Research on Critical Oxygen Concentration in Oxy-MILD Combustion of Syngas Fuel Using Artificial Neural Networks Method

This study investigates the critical oxygen concentration for oxy-fuel moderate or intense low oxygen dilution (Oxy-MILD) combustion of syngas. Syngas, composed of H<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>, CO, and CH<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</inf>, presents unique challenges for achieving stable MILD combustion due to its distinct chemical activity. This research employs two predictive models, multiple linear regression (MLR) and artificial neural networks (ANN), to analyze the relationship between oxygen concentration and the molar fractions of syngas components. The results show that while the MLR provides reasonable accuracy for data with small deviations, the ANN model outperforms it in terms of prediction accuracy, particularly for more complex, non-linear scenarios. This study highlights the potential of ANN models in predicting critical combustion parameters and offers valuable insights for optimizing combustion processes in industrial settings, demonstrating the applicability of advanced machine learning techniques in combustion research.