关键词: 采样数据模型, 质量-弹簧-阻尼, 零动态, 非线性系统, 系统辨识

Abstract: The approximate sampled-data models and discretization properties are derived for nonlinear multivariable mass-spring-damper mechanical vibration systems when the corresponding relative degrees of continuous plant are both two. The proposed sampled-data models use a more sophisticated derivative approximation than the simple Euler approach, and the local truncation error and global truncation error are also represented, respectively. An insightful interpretation of the given sampled-data models can be made in terms of an explicit characterization of nonlinear sampling zero dynamics which have no continuous-time counterpart. More importantly, nonlinear discrete-time controller design is also represented based on the above sampled-data model and zero dynamics. The ideas presented here generalize well-known results from the linear case to nonlinear plants. Finally, we also explore the implications of these results in nonlinear system identification.

Key words:

Sampled-data model, mass-spring-damper, zero dynamics, nonlinear system, system identification