Artificial intelligence based gene expression programming (GEP) model prediction of Diesel engine performances and exhaust emissions under Diesosenol fuel strategies

The study explores the affectability of oxygenated fuel on the performances and exhaust emissions of adulterated Diesel fueled engine. Diesel adulteration decreases brake thermal efficiency (B-th), brake specific energy consumption (BSEC), unburned hydrocarbon (UHC) and carbon monoxide (CO) emissions with significant reduction in NOX emissions. Ethanol blending improves the engine exhaust emissions without altering performance parameters of adulterated Diesel. In perspective of the experimental data, multi parametric artificial intelligence (AI) based gene expression programming (GEP) models have been developed for mapping the input (engine load, Kerosene share and Ethanol share) and output (B-th, BSEC, NOX, UHC and CO) relationship under Diesosenol platforms. The model predicted output has been validated with experimentally measured data and some statistical measures. The predicted model matched the experimental data with very lower mean square error (0.00002-0.00031). The statistical results such as correlation coefficient (0.99910-0.99995), absolute fraction of variance (0.99821-0.99989), Nash-Sutcliffe coefficient of efficiency (0.992-0.99974) and Kling-Gupta efficiency (0.98091-0.99736) obtained from the GEP model, along with mean absolute percentage error, mean squared relative error and prediction model uncertainty betokened itself as a real time robust machine identical tool under various Diesosenol stages. In addition, Pearson's chi-square test or goodness of fit measurement elevates the GEP model prediction quality to a higher level.