Superstructure Optimization of Petroleum Refinary Design : Processing Alternatives for Vacuum GaS Oil Produced from Distillation Unit

The objective of this research project is to develop an optimization-based
mathematical model in the form of a mixed-integer linear program (MILP) for
determining the optimal configuration of a petroleum refinery. The optimization
model serves as a decision support system taking into consideration the relevant
engineering knowledge and the past design experience. The scope of work of this
project focuses on the strategies and approaches for the optimal design of a petroleum
refinery configuration for vacuum gas oil processing using mathematical
programming in order to select the most economical and cost-efficient process route.
The approach involves the development of a superstructure representation in which all
feasible processing alternatives and routes for vacuum gas oil are considered.
Subsequently, the associated constraints and a suitable cost minimization objective
function are formulated to arrive at the optimal solution in terms of the optimal
process units to be selected and the corresponding optimal flowrates of the material
streams. Linear yield-based material balances are considered in the model constraints
by obtaining representative yield values from the literature. Logical constraints in the
form of logic cuts for stipulating the selection or no-selection of the process units
(tasks) and material streams (states) are included in the optimization model
formulation to further define the feasible region of solutions based on the relevant
design specifications and the interconnectivity relationships among the nodes of states
and tasks in the superstructure. Computational experiments are carried out by
implementing the model on GAMS. A numerical example with a hypothetical set of
product demands is illustrated, which is solved to optimality using GAMS, resulting
in a practically-viable optimal refinery configuration that agrees well with real-world
scenario