针对换热器本身具有较强的纯滞性、约束性等特性,提出将模型预测控制应用于换热器控制中。首先将换热器近似处理为一阶惯性环节,推导出其传递函数作为预测模型;而后以输出误差与控制量加权之和的二次型性能指标为目标函数对其进行滚动优化,使全过程达到最优;最后,通过反馈校正构成系统的闭环优化,提高控制的鲁棒性。通过仿真实例证明:模型预测控制能够较好地处理换热器控制上的纯滞后等问题,具有较好的鲁棒性。
Focusing on the strong pure lag, constraint and other features of heat exchanger, the model predictive control was applied to heat exchanger control. Firstly, a heat exchanger was approximated as first order inertia link, and its deduced transfer function was used as prediction model; then the quadratic performance index of the weighted sum of output error and controlled variable was used as the objective function, which was rollingly optimized, and the whole process was optimal; finally, through the feedback correction, the closed-loop optimization system was formed to improve the robustness of the control. The simulation example proves that the model predictive control can better handle the problems of pure time-delay etc. for heat exchanger control, and has good robustness.