Journal of Global Ecology and Environment

This study presents a machine learning (ML)-based framework to predict the combustion characteristics of ammonia-blended fuels for automotive applications. Specifically, the study examines the combustion kinetics of ammonia-blend automotive fuels using Artificial Neural Networks (ANN) and Support Vector Machines (SVM) for optimal parameters performance, including laminar burning velocity (LBV), ignition delay timing (IDT), and equivalence ratio under lean and near-stoichiometric conditions. A total of 300 data points, sourced from both experimental and simulated datasets published between 2015 and 2025, were preprocessed to ensure consistency and accuracy. The models were validated using k-fold cross-validation and evaluated using performance metrics such as R², RMSE, and MSE. A comparative analysis of various ammonia-blend fuels was conducted, including blends with hydrogen, di-methyl ether, methanol, methane, ethanol, and di-methoxy methane. Ammonia-hydrogen, ammonia-di methyl ether, ammonia-methanol, and ammonia-methane blends demonstrated favorable combustion characteristics with high LBV (R² = 0.94–0.96), and short ignition delay times, based on a proposed correlation ( y = 0.93x + 8.57 ) from test dataset. alternatively, ammonia-di methoxy methane and ammonia-ethanol blends exhibited weaker combustion performance, with lower LBV of R² = 0.84–0.88, and poor ignition characteristics. The outcome shows that the ANN model consistently outperformed the SVM model in the predictive accuracy of combustion parameters, particularly over LBV and IDT, as reflected by higher R² and lower RMSE and MSE scores, confirming the suitability of machine learning models, particularly ANN for capturing complex, nonlinear combustion behaviors across varied compositional blends. These insights support the development of optimized fuel blends for next-generation internal combustion engines. The study proposes a correlation for practical applications in engine design, fuel calibration, and emission reduction, contributing to the global transition toward cleaner automotive energy systems. In addition, the outcome can serve as a policy formulation and industrial innovation guide for the utilization of ammonia-based alternative fuels.