基于转速自适应多阶FRFT滤波提取瞬变工况齿轮微弱故障特征  

Extraction of gearbox fault features under transient-conditions based on self-adaptive segmentation and multilevel fractional fourier transformation filtering

作　　者：李枫[1] 梅检民[1,2] 肖云魁[1] 杨青乐 曾锐利[1] 

机构地区：[1]军事交通学院汽车工程系,天津300161 [2]军械工程学院火炮工程系,石家庄050003

出　　处：《振动与冲击》2013年第21期158-163,共6页Journal of Vibration and Shock

摘　　要：提出了一种基于转速信号自适应分段的多阶FRFT滤波方法,并应用于提取瞬变工况下变速器齿轮微弱故障特征。首先,根据设定的振动信号频率曲线曲率阈值将目标档位啮合频率时频曲线自适应分为若干段,使得每段内的信号频率近似线性变化;然后,在各分段内进行最小二乘拟合确定相应的FRFT最佳阶次,并在各段信号的最佳分数阶域进行滤波,实现基于转速自适应分段的多阶FRFT滤波。采用该方法分析实测变速器瞬变工况振动信号。结果表明,基于转速信号自适应分段确定多阶FRFT的最佳阶次,准确、快速、自适应性好;多阶FRFT滤波能够有效分离出瞬变工况下的啮合频率分量,隔离其他干扰;对分离出的啮合频率分量进行阶次包络解调分析,能有效提取出齿轮微弱故障特征。A method of multilevel fractional fourier transformation （FRFT） filtering based on self-adaptive order of FRFT was obtained for each segment with the least square curve fitting and the vibration signal was analyzed with FRFT to filter the interference components and noise. The test results showed that the best order of multilevel FRFT calculated based on self-adaptive segmentation of a rotating speed signal is accurate, quick and adaptive; the target order component under transient conditions can be extracted using multilevel FRFT with best order effectively and other interferences are isolated; gear fault features can be obtained with envelope demodulation of the extracted sole order component.

关 键 词：自适应分段 多阶FRFT滤波 瞬变工况 特征提取 

分 类 号：TH132.41[机械工程—机械制造及自动化]