Research On Optimization And Control Methods For Batch Process Considering The Loop Control

Batch process is a kind of earlier used mode of production operation in the industrial field. Owing to less space equipment and flexible operation, it is widely used by people. With the further improvement of batch process in worldwide, a more pressing need in production automation, optimization of production cost and process projects are proposed by manufacturing enterprises of batch process. Therefore, researching on optimization and control for batch process has important theoretical and practical significance.Due to key loop settings of optimization for batch process need to be achieved by controling of base level, therefore, researching on optimization and control for batch process must take into account coordination of the process-level and loop-level. Hence, an optimization control method for batch processes considering the loop control is proposed in this paper. Studying on optimization control for alcohol fed batch fermentation process, a typical batch process, is carried out in this paper. Firstly, the final quality prediction model of alcohol batch fermentation and controling prediction model of alcohol feeding are built based on Partial Least Squares (PLS) respectively, and then particle swarm optimization (PSO) algorithm is applied to pre-set of feeding optimization for alcohol batch fermentation process. Then, taking advantage of the similar idea between repetitive batch runs and iterations calculation during sequential quadratic programming (SQP) optimization, a batch-to-batch optimization control strategy coupled with MKPLS and SQP is proposed, after optimizing adjustments gradually, control trajectories based on the actual process may be greater optimal. And, simulation results have demonstrated that the proposed method can effectively eliminate the quality of deviation caused by error of the model and unknown disturbances. Due to model-plant mismatches, the optimization control methods base on confidence intervals of error and the uncertainty of error information are proposed for this problem, the effectiveness of proposed methods is validated with smiulation of feeding optimization control for alcohol batch fermentation.