【摘要】 作为旋转设备的关键部件之一，滚动轴承的故障诊断一直受到人们的关注，在故障分类研究中，传统的机器学习方法需要通过手动提取特征，因此特征的提取并不充分且自适应性不强。针对以上问题，本文使用一维卷积神经网络对对采集到的轴承振动信号进行数据集划分，使用训练集对1D-CNN进行训练，最后使用训练好的1D-CNN模型对滚动轴承的故障进行诊断，使用凯斯西储大学轴承数据中心提供的数据对1D-CNN模型进行实验，诊断正确率在99%以上，验证了1D-CNN模型的有效性与准确性。更多还原

【Abstract】 As one of the key components of rotating equipment, the fault diagnosis of rolling bearings has been concerned by people. In the research of fault classification, traditional machine learning methods need to extract features manually,so the feature extraction is not sufficient and self adaptability is not strong. To solve the above problems, the paper uses one-dimensional convolutional neural network to divide the data set of the collected bearing vibration signal, and then uses the training set to train 1D-CNN, and finally uses the trained 1D-CNN model to diagnose the rolling bearing fault.The 1D-CNN model is tested using the data provided by the Bearing Data Center of Case Western Reserve University, the diagnostic accuracy is above 99%, which verifies the effectiveness and accuracy of the 1D-CNN model.更多还原