基于XGBoost和自回归模型的轴承故障诊断和预警方法研究  

Research on Bearing Fault Diagnosis and Warning Based on XGBoost and Autoregressive Algorithm

作　　者：郝俊虎 胡毅[1,3] HAO Jun-hu;HU Yi(Shenyang Institute of Computing Technology,Chinese Academy of Sciences,Shenyang 110168,China;School of Computer and Control Engineering,University of Chinese Academy of Sciences,Beijing 101408,China;不详)

机构地区：[1]中国科学院沈阳计算技术研究所,沈阳110168 [2]中国科学院大学计算机与控制学院,北京101408 [3]沈阳高精数控智能技术股份有限公司,沈阳110168

出　　处：《组合机床与自动化加工技术》2020年第2期140-142,157,共4页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：2017国家智能制造综合标准化项目:数控装备故障信息数据字典标准研制及试验验证(20172150299)

摘　　要：针对轴承在数控车间生产中易发生故障且对轴承故障预警困难的问题,提出了一种基于XGBoost算法和AR(I)MA自回归模型的数据驱动的故障诊断和预警方法。首先使用XGBoost算法将轴承的历史数据划分为正常、滚珠故障、外圈故障和内圈故障4种状态,然后使用AR(I)MA模型来预测轴承在未来一段时间内的振动信号变化,再将预测出的振动信号进行降噪和特征提取后输入到训练好的XGBoost中进行故障诊断。使用PRONOSTIA平台采集的轴承工作数据进行实验,结果表明,文章方法可以准确预测出轴承短期内的振动信号并诊断出可能发生的故障,证明了该方法在轴承的故障诊断和预警中的可行性与正确性。Aiming at the problem that the bearing is prone to failure in the production of CNC workshop and difficult to predict the bearing failure,a data-driven fault diagnosis and early warning method based on XGBoost algorithm and AR(I)MA autoregressive model is proposed.First,the XGBoost algorithm is used to divide the bearing historical data into four states:normal,ball fault,outer ring fault and inner ring fault.Then the AR(I)MA model is used to predict the vibration signal change of the bearing in the future,and then the predicted vibration signal is denoised and extracted,and then input into the trained XGBoost for fault diagnosis.The bearing working data collected by the PRONOSTIA platform is used for experiments.The results show that the proposed method can accurately predict the vibration signal of the bearing in the short term and diagnose the fault,which proves the feasibility and correctness of the method in fault diagnosis and early warning of the bearing.

关 键 词：XGBoost ARMA 轴承 故障诊断 故障预警 

分 类 号：TH162[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]