一种带约束的广义预测控制器

A Generalized Predictive Controller with Constraints

传统的广义预测控制算法,计算量大,求解复杂,对有约束的控制问题,求解更是麻烦。本文针对这一问题,提出一种带约束的广义预测控制算法,通过对未来控制序列的离线近似计算以及考虑到约束情况来精确求解当前时刻的控制量。该算法简单,避免了求解Diophantine方程及逆矩阵,大大减小了计算量,仿真结果表明,该算法控制性能良好。

It is difficult to calculate the traditional generalized predictive control algorithm(GPC),especially for constraints problems.To avoid these problems,a new generalized predictive control with constraints is put forward.The algorithm calculates exactly present control sequences by off-line approximates.This algorithm is simple and not necessary to solve the Diophantine matrix,so that the on-line amount of computation is reduced greatly.The simulation results show that this algorithm has good control performance.