摘要
针对港口起重机轴承早期微弱故障特征易被强噪声淹没导致难以提取问题,结合广义似然比(generalized likelihood ratio,GLR)和变分模态分解(variational modal decomposition,VMD)算法的优势,提出了一种基于GLR引导VMD的算法。同时,针对传统包络谱检测识别故障需要依靠专业知识识别的情况,根据上述算法提出了一种基于包络谱的故障指标故障频率比(ratio of fault frequency,RFF)。利用仿真试验信号和实际数据对所提出方法进行验证,并且与其他VMD优化方法进行比较,研究结果表明,所提出算法能够准确提取高噪声强干扰下信号下滚动轴承的故障特征,所提出故障指标RFF能够实现微弱故障诊断。
Aiming at the problem that the early weak fault features of harbor crane bearings are easily submerged by strong noise,an algorithm based on the combined advantages of the generalized likelihood ratio(GLR) and variational modal decomposition(VMD) was proposed.A fault index ratio of fault frequency(RFF) based on envelope spectrum was proposed by virtue of the above algorithm in view of the fact that the traditional envelope spectrum detection and fault identification depend on professional knowledge.The proposed method was validated by experimental simulation signals and real data,and compared with other VMD optimization methods.The results show that the proposed algorithm can accurately extract the fault features of rolling bearings under the environment high noise and strong interference signals,and the proposed fault index RFF can realize weak fault diagnosis.
作者
冯文宗
张氢
张建群
孙远韬
秦仙蓉
FENG Wenzong;ZHANG Qing;ZHANG Jianqun;SUN Yuantao;QIN Xianrong(School of Mechanical Engineering,Tongji University,Shanghai 201804,China)
出处
《振动与冲击》
EI
CSCD
北大核心
2023年第22期264-272,共9页
Journal of Vibration and Shock
基金
国家自然科学基金面上项目(52075389)。
关键词
滚动轴承
故障诊断
统计模型
变分模态分解(VMD)
特征提取
rolling bearing
fault diagnosis
statistical model
variational mode decomposition(VMD)
feature ex-traction
