A MILP approach to the optimization of the operation policy of an emulsification process

This work addresses the model-based optimization of the operation policy of an emulsification process. In this paper the choice of the stirrer speed as function of the time, for reaching a certain predefined drop size distribution (DSD) in minimal time, is studied. It is argued that "standard" gradient based optimization techniques will fail to solve this optimization problem. An approach is suggested that approximates the original minimum time optimization problem as a Mixed Integer Linear Program (MILP). The MILP can be solved using well-proven, standard optimization codes and its solution is also a good solution of the original optimization problem. Applying it to the industrial process shows the feasibility of the approach and it illustrates the benefit of using model-based optimization for improving the operation policy of emulsification.