Zoom Link: https://kaust.zoom.us/j/96110627855

Abstract

The combustion of ammonia/hydrogen is currently gaining importance in the power generation sector as an alternative for hydrocarbon fuels and improved fundamental insights will facilitate its application. However, accurately modeling these flames remains a significant challenge due to complex phenomena such as differential diffusion and local extinction. While high-fidelity simulations may resolve these effects, they are often too computationally expensive for practical engineering use. This PhD research addresses these challenges by developing a novel reduced-order modeling approach that integrates principal component analysis (PCA) with deep neural networks (DNNs). The resulting PC-DNN model significantly enhances computational efficiency without compromising predictive accuracy. It fully incorporates differential diffusion and subgrid-scale closures within the framework of large eddy simulations (LES). This newly developed model has been successfully applied to a range of flame configurations and validated against recent experimental data, offering a scalable and accurate tool for the design of next-generation low-emission energy systems.

Biography

Suliman Abdelwahid is a PhD candidate in Mechanical Engineering at the Computational Reacting Flow Laboratory at KAUST, under the supervision of Prof. Hong Im. He obtained his bachelor's degree from KFUPM and then joined the Internal Combustion Engine Group of Politecnico di Milano, where he earned his master's degree in 2020. His work focuses on computational fluid dynamics (CFD) and modeling of turbulent reacting flows, particularly ammonia/hydrogen combustion systems.