基于AGSMD和TSVD的工业机器人柔性薄壁轴承故障诊断  

Fault diagnosis of thin-walled bearing for industrial robot based on AGSMD and TSVD

作　　者：张昌杰 马桂林 李超 Zhang Changjie;Ma Guilin;Li Chao(Zhengzhou Branch of China Nuclear Power Engineering Co.,Ltd.,Henan Zhengzhou,450052,China)

机构地区：[1]中国核电工程有限公司郑州分公司,河南郑州450052

出　　处：《机械设计与制造工程》2025年第3期77-83,共7页Machine Design and Manufacturing Engineering

摘　　要：由于柔性薄壁轴承结构特殊,振动信号复杂,因此为准确判断柔性薄壁轴承故障类别,提出一种基于自适应群稀疏模式分解和截断奇异值分解的故障诊断方法。首先采用功率谱密度方法得到预估信噪比参数,并利用电鳗觅食优化算法对自适应群稀疏模式分解方法中的惩罚因子参数及信号分量选取过程进行寻优,寻找出有效信号分量中的有效成分。然后利用截断奇异值分解方法对所得信号分量进行降噪处理,并提出一种新的奇异值能量比差分谱方法用来选取合适的重构阶数,从而准确地寻找出柔性薄壁轴承的故障特征信息。实验数据分析结果表明,所提方法能够有效地提取出柔性薄壁轴承的故障特征,实现对柔性薄壁轴承故障的准确诊断。Due to the special structure of flexible thin-wall bearings and complex vibration signal,a fault diagnosis method and this method is proposed to accurately determine the fault category of flexible singular value.First,the power spectral density method is used to obtain the estimated SNR parameter,and the electrical foraging optimization algorithm is used to optimize the penalty factor parameter and component selection process in the adaptive group sparse pattern decomposition method to find the active component in the signal.Then the cut-off singular value decomposition method is used for noise reduction,and a new singular value energy difference spectrum method is proposed to select the appropriate reconstruction order,so as to accurately find the fault feature information of flexible thin-wall bearings.The measured data analysis results show that the proposed method can effectively extract the fault characteristics of flexible thin-wall bearings,and realize the accurate fault diagnosis of flexible thin-wall bearings.

关 键 词：柔性薄壁轴承 自适应群稀疏模式分解 电鳗觅食优化算法 截断奇异值分解 

分 类 号：TH17[机械工程—机械制造及自动化]