基于FFT和CNN的滚动轴承故障诊断方法

A fault diagnosis method of rolling bearing based on FFT and CNN

为提高滚动轴承故障诊断的识别准确率、摆脱信号处理方法对专家经验的过度依赖,提出一种基于快速傅里叶变换(FFT)和卷积神经网络(CNN)的滚动轴承故障诊断方法。即对滚动轴承原始振动信号进行快速傅里叶变换,再将得到的一维数据转变为二维的特征图,输入到ResNet-18网络中进行学习训练,以实现滚动轴承的故障诊断。通过与其他几种输入方法进行对比分析,验证了该方法的有效性以及优越性。实验结果表明该方法诊断准确率高、收敛速度快,并且信号处理过程不需要设定相关预定义参数,摆脱了对专家经验的过度依赖。

In order to improve the recognition accuracy of rolling bearing fault diagnosis,and get rid of the excessive dependence of signal processing methods on expert experience,a rolling bearing fault diagnosis method based on fast Fourier transform(FFT)and convolutional neural network(CNN)is proposed in this paper.The fast Fourier transform is performed on the original vibration signal of the rolling bearing,and then the obtained one-dimensional data is transformed into a two-dimensional feature map,which is input into the ResNet-18 network for learning and training,so as to realize the fault diagnosis of rolling bearing.Through comparative analysis with several other input methods,the effectiveness and superiority of this proposed method are verified.This method has high diagnostic accuracy and fast convergence speed,and the signal processing process in this method does not need to set relevant predefined parameters,thus getting rid of the excessive dependence on expert experience.