摘要
针对电机滚动轴承故障检测的复杂性,采用了理论成熟且应用较多的BP神经网络和RBF神经网络两种故障诊断方法。首先通过经验模态分解的方法对滚动轴承的振动信号进行故障特征提取,并将故障特征向量输入到BP神经网络和RBF神经网络进行达标训练,最后对两种神经网络在滚动轴承故障诊断方面进行了比较分析,结果表明,两种神经网络的故障诊断效果均理想,但是RBF神经网络故障诊断结果较准且训练速度快,具有一定的优越性。
Aiming at the complexity of motor rolling bearing fault detection,two fault diagnosis methods,BP neural network and RBF neural network,which are mature in theory and widely used,are adopted.First,this paper extracts the fault features from the vibration signals of rolling bearings by the method of empirical mode decomposition,then inputs the fault feature vector to the BP neural network and the RBF neural network to achieve the standard training.Finally,it compares the two neural networks in the fault diagnosis of the rolling bearing.The analysis shows that the fault diagnosis results of the two neural networks are ideal,but the RBF neural network fault diagnosis results are accurate and the training speed is fast,which has certain superiority.
作者
凌标灿
杨佳滨
LING Biaocan;YANG Jiabin(School of Safety Engineering,North China Institute of Science and Technology,Yanjiao,065201,China)
出处
《华北科技学院学报》
2018年第6期53-57,共5页
Journal of North China Institute of Science and Technology
基金
中央高校基本科研业务费资助(3142018040)
