Predicting Minimum Energy Conditions for a Distillation Column by Design of Experiments and Process Simulation

Abstract

Distillation is the most commonly used and the most versatile separation method for liquid components in boiling mixtures. Unfortunately, this unit operation is often one of the biggest energy consumers in industrial processes. The energy consumption of the distillation column is dependent on several operation variables; optimization of these variables means to minimize the energy demand while maintaining good product quality. In the classical optimization approach, only one variable is varied at a time, and its effect on the system is recorded. This so-called “univariate” approach often requires a considerable experimental or computational effort, and it neglects relationships between the variables. In the multivariate optimization approach, the variables are varied in a more efficient way, and their possible interaction is taken into account. In the present work, a multivariate approach is used to define the optimal operation variables for minimum energy consumption of a distillation column. In the studied case, the distillation column is the solvent recovery in an organosolv process. The system is set up in the process simulation software Aspen Plus. In the given set-up, three independent column variables are identified for optimization: number of column stages, feed stage location, and solvent concentration at the column top. The multivariate optimization is performed in software Design Expert, and combined with process simulation results. The result of the multivariate approach is an empirical regression model for calculating the energy demand of the column and the optimum operation conditions.