基于终端凸集约束的新MPC控制器

New MPC Controller Based on a Terminal Convex Set Constraint

针对一类离散系统,研究了带有终端约束凸集的M PC控制问题.通过离线设计一组椭圆不变集,并将其组合成一个终端约束凸集,其中凸集参数作为在线优化变量.在线运算时,根据实际的终端状态即时地选择合适的终端不变集,从而有效地扩大了系统的可行域.分别给出了设计M PC控制器的离线和在线算法,仿真实例说明了该方法的有效性.

For a class of discrete systems, the problem of MPC with terminal constraint convex set is investigated. A group of ellipsoidal invariant sets is designed off-line, and then constitutes a terminal constraint convex set whose coefficients are taken as on-line optimization variables. During on-line computation, an appropriate invariant ellipsoid is chosen from the terminal convex set according to the real state, which enlarges the initial feasible region efficiently. Both the off-line and the on-line algorithms are presented to design MPC controllers. A simulation example shows the validity of the proposed approach.