基于数据驱动的锅炉主蒸汽温度预测控制

Data Driven-Based Predictive Control for Main Steam Temperature

基于数据驱动控制(data driven control,DDC)理论,采用无模型自适应控制,并结合预测控制在线滚动优化、反馈校正的优点,设计了无模型自适应预测控制器(model free adaptive predictive control,MFAPC)。将上述控制器应用到锅炉主蒸汽温度串级控制系统,仿真结果表明,上述控制策略对大惯性、大迟延的主蒸汽温度过程控制效果良好,而且系统响应速度快、抗干扰性能强,对系统参数时变具有较好的鲁棒性。

On the basis of data driven control theory, a model-free adaptive predictive controller is proposed by u- sing model-free adaptive control and predictive control, which combines both advantages of on-line receding optimi- zation, feedback correction and model free adaptive technique. Through the simulation on main steam temperature cascade control system, the results illustrate that the control strategy has good control effect for the large inertia and long time delay property of the main steam temperature process. And the control system can work well with fast responsive speed, strong disturbance-rejection ability, and robustness in controlling time-varying systems.