Reference-Based Model Predictive Control Scheme for Fed-Batch Bioprocess Optimization

概要

論文の詳細を見る
In the study, a semianalytic optimization algorithm was used to achieve the suboptimal performance under specific operating constraints of the fed-batch fermentation process. By applying Pontryagin’s minimum principle, the substrate feed rate was taken as the manipulated variable and characterized in a piecewise manner. On the basis of the repetitive optimization algorithm, a varied reference was developed by adjusting the switching times and other parameters. The reference-based model prediction strategy was successfully implemented to ensure the satisfactory output regulation, whereas the amount of product obtained was measured separately. The symbolic software MapleTM was used to implement the complete control structure, and the effectiveness of the proposed methodology was verified by performing closed-loop simulation tests.
社団法人化学工学会の論文