基于VMD的自适应复合多尺度模糊熵及其在滚动轴承故障诊断中的应用

发布时间：2018-01-08 00:22

本文关键词：基于VMD的自适应复合多尺度模糊熵及其在滚动轴承故障诊断中的应用　出处：《航空动力学报》2017年07期 　论文类型：期刊论文

【摘要】：提出了一种基于自适应多尺度模糊熵、ILS(迭代拉普拉斯得分)特征选择和粒子群优化支持向量机(PSO-SVM)的滚动轴承故障诊断方法。该方法采用变分模态分解对振动信号进行分解和重构,并计算重构信号的复合多尺度模糊熵;同时采用迭代拉普拉斯得分选择敏感故障特征,并将特征选择结果输入到基于粒子群优化支持向量机的多故障分类器进行识别。将提出的方法应用于滚动轴承试验数据分析。结果表明:该方法对试验数据的故障识别率为100%。并将基于ILS特征选择方法与基于SFS(sequential forward selection)特征选择进行了对比,表明基于SFS特征选择的最高识别率为92.86%,而基于ILS特征选择的故障识别率达到100%。
[Abstract]:Propose a method based on adaptive multi scale fuzzy entropy, ILS (iterative Laplasse scoring) feature selection and particle swarm optimization support vector machine (PSO-SVM) of the rolling bearing fault diagnosis method. This method adopts variational modal decomposition and reconstruction of vibration signal, multi scale calculation and signal reconstruction of fuzzy entropy; at the same time using the iterative Laplasse scoring option sensitive fault feature, and the feature selection result is input to the recognition of particle swarm optimization in multi fault classifier based on support vector machine. The proposed method is applied to rolling bearing test data analysis. The results show that the method of fault identification of the test data rate for 100%. and ILS feature selection method based on and based on SFS (sequential forward selection) were compared for feature selection, feature selection of SFS showed the highest recognition rate based on 92.86%, and based on ILS feature selection The selective fault recognition rate is up to 100%.

【作者单位】：安徽工业大学机械工程学院;
【基金】：国家自然科学基金(51505002) 安徽省高校自然科学研究重点资助项目(KJ2015A080) 安徽工业大学研究生创新研究基金(2016062)
【分类号】：TH133.33
【正文快照】： 引用格式:郑近德,姜战伟,代俊习,等.基于VMD的自适应复合多尺度模糊熵及其在滚动轴承故障诊断中的应用[J].航空动力学报,2017,32(7):1683-1689.ZHENG Jinde,JIANG Zhanwei,DAI Junxi,et al.VMD based adaptive composite multiscale fuzzyentropy and its application to faul

【参考文献】