基于迭代广义解调算法的变转速滚动轴承复合故障特征的提取  

Multi-fault feature detection of rolling element bearing by an iterative generalized demodulation algorithm under time-varying rotational speed

作　　者：赵德尊 李建勇[1,2] 程卫东 温伟刚[1] ZHAO Dezun;LI Jianyong;CHENG Weidong;WEN Weigang(School of Mechanical Electronic and Control Engineering,Beijng Jiaotong University,Beijing 100044,China;Key Laboratory of Vehicle Advanced Manufacturing,Measuring and Control Technology,Ministry of Education,Beijing 100044,China)

机构地区：[1]北京交通大学机械与电子控制工程学院,北京100044 [2]载运工具先进制造与测控技术教育部重点实验室,北京100044

出　　处：《振动与冲击》2018年第4期177-183,190,共8页Journal of Vibration and Shock

基　　金：中央高校基本科研业务费专项(2016YJS129)

摘　　要：变转速工作模式使得本来就互相干扰、彼此联系的滚动轴承复合故障特征的提取更加困难。为此提出了基于迭代广义解调算法的变转速滚动轴承复合故障特征提取方法。该方法根据复合故障轴承信号包络时频谱中代表故障特征频率的时频曲线的突出性,结合迭代广义解调算法可以将特定时频曲线转换成平行于时间轴的直线这一特点,直接对滚动轴承振动信号中的特定成分进行分析和提取。整个算法由以下四部分组成:对同步测取的故障轴承转速脉冲信号进行处理得到转频曲线,根据转频曲线以及目标轴承的故障特征系数确定迭代广义解调算法需要的相位函数;其次,对故障轴承信号进行包络分析获取包络信号;根据计算的相位函数对故障轴承包络信号进行迭代广义解调;对解调信号进行频谱分析,通过分析频谱中独立峰的位置对滚动轴承的健康状况进行判断。仿真与实验结果表明,该算法可以消除转速变化对滚动轴承复合故障特征的影响,有效实现复合故障特征的识别和提取。Due to variable rotational speed,it is difficult to extract multi-fault features of rolling element bearing,which may tightly couple together and interfere with each other.As such,a new multi-fault feature extraction method based on the iterative generalized demodulation algorithm was proposed.We utilized the time-frequency characteristics of the compound faults bearing signal and the merits of the generalized demodulation algorithm which can transform an interested curved time-frequency component of the non-stationary signal into linear path paralleling to the time axis to extract the multi-fault features directly.The proposed method can be summarized as four parts.(1)The equation of rotational frequency curve is obtained by rotational speed pulse signal.The phase functions and phase points are calculated based on the fault characteristic coefficient and the rotational frequency equation.(2)The envelope signal is calculated by the Hilbert transform.(3)The envelope signal is demodulated by the iterative generalized demodulation algorithm.(4)The frequency spectrum of the demodulated signal is obtained to detect fault feature.The studies of simulated and experimental multi-fault bearing signals validate that the proposed method is reliable to diagnose multi-fault bearings under time-varying rotational speed.

关 键 词：滚动轴承 复合故障特征提取 变转速 迭代广义解调算法 

分 类 号：TH113.1[机械工程—机械设计及理论]