基于粒化散布熵和SSA-SVM的轴承故障诊断  

Bearing Fault Diagnosis Based on Granulation Dispersion Entropy and SSA-SVM

作　　者：叶震 李琨[1] YE Zhen;LI Kun(School of Information Engineering and Automation,Kunming University of Technology,KunmingYunnan 650500,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南昆明650500

出　　处：《机床与液压》2022年第22期157-162,共6页Machine Tool & Hydraulics

摘　　要：针对轴承故障振动信号在单一尺度下提取故障特征信息不完备,导致故障诊断识别率较低的问题,提出基于粒化散布熵(FIG-DE)和麻雀搜索算法(SSA)参数优化的支持向量机(SVM)的轴承故障诊断方法。利用模糊信息粒化对轴承振动信号进行粒化处理,得到f_(Low)、f_(R)、f_(tp)3个尺度下的模糊信息粒;分别计算3组信号的散布熵;将所得的熵值组成特征向量矩阵,输入SSA-SVM进行轴承故障分类。结果表明:利用SSA-SVM进行滚动轴承故障诊断,准确率有明显的提高。Aiming at the problem that the fault feature information of rolling bearing fault vibration signals is not fully extracted at a single scale, which results in low fault diagnosis recognition rate, a bearing fault diagnosis method was proposed based on granular dispersion entropy(FIG-DE) and SSA-SVM.The fuzzy information granulate was used to granulate the bearing vibration signal, and the fuzzy information granulation in three scales of f_(Low),f_(R) and f_(t-p) was obtained;the dispersion entropy was calculated for the three groups of signals;the obtained entropy was composed of eigenvector matrix and input into SSA-SVM to carry out the bearing fault classification.The results show that the accuracy of SSA-SVM is significantly improved.

关 键 词：模糊信息粒化 散布熵 支持向量机 故障诊断 

分 类 号：TH165.3[机械工程—机械制造及自动化]