中文摘要：

滚动轴承故障信号具有较强的非平稳特性,并且极易受齿轮等噪源污染,故障特征信息微弱,特别当滚动轴承处于故障早期,上述问题尤为严重。针对这一难点,提出基于瞬时包络尺度谱熵的滚动轴承早期故障奇异点识别及特征提取方法。应用重分配尺度谱对轴承的包络信号进行时频分解,计算每一时刻的功率谱熵,以获取信号的瞬时包络尺度谱熵（Instantaneous envelope scalogram entropy,IESE）,则信号IESE曲线发生畸变的位置,即是轴承故障表征最为明显的时刻,进而可以提取轴承故障信号的最优故障表征时段（Optimal fault characterization phase,OFCP）,应用包络解调和包络尺度谱分析OFCP,以提取轴承故障特征频率。实测信号分析结果表明,该方法能有效提取轴承故障早期的微弱故障特征信息。

英文摘要：

The non-stationary characteristics of the bearing fault signal is strong,and the signal is easy to be contaminated by the gear and other noise sources,the fault feature information is weak,especially when the rolling bearing is at the early fault stage,the above problem is particularly serious.To solve the problem above,a method of singular point recognition and feature extraction for the incipient bearing fault is presented based on instantaneous envelope scalogram entropy（IESE）.The envelope signal is decomposed by reassigned wavelet scalogram,and the power spectrum entropy of each moment is calculated to obtain the instantaneous envelope scalogram entropy of the signal.The distortion position of the IESE curve of the signal is the most obvious time of bearing fault characterization.And then the optimal fault characterization phase（OFCP） of bearing fault signal could be extracted,and the OFCP could be used to extract the characteristic frequency of bearing fault.The results showed that the method could effectively extract the weak fault feature information of the signal of early fault bearing.1