Optimal Design of Petroleum Refinery Configuration Using a Model-Based Mixed-Integer Programming Approach with Practical Approximation

Abstract

We present a model-based optimization approach to determine the configuration of a petroleum refinery for grassroots (new) or existing site that considers a large number of commercial technologies particularly for heavy oil processing of crude oil residue from an atmospheric distillation unit. First, we develop a superstructure representation for the refinery configuration to encompass all possible topology alternatives comprising 96 technologies and their interconnectivities. The superstructure is postulated by decomposing it to incorporate representative heavy oil processing scheme alternatives that center on the technologies for atmospheric residual hydrodesulfurization (ARDS), vacuum residual hydrodesulfurization (VRDS), and residual fluid catalytic cracking (RFCC). We formulate a mixed-integer linear program (MILP) based on the superstructure by devising logic propositions on design and structural specifications that represent these processing options to aid convergence to an optimal refinery configuration. A numerical example is illustrated to implement the proposed technique in which an equivalent of more than two million refinery plot plans is evaluated. To assess the applicability and value of the approach, we validate the results against the literature as well as compare with existing real-world refinery configurations. A main contribution of this work is to demonstrate how a mixed-integer programming approach can be applied to a large-scale petroleum refinery design problem with suitable approximations informed by practical considerations to obtain results with reasonable computational load.

Simulation of Liquefied Petroleum Gas Recovery from Off Gases in a Fuel Oil Refinery

“…Kerosene, now rich with LPG, is taken out from the bottom and introduced into a stripper where LPG and kerosene are separated. LPG is taken from the top of the stripper and kerosene from its bottom [12,13].…”

“…Each plate has two conductors on both sides, called down comers. Fluid falls through the down comers by gravity from one plate to another below it [4,12]. A weir on the plate assurances that there is some fluid (holdup) on the plate and is structured such that the holdup is at a suitable height, to such an extent that the bubble caps are protected by fluid.…”

Sustainable Biorefineries: What was Learned from the Design, Analysis and Implementation

“…It is important to note that the most important developments reached last years in sustainability for oil refineries are related to computers and programming advances, making the consideration of a huge number of variables during the design stages possible. For that, optimization strategies to get the best configuration are usually used, including the Model-Based Mixed-Integer Programming Approach commonly used in process engineering [10].…”

An Efficient Modified Particle Swarm Optimization Algorithm for Solving Mixed-Integer Nonlinear Programming Problems

“…Benati et al [6] established an MIP model to solve the feature selection problem in clustering and a heuristic method was given. There are many other studies on MIPs [7][8][9][10].…”