摘要
介绍了基于主元分析(PCA)理论改进后的BP神经网络在拖拉机齿轮故障诊断中的应用;试验中以齿轮振动信号的频域特征为神经网络输入,齿轮的主要故障模式为神经网络的输出,发现训练过的神经网络能很好的满足齿轮故障诊断的要求。
A new method for applying to the study of tractor′s gear fault diagnosis,which based on BP neural network and improved by PCA theory,is introduced.In the experiment,it takes the real frequency characteristics of gear vibration signal as the input of neural network and the main fault patterns of gear as the output of neural network,the result proves that the trained neural network could reach the requirement of gear fault diagnosis perfectly.
出处
《拖拉机与农用运输车》
北大核心
2005年第2期18-20,共3页
Tractor & Farm Transporter
基金
江苏省教育厅自然科学研究基金资助项目(00KJB460003)
