摘要
针对非均匀周期多采样率系统,在状态估计为已知的情况下,提出了基于奇异值分解的模型参数的最小二乘辨识方法.首先,根据系统的连续时间状态空间模型,在满足因果关系基础上,推导了含有提升变量的离散状态空间模型.然后,为了克服辨识误差积累和传递,采用基于奇异值分解的递推最小二乘方法确定模型参数.最后,仿真结果表明提出方法的有效性.
The least-squares method is proposed via the model decomposition （SVD） specific to the non-uniformly sampling system under estimates are known. Firstly, the discrete state-space model with the lifting continuous state-space model on the basis of realizing accumulation and the transmission of the identification singular values decomposition is developed simulation results show the effectiveness of parameter of singular value the assumption that the state variables is derived from the the causality constraints. Secondly, to overcome the errors, the recursive least-squares method based on to determine the parameter of the the proposed method.
出处
《集宁师范学院学报》
2014年第1期93-99,共7页
Journal of Jining Normal University
基金
国家自然科学基金"用显卡通用计算方法设计超导/铁磁异质结构的磁通量子器件"(项目编号:11064008)
关键词
状态空间模型
奇异值分
多采样率系统
非均匀采样
state-space model
singular value decomposition
multi-rate sampled systems
non-uniform sampling
