Measurement-based Run-to-run Optimization of a Batch Reaction-distillation system

Recently, measurement-based optimization schemes have been developed to deal with uncertainty and process variations. One of the methods therein, labeled NCO tracking, relies on appropriate parameterization of the input profiles and adjusts the corresponding input parameters using measurements so as to satisfy the necessary conditions of optimality (NCO). The applicability of NCO-tracking schemes has been demonstrated on several academic-size examples. The goal of this paper is to show that it can be applied with similar ease to more complex real-life systems. A batch reaction-separation system for the production of propylene glycol is used for illustration. A simple parameterization of the optimal solution with four input parameters is derived and only the active terminal constraints are tracked in a run-to-run manner. With this strategy, near optimality is achieved within a few batches.