中文摘要：

无量纲指标作为新的理论工具应用于故障诊断研究中,虽然取得了一定的进展,但在应用时没有考虑到其他噪声干扰信号的影响,对结果分析有一定的干扰.而经验模态分解（EMD）技术能够提取出振动数据的故障特征信号,针对无量纲指标分析数据时的噪声干扰,提出了基于EMD的无量纲指标处理算法.首先对采集到的振动数据做EMD,分解出的前几个固有模态函数（IMF）分量中包含了振动数据的故障特征,去除其他噪声干扰信号的影响;其次求出含有振动数据特征信号的IMF分量的无量纲指标值,做出其无量纲指标的特征范围值;最后进行故障诊断分析.将此算法应用于旋转机械的故障诊断实验中,通过实验验证了该方法的可行性和有效性.

英文摘要：

Dimensionless index as a new theory tool has been applied in fault diagnosis study.Although there has been some progress,the application of dimensionless index often fails to consider the influence of other noise interference signal,and the results analysis will be interfered.While the fault characteristic signal of vibration data can be effectively extracted by empirical mode decomposition（EMD）.In view of the noise interference of dimensionless index analyzing data,dimensionless index processing algorithms was proposed on the basis of EMD decomposition.Firstly,EMD method was employed to decompose the collected vibration signals,the first few intrinsic mode functions（IMF）components containing the fault characteristic of vibration data were decomposed by EMD,and the effects of other noise signals was removed at the same time.Secondly,dimensionless parameter values to the IMF components with characteristic signal of vibration data were calculated,and the range of characteristic value of their dimensionless index was obtained,then fault characteristics of the equipment were diagnosed and analyzed.The proposed method was applied to fault diagnosis test analysis of rotating machinery,and its feasibility and effectiveness were verified by the experiment.