Optimization Approaches & Strategies for Distillation Column Sequencing Separation for Olefins Production

This main objective of this project is to develop a mathematical model for determining the optimal design of distillation sequencing for olefin production. The mathematical model with optimization procedure for the integration of olefin flow from refinery to a petrochemical plant is based on a process flowsheet superstructure representation that embeds all possible alternatives for the distillation sequencing. The model formulation includes material balances with fixed split fractions and logical constraints for representing design specifications and structural specifications based on engineering knowledge and past design experience and heuristics. Additionally, big-M logical constraints relating the continuous variables of flowrates to the binary 0–1 variables of column existence are incorporated. In this work, the intermediate superstructure representation is adopted to represent the distillation sequencing for olefin production because it has been shown to provide good computational performance in obtaining the global optimal solution (Caballero and Grossmann, 1999).
The optimization model is investigated using different feedstock; ethane from Ethylene Polyethylene (M) Sdn. Bhd (EPEMSB) and naphtha from University of Manchester‟s Process Integration (2005). By using different feedstocks, the computational results yield the same optimal sequencing. Furthermore, The the optimal distillation sequencing with this model formulation is validated with the existing olefin plant. It is proved that the optimal distillation sequencing is consistent with the common heuristic in process plant synthesis. The optimization model is also investigated using integer cuts in order to check that they agree or conform to the heuristic for distillation sequencing. It is proved that the optimum solution has the least total mass flow rate.