Machine learning driven chemiluminescence-based modelling of combustion parameters in premixed swirling NH3/H2 flames

Agwu, Nwode, Davies, Jordan, Sato, Daisuke, Mashruk, Syed and Valera Medina, Agustin ORCID: https://orcid.org/0000-0003-1580-7133 2025. Machine learning driven chemiluminescence-based modelling of combustion parameters in premixed swirling NH3/H2 flames. International Journal of Hydrogen Energy 145 , 717 - 731. 10.1016/j.ijhydene.2025.06.062

PDF - Published Version Available under License Creative Commons Attribution. Download (8MB)

Abstract

This work develops predictive models for estimating equivalence ratio (ϕ), ammonia fraction (xNH3) and noxious emissions (NOx) from the flames of turbulent premixed NH3/H2 fuel blend stabilised using a tangential swirl burner. Bayesian Regularisation Artificial Neural Network (BR-ANN) is utilised to estimate both ϕ and xNH3 with excited ratios NH*/OH*, violet/OH* and NH2*/NO2* as inputs. NOx was predicted with NO*, OH*, NH*, NH2* and NO2* as input variables. The coefficient of determination(R2) was 0.98,0.95, 0.99,0.99 and 0.97 for the ϕ, xNH3, NO, NO2 and N2O models, respectively. The models show better performance when compared to the conventional ratio-based method of inferring crucial combustion features. The developed models operate within the ranges of ammonia-hydrogen blend (0.55≤ xNH3 ≤ 0.90), Reynolds numbers (4000 ≤ Re ≤ 7000), equivalence ratios (0.60 ≤ϕ ≤ 1.40), room temperature and atmospheric pressure. The models have been explicitly presented in mathematical equations enabling easy deployment in a software. These models will serve as a crucial step towards the development of non-invasive sensors that will help designers easily determine location of interest, predict reaction zones formation according to ϕ and xNH3 and advance approaches that abate NOx emissions.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Schools > Engineering
Subjects:	T Technology > TJ Mechanical engineering and machinery
Publisher:	Elsevier
ISSN:	0360-3199
Funders:	Cardiff University
Date of First Compliant Deposit:	6 June 2025
Date of Acceptance:	3 June 2025
Last Modified:	13 Jun 2025 15:15
URI:	https://orca.cardiff.ac.uk/id/eprint/178828

Actions (repository staff only)

Edit Item