非均匀采样数据系统的AM-RLS辨识方法及仿真研究

AM-RLS Identification and Simulation Studies for Non-uniformly Sampled-data Systems

推导了一类非均匀采样数据系统的提升状态空间模型,进而得到其对应的输入输出表达。针对辨识模型信息向量中存在未知的中间变量,提出非均匀采样系统基于辅助模型的递推最小二乘(AM-RLS)辨识方法,基本思想是用辅助模型的输出代替信息向量中的未知变量。并对算法进行了计算机数值仿真研究,来说明提出辨识算法的有效性。

The lifted state-space models for a class of non-uniformly sampled-data systems from their continuous-time systems were derived and the corresponding input-output relationship was obtained. For unmeasurable information vectors arising in the identification models, an auxiliary model based recursive lest squares identification algorithm was proposed to estimate the parameters of the non-uniformly multirate systems using the auxiliary model technique. The basic idea is to replace the unknown variables in the information vector with the outputs of the auxiliary model. Finally, the numerical simulation studies on computers show the merits of the proposed algorithm.