Skip to main content

Aerospace: Accelerating the development of greener, safer fuels with AI

June 3, 2024

iew from an airplane window showing the wing, with a clear blue sky and the sun shining brightly above a sea of fluffy white clouds.
Portrait of Hoi Dick Ng, a professor at Concordia University's Gina Cody School of Engineering and Computer Science, wearing glasses and a dark suit. Hoi Dick Ng

In the fight against climate change, developing greener fuels that are hydrogen and ammonia based is crucial as they are zero carbon fuel. Artificial intelligence (AI) is aiding in this effort by helping to design safer fuel mixtures for rockets and jet engines, which could significantly reduce the carbon footprint of the aerospace industry.

Researchers at Concordia’s Gina Cody School of Engineering and Computer Science are at the forefront of this initiative, leveraging AI to enhance fuel safety and efficiency.

In a recent study published in Energies, Georgios Bakalis and Hoi Dick Ng from the Department of Mechanical, Industrial and Aerospace Engineering have employed artificial neural networks (ANNs) to predict the behavior of explosive fuel mixtures. Their research particularly focuses on the 'cell size' of detonations, which is a critical factor in understanding the sensitivity and behavior of a detonation wave. Since smaller cell sizes typically indicate higher sensitivity and greater risks, accurately predicting these sizes is essential for both safety and design.

ANNs, which are forms of AI modeled after the human brain, are capable of learning from data to make predictions. Using a previously developed ANN model, Bakalis and Ng predicted the cell size of various fuel mixtures, including hydrogen, biogas, and ammonia. These fuels are targeted due to their potential as sustainable alternatives.

The model, which was initially trained with data from previous experiments and chemical calculations, has shown the ability to predict cell sizes for new mixtures not included in the original training set, demonstrating both versatility and robustness. The results from the ANN model closely align with actual experimental data for a diverse range of fuel mixtures. Furthermore, the inclusion of new experimental data into the model has improved its accuracy, enhancing its utility for future research and development.

Learn more about the Department of Mechanical, Industrial and Aerospace Engineering at the Gina Cody School

Back to top

© Concordia University