基于集成学习算法的轴承故障诊断方法研究

Research on Bearing Fault Diagnosis Method Based on Ensemble Methods

针对轴承故障诊断严重依赖人工经验、难以标准化的问题,建立了一种基于集成学习算法的轴承故障诊断模型。通过对轴承进行振动信号的采集,依据机器学习的流程,对振动信号进行特征工程处理,采用XGBoost集成算法构建故障诊断模型,并且与随机森林算法对照,验证模型效果。试验证明该模型准确率为92.89%,可以快速准确地以数据化的方式诊断出轴承当前所处的状态,降低对人工经验的依赖,提高轴承的检验效率,标准化轴承检验流程。

Aiming at the problem that bearing fault diagnosis relies heavily on manual experience and is difficult to standardize,a bearing fault diagnosis model is established.Through the collection of vibration signals on the bearings,according to the process of machine learning,the data is subjected to feature engineering processing,and the XGBoost integrated algorithm is used to build a fault diagnosis model.The model can accurately diagnose the current state of the bearing in a digital way,reduce dependence on human experience,improve bearing inspection efficiency,and standardize bearing inspection procedures.