Computational Fluid Dynamics (CFD) Modelling for Process Intensification of Fluidized Dry Methane Reforming

Abstract

Dry Methane Reforming (DMR) is a process that produces synthesis gases (mixtures of hydrogen and carbon monoxide) from the reaction of carbon dioxide with methane. DMR is emerging as a replacement for the steam methane reforming process to mitigate the alarming concerns about global warming caused by greenhouse gases (GHG) emissions and this by replacing the reactant steam with CO2 while producing valuable chemicals (CO+H2) with the desired syngas H2/CO ratios suitable for downstream processes such as Fischer–Tropsch synthesis. Different configurations of packed bed reactors have originally been developed to host DMR. However, they all suffer shortcomings such as difficult temperature control, poor heat transfer, conversions/yields issues, and carbon deposition, to name a few. To remedy these inefficiencies, circulating fluidized beds (CFBs) can be deployed with the potential to meet the same success reported for other valuable petrochemical processes. However, challenges related to their design should be first alleviated. For this purpose, numerical models with sufficient accuracy and low computational cost are needed to study them. In the present research, hydrodynamics, heat transfer, and reforming reactions in a riser reactor of commercial scale are investigated using the Multi-Phase Particle-In-Cell approach (MP-PIC). The performance of the proposed fluidized bed is compared with available experimental and numerical literature reported data. Also, sensitivity analysis on various operating conditions are conducted. Expanding upon the aforementioned results, the combination of MP-PIC, Machine Learning through Artificial Neural Networks, and multi-objective genetic algorithms has been deployed to gain further insights into optimizing the process operating conditions. This technique facilitated the identification of optimal conditions by enhancing conversion and yield rates while suppressing the undesired coke formation.

Date of Award	18 Dec 2023
Original language	American English
Supervisor	Abdallah Berrouk (Supervisor)