基于MED和ICA的滚动轴承循环冲击故障特征增强

发布时间：2019-04-25 10:49

【摘要】：针对强噪声背景下多个传感器对同一振动源同步采集多组数据的情况,提出一种基于最小熵解卷积和独立成分分析的联合降噪方法,并用于滚动轴承循环冲击故障特征的提取。利用最小熵解卷积对各传感器的信号分别进行盲解卷滤波,消除信号传递路径的影响,从噪声信号中初步提取出故障冲击特征;对各传感器的滤波信号进行独立成分分析处理,将信号进行重组后得到重构分量,进一步消除噪声成分,使故障冲击特征成分得到二次增强;选取峭度最大的最优独立成分分析重构分量并进行包络谱分析,得到诊断结果。通过仿真数据和实验室数据分析验证了该方法能够增强滚动轴承的循环冲击特征,便于识别故障类型。
[Abstract]:Based on minimum entropy deconvolution and independent component analysis, a joint de-noising method based on minimum entropy deconvolution and independent component analysis is proposed for simultaneous acquisition of multiple sets of data from the same vibration source by multiple sensors in the background of strong noise. The method is used to extract the features of rolling bearing cyclic shock fault. The minimum entropy deconvolution is used to filter the signals of each sensor in blind deconvolution to eliminate the influence of the signal transmission path and to extract the fault impact characteristics from the noise signals. The filter signal of each sensor is analyzed and processed by independent component analysis, the reconstructed component is obtained after the signal is reorganized, the noise component is further eliminated, and the characteristic component of fault shock is second enhanced. The reconstructed component of the optimal independent component analysis with the largest kurtosis is selected and the envelope spectrum analysis is carried out to obtain the diagnosis result. The simulation data and laboratory data show that the proposed method can enhance the cyclic impact characteristics of rolling bearings and facilitate the identification of fault types.
【作者单位】：华东交通大学机电与车辆工程学院;
【基金】：国家自然科学基金资助项目(51665013,51265010) 江西省科协重点活动资助项目(赣科协字2014-154) 江西省青年科学基金资助项目(20161BAB216134) 江西省研究生创新专项资金资助项目(YC2015-S239)~~
【分类号】：TH133.33

【相似文献】