基于BEMD的二元滤波器及其在转子轴心轨迹提纯中的应用

摘要

在二元经验模态分解(Bivariate Empirical Mode Decomposition,BEMD)的基础上设计了一种二元时空滤波器(Bivariate-Time-Space Filter Banks,BTSFB),并将其用于转子轴心轨迹提纯.首先,介绍了BEMD原理;其次,根据BEMD方法获得的复固有模态函数(Complex Intrinsic Mode Functions,CIMFs)按旋转速度从高到低排列的特点构造了具有良好自适应能力的BTSFB;然后,联合BEMD和全矢谱方法确定BTSFB用于转子轴心轨迹提纯的CIMFs的阶数,以便提纯轴心轨迹.转子不对中和动静部件碰摩的实验结果表明BTSFB不仅可提纯轴心轨迹,还可进行轴心轨迹的三维显示以及判断轴心轨迹运动方向.%Bivariate-time-space filter bank (BTSFB) was designed based on the bivariate empirical mode decomposition (BEMD) and applied to purify rotor shaft center trajectory. Firstly,the principle of BEMD was introduced. Secondly,the BTSFB with a good adaptive ability was constructed with complex intrinsic mode functions (CIMFs) obtained by BEMD and arranged according to rotating speed from high to low. Finally,the BEMD combined with the total vector spectrum method was used to determine the order of CIMFs in the BTSFB so as to purify rotor shaft center trajectory. The test results of rotor misalignment and rubbing between dynamic parts and static ones showed that the BTSFB can not only purify rotor shaft center trajectory,but also display the 3-D rotor shaft center trajectory and judge the trajectory motion direction.