基于深度学习与电子听诊器的轴承故障诊断

针对滚动轴承故障诊断，受启发于传统人工听诊的做法，以及利用传统机器学习方法提取故障特征过度依赖人工、诊断正确率低等问题，提出一种基于深度学习与电子听诊器相结合的滚动轴承故障诊断方法。该方法利用电子听诊器获取轴承不同健康状态下运行的声音信号，以轴承转动周期为数据样本长度，采用重采样数据集增强方法提高模型的泛化性。搭建基于TensorFlow的一维卷积神经网络深度学习模型进行实验验证，并利用t-SNE对分类过程进行可视化，诊断正确率达到99%。

Rolling Bearing Fault Diagnosis Based on Electronic Stethoscope and Deep Learning

For rolling bearing fault diagnosis,inspired by traditional manual auscultation practices,to solve the problem that the fault features extraction by traditional machine learning method depends too much on manual work,and the diagnosis accuracy is low,a new rolling bearing fault diagnosis method based on acoustic signal and one-dimensional convolutional neural network model was proposed.The sound signals of the bearing in different health states were obtained by using an electronic stethoscope.The rotation period of the bearing was taken as the data sample length,and the resampling data set enhancement method was used to improve the generalization of the model.A one-dimensional convolutional neural network deep learning model based on TensorFlow was built for experimental verification,and t-SNE was used to visualize the classification process.The diagnostic accuracy reaches 99%.