Laminar Flame Speed modeling for Low Carbon Fuels using methods of Machine Learning

Application and comparison of multiple machine learning models for the prediction of the laminar burning velocity for CH3OH/H2/air mixtures

“…Eckart et al found that the best performance for predicting the LBVs of methane/hydrogen/air mixtures can be achieved with the ANN model [13]. Shahpouri [14] found that for the prediction of the LBVs of NH 3 /H 2 /CH 3 OH mixtures, ANN shows better performance than SVM for single fuels, while SVM has better performance for fuel blends. However, there are limited studies discussing the effect of various normalization methods on the predicted performance.…”

Data-Driven Prediction of Laminar Burning Velocity for Ternary Ammonia/Hydrogen/Methane/Air Premixed Flames

“…Nevertheless, the model was not validated for high-pressure and high-temperature conditions. More recently, Shahpouri et al [35] investigated the MLbased prediction of laminar flame speed for low-carbon fuels like NH 3 , H 2 , CH 3 OH, and their combinations. The study employed 1D simulations to generate a substantial LBV database and subsequently trained the models using ANN and Support Vector Machine (SVM) algorithms.…”