基于BAACMD-NGO-TMSST的变转速滚动轴承故障诊断  

Fault Diagnosis of Variable Speed Rolling Bearings Based on BAACMD-NGO-TMSST

作　　者：吴欢 马洁 WU Huan;MA Jie(Mechanical Electrical Engineering School,Beijing Information Science&Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学机电工程学院,北京100192

出　　处：《机床与液压》2024年第22期227-240,共14页Machine Tool & Hydraulics

基　　金：国家自然科学基金面上项目(61973041)。

摘　　要：在变转速工况下,滚动轴承时变故障特征提取困难,且时间重分配的多重同步压缩变换(TMSST)易受噪声干扰,其相关参数的确定不具备自适应性。针对此问题,提出一种基于BAACMD-NGO-TMSST的变转速滚动轴承故障诊断方法。通过BAACMD将故障信号分解为多个分量,采用基尼指数和包络熵作为综合指标选取最优分量,从而去除噪声干扰;采用北方苍鹰优化算法(NGO)对TMSST进行参数寻优;最后,利用优化后的TMSST对最优分量进行时频分析,并计算最大TF包络谱(TFES)提取故障特征。通过仿真信号和渥太华轴承数据集,验证了所提方法的可行性和有效性;与其他降噪方法对比,BAACMD在降噪方面具有优越性;与其他时频分析方法对比,所提方法具有更好的特征提取效果。It is difficult to extract time-varying fault features of rolling bearings under variable speed conditions,and the time-reassigned multisynchrosqueezing transform(TMSST)is susceptible to noise interference,and the determination of its related parameters is not adaptive.Aiming at the problems,a fault diagnosis method for rolling bearings with variable speed based on BAACMD-NGO-TMSST was proposed.The fault signal was divided into multiple components by BAACMD,and the Gini index and envelope entropy were used as comprehensive indexes to select the optimal components,so as to remove the noise interference.The northern goshawk optimization algorithm(NGO)was used to optimize the parameters of the TMSST.Finally,the optimal components were subjected to time-frequency analysis by using the optimized TMSST,and the maximum TF envelope spectra(TFES)was calculated to extract fault features.The feasibility and effectiveness of the proposed method were verified by simulated signals and Ottawa bearing dataset.Compared with other noise reduction methods,BAACMD is superior in noise reduction;compared with other time-frequency analysis methods,the proposed method has better feature extraction effect.

关 键 词：变转速 故障诊断 带宽感知自适应线性调频模态分解(BAACMD) 时间再分配多重同步压缩变换(TMSST) 

分 类 号：TH165.3[机械工程—机械制造及自动化]