基于机器学习算法的滚动轴承故障诊断研究

为了解决滚动轴承故障诊断过程中特征提取困难以及数据处理缓慢等主要问题,提出了基于5种机器学习算法且仅需提取4种简单特征的滚动轴承故障诊断方法。首先,对不同故障类型的滚动轴承振动信号的时域信号进行了分析,并提取时域信号的4种简单特征输入到分类模型,然后,采用机器学习算法对滚动轴承进行故障分类与诊断。实验结果表明,与传统的轴承故障诊断方法相比,用机器学习方法对轴承进行故障诊断更简单且具有更好的诊断效果。研究内容为以后用机器学习分类算法来研究轴承的故障诊断问题提供了参考。

Research on Fault Diagnosis of Rolling Bearing Based on Machine Learning Method

In order to solve the difficulties of feature extraction and slow data processing in fault diagnosis of rolling bearings, a fault diagnosis method for rolling bearings based on five machine learning algorithms and only four simple features is proposed. Firstly, the time-domain signals of bearing vibration signals of different fault types are analyzed, and four simple features of time-domain signals are extracted and input into the classification model. Then, machine learning algorithm is used to classify and diagnose the fault of rolling bearings. The experimental results show that the machine learning method is simpler and more effective than the traditional bearing fault diagnosis method. The research content provides a reference for the future study of bearing fault diagnosis using machine learning classification algorithm.