基于VMD瞬时能量与GA- RBF的滚动轴承故障诊断

针对滚动轴承早期故障的有效识别,提出了一种基于VMD瞬时能量与GA优化的RBF神经网络的滚动轴承故障诊断方法,可以有效对滚动故障做出诊断。首先,VMD将滚动轴承振动信号进行分解成合适数目的本证模态函数;其次,计算本证模态函数分量的瞬时能量并组成特征向量;最后,将特征向量输入到GA优化的RBF神经网络实现轴承故障识别。通过滚动轴承故障诊断实验对该方法进行验证。结果表明,该方法识别滚动轴承故障的准确率为96.43%,较默认参数的RBF神经网络和EEMD瞬时能量与GA-RBF神经网络有明显的提高,证明了所提方法的可行性。

Rolling Bearing Fault Diagnosis Based on VMD Instantaneous Energy and GA Optimized RBF Neural Network

The rolling bearing is an important supporting component of the CNC machine feed axis.The working condition of the rolling bearing is related to the accuracy of the feed axis and the machine tool.For the effective identification of the early failure of the rolling bearing,a RBF neural network based on VMD instantaneous energy and GA optimization is proposed.The rolling bearing fault diagnosis method can effectively diagnose the rolling fault.First,the VMD decomposes the rolling bearing vibration signal into a suitable number of the modal function;then,calculates the instantaneous energy of the modal function component and forms the eigenvector;finally,inputs the eigenvector into the GA-optimized RBF neural network.Bearing fault identification.The method is validated by rolling bearing fault diagnosis experiment.The results show that the accuracy of the method for identifying rolling bearing faults is 96.43%,which is significantly higher than the default parameters of RBF neural network and EEMD instantaneous energy and GA-RBF neural network.The result verified the feasibility of the proposed method.