Sensitivity of nonlinear dynamic modeling
暂无分享,去创建一个
Abstract The paper discusses the sensitivity of modeling with respect to the discretization and structural realization of sampled continuous nonlinear dynamic processes. As a tool for the sensitivity analysis equivalent input-output equivalent structures are introduced. The aim of the paper is to explain the relationships behind the modeling error and to offer a systematic approach to discuss input/output equivalent structures and to show how to reduce sensitivity by choosing the proper structure. The results are verified by simulation examples, as well.
[1] Yuh-tay Sheen,et al. Neural network for system identification , 1992 .
[2] Kumpati S. Narendra,et al. Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.
[3] Susanne Ernst,et al. Identification with Dynamic Neural Networks - Architectures, Comparisons, Applications , 1997 .