Model reduction for Lur'e type systems
msra(2009)
摘要
In the design of complex high-tech systems, predictive mod- els are typically of high order. Model reduction can be used to obtain a low-order approximation of these models, allow- ing for efficient analysis or control design. Several method s exist for the reduction of linear systems. Balanced trunca- tion (1) is among the most popular, since it provides guaran- tees on both stability and error bounds of the reduced-order model. Nonlinear model reduction techniques such as non- linear balancing or proper orthogonal decomposition lack such an error bound or require simulations of the full-order model. In this work, (linear) balanced truncation is applie d as a tool for model reduction of Lur'e type systems, repre- senting a specific class of nonlinear systems. Next, condi- tions for stability of the reduced-order model are stated an d an error bound is presented.
更多查看译文
关键词
linear system,nonlinear system,type system
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络