【摘要】 针对传统故障诊断方法耗时长、人工成本高且工作效率低以及现代故障诊断过于复杂的问题,提出了一种直接对振动信号进行关键特征筛选提取的方法,并结合经典的C4.5算法、Cart算法、BP算法和SVM算法对滚动轴承进行在线的故障诊断。研究结果表明,基于C4.5算法、Cart算法、BP算法和SVM算法模型的诊断方法均可对滚动轴承在运行过程中是否出现了故障以及出现了什么故障进行较为精准的识别和分类,且具有较高的准确率,可以达到很好的故障监测效果以及故障诊断效果。基于SVM算法的故障诊断模型诊断准确率优于其他三种算法,更加适用于滚动轴承的故障诊断。更多还原

【Abstract】 To addresses the problems of time-consuming,high labour cost and low efficiency of traditional fault diagnosis methods as well as the problem that modern fault diagnosis is too complex.In view of the time-consuming,high labor cost and low officiency of traditional fault diagonosis methods and the complexity of modern fault diagonosis,a method is proposed to directly extract key features from vibation signals and combine the classical C4.5 algorithm,Cart algorithm,BP algorithm and SVM algorithm to perform online fault diagonosis of rolling bearings.The research results show that the diagnosis methods based on C4.5 algorithm,Cart algorithm,BP algorithm and SVM algorithm model can identify and classify whether and what faults occur in the rolling bearing during operation with high accuracy,and can achieve good fault monitoring and fault diagnosis effect.The diagnosis accuracy of the fault diagnosis model based on SVM algorithm is better than the other three algorithms and is more suitable for the fault diagnosis of rolling bearings.更多还原