摘要
通过对轴承的振动特征的分析 ,确定了基于故障的征兆频率 ,进而又构造出了轴承的征兆空间和故障空间的模式 ;采用多层前馈型神经网络 ,通过网络的自学习和训练 ,实现了两个空间之间的非线性映射 ;最后 。
By the analysis of the vibration of bearing,the relationship between the bearing fault and the vibration frequency was foumd out,and the symptom space and fault space mode was built up.By using the ability of self-learning and training of Artificial Neural Network,the non-linear transformation from symptom space to fault space was implemented,and the whole bearing fault diagnosis system was designed.
出处
《机床与液压》
北大核心
2004年第11期208-210,176,共4页
Machine Tool & Hydraulics
关键词
轴承
神经元网络
特征频率
故障诊断
Bearing
Artificial neural network
Specified frequency
Fault diagnosis
