PMA-ASTFA及其在齿轮裂纹定量诊断中的应用

发布时间：2018-09-11 15:55

【摘要】：目前对齿轮裂纹的诊断研究多采用定性诊断,而工程实际中往往更关注定量诊断。由于齿轮裂纹信号往往表现出非线性非平稳特征,处理这类信号通常采用时频分析。自适应最稀疏时频分析(Adaptive and Sparsest TimeFrequency Analysis,简称ASTFA)是一种新的时频分析方法,相比于经验模态分解(Empirical Mode Decomposition,简称EMD)方法,ASTFA方法能更好地抑制端点效应和模态混淆,但ASTFA方法也存在分解得到的分量排列不规律的缺陷,从而给特征提取时分量的选择带来困难。针对这一问题,提出了一种改进ASTFA算法,即基于主模态分析(Principle Mode Analysis,简称PMA)的自适应最稀疏时频分析(PMA-ASTFA)方法,该方法可以根据所选择的故障特征参数(一个或多个)对内禀模态函数(Intrinsic Mode Function,简称IMF)分量进行排序。根据齿轮故障实验台建立齿轮动力学模型,选择对齿轮裂纹敏感的故障特征参数,再把PMA-ASTFA方法用于实测的齿轮裂纹故障信号处理。实验信号的分析结果表明,提出的方法可以有效地实现齿轮裂纹故障的定量诊断。
[Abstract]:At present, qualitative diagnosis is often used in the diagnosis of gear cracks, while quantitative diagnosis is often paid more attention in engineering practice. Because gear crack signals often show nonlinear and non-stationary characteristics, time-frequency analysis is usually used to deal with these signals. Adaptive most sparse time-frequency analysis (ASTFA) is a new time-frequency analysis method. Compared with the empirical mode decomposition (Empirical Mode Decomposition,) method, the EMD method can better suppress the endpoint effect and modal confusion. However, the ASTFA method also has the defect of irregular arrangement of the components obtained by decomposition, which makes it difficult to select the components in feature extraction. In order to solve this problem, an improved ASTFA algorithm is proposed, that is, adaptive most sparse time-frequency analysis (PMA-ASTFA) method based on main modal analysis (Principle Mode Analysis,) for PMA. The method can sort the intrinsic mode function (Intrinsic Mode Function,) components according to the selected fault feature parameters (one or more). According to the gear dynamic model of the gear fault test bench, the fault characteristic parameters sensitive to the gear crack are selected, and then the PMA-ASTFA method is used to deal with the measured fault signal of the gear crack. The analysis results of experimental signals show that the proposed method can effectively realize the quantitative diagnosis of gear crack faults.
【作者单位】：湖南大学汽车车身先进设计制造国家重点实验室;
【基金】：国家重点研发计划项目(2016YFF0203400) 国家自然科学基金资助项目(51575168,51375152) 智能型新能源汽车国家2011协同创新中心湖南省绿色汽车2011协同创新中心资助项目
【分类号】：TH132.41

【相似文献】