Optimization of multiple-fraction batch distillation by nonlinear programming



Batch distillation processes are very attractive for the recent development of the chemical industry: multipurpose, flexible plants and fine chemistry. For many separations of high-added value products, even a modest change in operating conditions has a significant economic impact—there is an important challenge for optimizing such processes.

Short-cut and dynamic models are the two classical approaches to the simulation of batch distillation columns. For problems without holdup, an intermediate procedure based on a decoupling method is validated.

For a multifraction separation problem with fixed final time, the reflux policies for each period and the period switching times constitute the set of decision variables. For predefined reflux policies, we apply an engineering approach to the solution of a such constrained variational problem, based on its transformation into a nonlinear programming problem. In this computer-implementable algorithm, the gain in distillate for the optimal linear or exponential reflux policies is significant (about 10%) compared with the optimal constant reflux policy.