Abstract: Due to easy modeling and good robust properties of non-parametric model predictive control based on the step response of a process,it is extensively used in process control. However, the superfluous parameters of non-parametric model predictive control lead to heavy computational burden in the computation of process predictive output and online receding horizon optimization. To overcome the above demerits of non-parametric model predictive control, on the basis of step response of a process, the constrained model predictive control algorithm with redundant parameters is presented in this paper by employing a model with a few redundant parameters to approach the step response of the process. Because of far fewer parameters of the redundant-parameter model than the non-parametric model, the computational burden of multi-step prediction and online receding horizon optimization is relieved greatly. To effectively overcome the adverse influence of modeling errors, a PID feedback correction scheme is introduced. The effectiveness of the presented approaches is verified through computer simulations employing step response data collected from a heat exchanger.

摘要： 在系统阶跃响应的基础上，根据最小绝对误差准则，用一个具有少量冗余参数的模型逼近实际系统的阶跃响应，提出了一种有约束冗余参数模型预测控制算法.由于冗余参数模型的参数比非参数模型少得多，多步预测和滚动优化的计算量大为减小.为了校正模型逼近带来的误差，引入了PID反馈校正环节，从而有效地克服了模型误差的不利影响.最后利用热交换过程的阶跃响应数据进行了计算机仿真研究，使算法的有效性得到验证.