摘要
针对风机齿轮箱振动信号的故障特征提取与故障诊断问题,文章提出了一种基于MEEMD信号分解、样本熵和KFCM的齿轮箱故障诊断方法。首先,采用一种改进的集合经验模态分解方法(MEEMD)对采集的齿轮箱振动信号进行分解,得到了多个本征模态函数(IMF)分量;然后,计算每个IMF分量的样本熵作为齿轮箱故障诊断的特征向量;最后,使用核化的模糊聚类算法(KFCM)对齿轮箱故障样本进行聚类。通过实验数据对比表明:基于MEEMD-KFCM算法的风机齿轮箱故障诊断方法可以更加有效地识别齿轮箱故障。
Aiming at the problem of fault feature extraction and fault diagnosis of vibration signal of wind turbine's gearbox,this paper proposes a fault diagnosis method of gearbox based on MEEMD,sample entropy and KFCM.Firstly,a modified EEMD method(MEEMD)is adopted to decompose the vibration signal of the wind turbine's gearbox,and anumber of IMF components are obtained.Then,sample entropy of each IMF component iscalculated as the characteristic vector of gear box fault diagnosis.Finally,Kernel-based fuzzy C-means clustering method(KFCM)is used to cluster the gearbox fault samples.The comparison of experimental data shows that the fault diagnosis method of wind turbine's gearbox based on MEEMD-KFCM algorithm can identify gearbox fault more effectively.
作者
郑坤鹏
丁云飞
Zheng Kunpeng;Ding Yunfei(School of Electrical Engineering,Shanghai Dianji University,Shanghai 200240,China)
出处
《可再生能源》
CAS
北大核心
2020年第9期1192-1196,共5页
Renewable Energy Resources
基金
国家自然科学基金(11302123)
上海市浦江人才计划(15PJ402500)。