摘要
针对非平稳时变信号,提出一种基于隐Markov模型(HMM)的机械振动源数估计方法。该方法结合隐Markov模型理论与自相关测定,通过比较不同模型的信度来确定信源的个数。实验结果表明该方法能够有效地估计出非平稳时变信号的信源个数,为机械振动故障诊断中的振动源分析提供了方法保障。
To estimate the number of mechanical vibration sources from nonstationary and timevarying signals,a new method based on hidden markov models （HMM） is proposed. The method combined hidden Markov models with automatic relevance determination to determine the number of sources through comparing the reliability of different models. The experimental results show that the method can effectively estimate the number of nonstationary and time varying signal sources,and this work is helpful to analyze the mechanical vibrations in fault monitoring system.
出处
《机械工程师》
2013年第5期13-15,共3页
Mechanical Engineer
关键词
隐MARKOV模型
自相关测定
故障诊断
源数估计
hidden markov models
automatic relevance determination
fault diagnosis
source number estimation
