A predictive controller with self-renewal model observer for continuous sterilization of biochemical process

Continuous sterilization is conducted on medium before fermentation processes, and the aim of sterilization is effective destruction of microorganisms while less that of the fermentable substrate. Problems exist in control process of continuous sterilization for the system variety and strong disturbances, bringing out bad control accuracy and performance of sterilization. In this paper, a predictive controller with self-renewal model observer is proposed to obtain high control accuracy. An approximation model is obtained based on industrial data, and model predictive control strategy is implemented to overcome the bottleneck of complex precise modeling, at the same time guarantee the control accuracy. Besides, a model observer is designed to deal with strong disturbances in industrial conditions. The parameters of the model observer is updated as the sterilization process goes on, so as to follow the system variety well. Simulations are carried out, and results show the effectiveness of the proposed strategy in theory. The strategy is applied in a fermentation workshop in Yantai, Shandong province, and operating results show that it proves good control accuracy, high tolerance for variety and disturbance in practice.