Evolutionary Multi-Objective Optimization and Applications
12th September 2018, Wednesday, 04:30‐05:00 pm (25+5 mins), ChE Seminar hall
Abstract
Scheduling of crude oil refinery process involves decomposing of the refining process into three sub-parts and then recombining an individually optimized one. Crude oil operation is
the front side process which involves unloading of crude from vessels to storages, transferring to blending and desalting tanks and then charging to distillation units (CDUs).
Following the crude oil operations, products processing and products blending and distribution come, consecutively. In the present study, a desalting as a separate task on the
ordinary crude oil operation scheduling, a state task network (STN) model is formulated and its efficacy checked on four case studies1. The results show, the proposed model leads a step
closure to the reality and enables to know the amount of freshwater used and waste discharged. After evaluating the proposed model for crude oil operation, a graphically based
formulation called structure adapted genetic Algorithm (SAGA) is developed to optimize gasoline blending and distribution. The performance of the model is checked by solving an
industrial problem from literature2 which gives 15% lower operating cost than the corresponding MINLP model.
Reference
1. Yadav, S. & Shaik, M. A. Short-term scheduling of refinery crude oil operations. Ind Eng Chem Res. 51, 9287–9299 (2012).
2. Li, J., Karimi, I. A. & Srinivasan, R. Recipe determination and scheduling of gasoline blending operations. AIChE J. 56, 441–465 (2010).