Neural Network-Based Robust Control for Uncertain Systems with Nonlinearity
The linear difference inclusion (LDI) design methodology of neural network-based control systems is applied for a class of uncertain systems with nonlinear component. By means of quadratic stability theory of uncertain linear systems, robustness of a guaranteed LQ index controller is discussed. The robust analysis and design approach is formulated in a set of linear matrix inequalities (LMIs). An application example shows that the LQ control systems will work well against not only bounded error between a nonlinear dynamical system and a multi-layer neural network, but also against the bounded time-varying parameter uncertainties.
Xiaoli Luan Fei Liu
Institute of Automation Southern Yangtze University Wuxi214122,Jiangsu, China
国际会议
青岛
英文
2006-07-21（万方平台首次上网日期，不代表论文的发表时间）