项目名称
动手教你学故障诊断:Python实现基于Tensorflow+CNN深度学习的轴承故障诊断（西储大学数据集）（含完整代码）
项目介绍
该项目使用tensorflow和keras搭建深度学习CNN网络，并使用西储大学数据集作为训练集和测试集，对西储大学mat格式数据进行处理，将数据放入搭建好的网络中进行训练，最终得到相关故障诊断模型。
背景
最近，在故障诊断的过程中，老师给我们发来了西储大学的轴承故障数据集，让我们自己来做。巧合的是，我前段时间学习了深度学习的课程，所以我想构建一个深度学习网络来诊断相关故障。参考相关文献，使用深度学习的故障诊断方法主要有两种形式，一种是直接将相关的加速度一维数据放入深度学习网络学习，另一种是利用相关变化将加速度数据转换成二维图像，再将二维图像放入深度学习网络进行学习。本文采用第一种方法，然后介绍代码的相关部分，那些想学习和实践的人也可以跳到最后才有完整的代码。
[En]
Recently, in the course of fault diagnosis, the teacher sent us the bearing fault data set of Western Reserve University, and let us do it by ourselves. Coincidentally, I learned the course of deep learning some time ago, so I want to build a deep learning network to diagnose the related faults. Referring to the relevant literature, there are mainly two forms of fault diagnosis methods using deep learning, one is to directly put the relevant acceleration one-dimensional data into the deep learning network to learn, the other is to use the relevant changes to convert the acceleration data into two-dimensional images, and put the two-dimensional images into the deep learning network for learning. This paper uses the first method, and then introduces the relevant parts of the code, and those who want to learn and practice can also skip to the end to have the complete code.