摘要
动态矩阵控制（ＤＭＣ）算法是一种基于被控对象的阶跃响应、适用于渐进稳定的线性对象的预测控制算法，它具有结构清晰、算法简单的特点．但由于其参数仅间接影响控制系统的性能，与控制的快速性．稳定性、鲁棒性、抗干扰性等没有直接的解析关系可作为设计的定量依据，给设计者造成很大的困难．本文提出了运用满意解原理，通过仿真进行参数优化的方法，可为ＤＭＣ算法的参数设计提供参考．
Dynamic Matrix Control algorithm is one of predictive control algorithms. Based on the unit step response, it fits to linear controlled objects with progressive stability, and its structure and algorithm are simple. But the parameters of DMC only effect performance of control system indirectly, can't be as quantitative measure of design because they haven't direct analytic relation with speediness, robustness, stability and anti jamming of control. This is very difficult for designer. Using satisfactory optimization concept, this paper proposes parameters' optimization method by simulation, it might be a reference of design in DMC algorithm.
出处
《信息与控制》
CSCD
北大核心
1999年第1期75-79,共5页
Information and Control
基金
国家自然科学基金
关键词
动态矩阵控制
预测控制
遗传算法
满意优化
仿真
dynamic matrix control, predictive control, genetic algorithm, satisfactory optimization, simulation
