基于啤酒发酵温度多胞模型的鲁棒预测控制器设计

Design of a Robust Predictive Controller Based on Multi-Cell Model of Beer Fermentation Temperature

针对具有时滞性和输入输出约束的啤酒发酵系统,建立其多胞模型,并设计了鲁棒预测控制器。在啤酒发酵过程中,用一阶惯性纯滞后模型近似描述控制系统,并离散化。分析模型参数和采样周期的关系,得到发酵系统的多胞不确定时滞模型和约束关系。利用参数估计设计了鲁棒预测控制器。系统仿真表明,啤酒发酵温度能较准确地跟踪工艺曲线,满足工艺需求。

In this study, the multi-cell model was established for beer fermentation system with time-delay and input/output constraints and the robust predictive controller was further designed. Firstly, beer fermentation process was introduced and the first order plus time-delay model was used to describe the control system and then was discretized. The relationships between the model parameters and the sampling period were analyzed to obtain the multi-cell uncertain time-delay model and the constraint relations of the fermentation system. Then robust predic- tive controller was designed by using parameter estimation. The simulation test showed that beer fermentation temperature could track technical curve accurately and meet technical requirements.