基于人工蜂群算法优化VMD的旋转机械故障诊断方法  

Rotating Mechanical Fault Diagnosis Method Based on VMD Optimized by Artificial Bee Colony Algorithm

作　　者：朱兴统[1,2] Zhu Xingtong(School of Automation,Guangdong University of Technology,Guangzhou 510006,China;School of Computer,Guangdong University of Petrochemical Technology,Maoming 525000,China)

机构地区：[1]广东工业大学自动化学院,广州510006 [2]广东石油化工学院计算机学院,广东茂名525000

出　　处：《煤矿机械》2020年第4期159-161,共3页Coal Mine Machinery

基　　金：广东省自然科学基金项目(2018A030307038)。

摘　　要：旋转机械在运行复杂环境下振动信号复杂且有许多噪声,难以有效提取故障特征并准确诊断,为此提出一种人工蜂群优化变分模态分解(VMD)的机械故障诊断方法。首先采用人工蜂群算法优化VMD的参数,对振动信号进行变分模态分解,获得相应的本征模态函数分量;然后对各模态分量信号计算多尺度熵,构造故障特征向量;最后利用最小二乘支持向量机(LS-SVM)进行故障诊断。实验结果表明,该方法具有较高的准确率,准确率达到97.5%,可以满足旋转机械故障诊断的要求。The vibration signal of rotating machinery is complex and has a lot of noise in the complex operation environment.So,it is a challenge to extract fault features and diagnose faults accurately.Therefore proposed a machinery fault diagnosis method based on variational mode decomposition(VMD)optimized by artificial bee colony.Firstly,artificial bee colony algorithm was used to optimize the parameters of VMD,and decomposed vibration signals to obtain a series of intrinsic mode functions.Then,the multi-scale entropy of each modal component was calculated and used to construct the fault eigenvector.Finally,the least square support vector machine(LS-SVM)was used to diagnose the mechanical fault.The experimental results showed that the method had relatively high accuracy(about 97.5%),which can meet the requirements of rotating machinery fault diagnosis.

关 键 词：旋转机械 故障诊断 VMD 人工蜂群算法 LS-SVM 

分 类 号：TH133[机械工程—机械制造及自动化]