摘要
由独立成分分析(ICA)的顺序不确定性带来的源数估计和对传感器个数的估计问题使得ICA在机械故障诊断中的广泛应用受到了限制,而约束独立成分分析(CICA)充分利用了设备的先验知识作为ICA的约束条件,可以使ICA算法收敛到感兴趣的故障信号。本文提出了一种基于滚动轴承模型的约束独立成分分析(CICA)方法,该方法可以从传感器信号中快速诊断出设备是否发生了滚动轴承故障,并用仿真和实验验证了该方法在滚动轴承故障诊断中的有效性。
The order ambiguity of independent component analysis (ICA ) makes it very difficult to estimate numbers of sources and sensors.Constrained independent component analysis (CICA)can use some prior knowledge of equipments as a constraint of ICA to make the ICA algorithm converge to fault signals to be interested.Here,a model-based constrained independent component analysis method for fault diagnosis of rolling element bearings was presented.Its effectiveness was verified with simulations and tests.
出处
《振动与冲击》
EI
CSCD
北大核心
2015年第8期66-70,共5页
Journal of Vibration and Shock
基金
国家自然科学基金项目(U1304523
51205371)
河南理工大学博士基金项目(648491)
关键词
独立成分分析
约束独立成分分析
盲源分离
机械故障诊断
滚动轴承
independent component analysis (ICA)
constrained independent component analysis (CICA)
blindsource separation
machine fauh diagnosis
rolling element bearing