Control vector optimization and genetic algorithms for mixed-integer dynamic optimization in the synthesis of rice drying processes

Rice drying synthesis is an essential operation that has to be done carefully and cost-effectively. Rice is harvested at high moisture content and hence must be dried within 24 h for safe storage. However, improper drying can cause fissuring in the rice grain, and thus greatly reduce its market value. Multi-pass drying systems are therefore used to gradually bring moisture content to desired level.The problem of rice synthesis, considered in this study, seeks the configuration of units and their corresponding operating conditions that maximize rice quality. This problem is formulated as a mixed-integer dynamic optimization problem. The integer part of the problem reflects process alternatives while the dynamic part originates from nonlinear differential-algebraic equations describing the drying behavior of a rice grain.Clearly such a formidable problem is not easy to solve. Hence, we propose an approach that makes use of two algorithms: a genetic algorithm to search for the best configuration of units and a control vector parameterization approach that optimizes the operating conditions for each configuration. We demonstrate the effectiveness of the approach on a case study.