摘要
齿轮箱在运行的过程中存在众多的干扰直接影响到传动效率的管理水平。针对传统的支持向量机(SVM)存在收敛效率偏低的问题,设计了一种自适应局部迭代滤波(ALIF)和多尺度排列熵(MPE)的组合方法,并成功应用于齿轮箱轴承早期故障诊断领域。研究结果表明:选择前3个主成分组成特征量,输入MPE中进行训练。各状态准确识别出了故障区域,表明使用MPE作为故障识别器具备较高可靠性。分类器选择ALIF-MPE时,相比EMD-MPE获得更理想分类效果,标准差相对较小,说明已进入稳定分类阶段。该研究能够准确识别齿轮箱轴承故障程度,对提高机械传动设备的障诊断水平具有很好的理论支撑。
There are many disturbances in the process of gear operation that directly affect the management level of transmission efficiency.The traditional support vector machine(SVM)has the problem of low convergence efficiency.A combination method of adaptive local iterative filtering(ALIF)and multi-scale permutation entropy(MPE)is designed and successfully applied in the field of early fault diagnosis of gear bearings.The results show that the first 3 principal components are selected to form characteristic quantities and input into MPE for training.Each state accurately identifies the fault area,which indicates that MPE as the fault identifier has high reliability.When the classifier selects ALIF-MPE,it can obtain better classification effect than EMD-MPE,and the standard deviation is relatively small,indicating that it has entered the stable classification stage.This research can accurately identify the fault degree of gear bearing,and has a good theoretical support for improving the fault diagnosis level of mechanical transmission equipment.
作者
宋金朋
Song Jinpeng(Xinxiang Polytechnic,Xinxiang Henan 453000,China)
出处
《机械管理开发》
2025年第10期86-88,共3页
Mechanical Management and Development
关键词
轴承
故障诊断
排列熵
传动效率
bearing
fault diagnosis
permutation entropy
transmission efficiency
