中南大学学报（自然科学版）2011，Vol.42Issue(12)：3728-3734,7.
基于AR双谱的超声波电机故障诊断
Fault diagnosis of ultrasonic motor using AR bispectrum
摘要
Abstract
Owing to the fact that it is difficult to diagnose the fault status of ultrasonic motor when it is working, a new method of diagnosing the fault of ultrasonic motor based on high order spectra was presented. As high order spectra was sensitive to non-Gaussian colored noise, the high order cumulant technique was employed to set up the time series AR models, and bispectrum analysis was performed. The experimental results show that the energy of AR bispectra dispersive considerably and its peak value decreases significantly in the fault state compared with that in the normal operation state. Meanwhile, the contours and the slices of AR bispectra in the two states are different. Therefore, it can be used as a basis of fault diagnosing.关键词
AR模型/双谱/超声波电机/故障诊断Key words
AR model/ bispectrum/ ultrasonic motor/ fault diagnosis分类
信息技术与安全科学引用本文复制引用
陈智博,黄宜坚..基于AR双谱的超声波电机故障诊断[J].中南大学学报（自然科学版）,2011,42(12):3728-3734,7.基金项目
国家自然科学基金资助项目(50975098) （50975098）
