期疲劳损伤的诊断是可行的。
The vibrating screen under the beam as a damage structure to identify the object of study, the laboratory loading test. The acoustic emission waveform signals related to the fatigue damage of components were collected by using acoustic emission detection technique. The signal features were extracted by using “wavelet packet-energy” method, which was regarded as the input vector of neural network. The neural network identification and calculation in the MATLAB 6.5 environment shows that it is feasible to use the wavelet packet-neural network in acoustic emission signal processing and early fatigue damage diagnosis.