独立分量分析在转子多源故障自适应时频分析中的应用  

Application of Independent Component Analysis in Adaptive Time-Frequency Analysis of Rotor Multi-Source Fault

作　　者：李斌[1] 罗敏[1] 冯锦春[1] 肖健[1] 

机构地区：[1]四川工程职业技术学院,四川德阳618000

出　　处：《机械设计与制造》2016年第1期73-75,79,共4页Machinery Design & Manufacture

基　　金：四川工程职业技术学院2014-2015年度科研专项项目(YJ2014KY-17);云南省教育厅科学研究基金(2010Y379)

摘　　要：针对短时傅里叶变换(STFT)和Wigner-Ville分布等传统时频分析方法对转子多源故障分析效果不理想的问题,提出了一种基于独立分量分析(Independent Components Analysis,ICA)和自适应时频分析结合的抗干扰转子故障非平稳信号分析方法。该方法首先对采集的信号进行ICA处理得到独立分量,然后对独立分量进行自适应时频分析,提取出故障频率信息,最后对独立分量进行阶比跟踪分析得到的瀑布图,更加充分地验证了试验结果。仿真和试验对转子不平衡和地脚螺栓松动故障进行了研究,分析结果验证了本方法的可行性和有效性。Traditional time-frequency analysis methods such as short-time Fourier transform（STFT） and Wigner-Ville distribution are not effective for the multiple fault analysis in the rotor systems. A novel anti-interference non-stationary signal analysis method used for the rotor fault detection is proposed,in which the Independent Components Analysis（ICA） is incorporated with the adaptive time-frequency analysis. The system signals are firstly dealt by means of the ICA providing their independent components. Then the obtained independent components are further analyzed via the adaptively time-frequency analysis to derive the fault information. Finally,their independent components are dealt by order tracking analysis providing their waterfall diagrams,the results of the test are more fully verified by waterfall diagrams. Simulation and experimental results on the rotor imbalance and flexible foundation bolt faults are included to verify the feasibility and validity of the proposed scheme.

关 键 词：独立分量分析 自适应时频分析 转子系统 故障诊断 

分 类 号：TH16[机械工程—机械制造及自动化]