Characterization of emission-performance paradigm of a DI-CI engine using artificial intelligent based multi objective response surface methodology model fueled with diesel-biodiesel blends

Abstract

ABSTRACT The trade-off between performance and emissions must be in proper balance with the mineral fuels in executing a novel idea of renewable fuel design for existing and future combustors. Higher alcohols have superior activation energy and are an ideal choice as a diesel-biodiesel additive. Therefore, an investigative study is carried out to analyze the compatibility of biodiesel-diesel with n-Octanol in different proportions. An ignition improver, DEE is also added in different proportions in lower heating values and Cetane numbers standpoint. Results enumerated the benefit of prolonged ignition delay of higher alcohol and DEE have raised the peak pressures and better performance in terms of BSEC is recorded, and the emissions became leaner as NOx, unburnt Hydro Carbons and CO emissions are decreased. Furthermore, the engine responses were fed to an Artificial Intelligent model, Multi-Objective Response Surface Methodology to check the inherent capability of predicting the engine behavior and to optimize the input parameters. The model was tested on a statistical base and found robust as the regression coefficient R, R2 (Adj.), R2 (Pred.), MSRE and a special performance correlation metric like NSE which were found satisfactory. The model fetched a range of solutions based on the desirability criteria and suggested optimum blend composition as 20% biodiesel, 10% n-octanol, 2.526% DEE with 67.474% petrodiesel with the desirability of 0.985. The biodiesel along with the additives replaced diesel up to 34% by volume. Abbreviations: AI: Artificial Intelligence; FFD: Full Factorial Design; ANOVA: Analysis of Variance; HC.u: Hydrocarbons (Unburnt); BSEC: Brake Specific Energy Consumption; MORSM: Multi Objective Response Surface Methodology; CO: Carbon Monoxide; MSRE: Mean Squared Relative Error; DEE: Di Ethyl Ether; NSE: Nash–Sutcliffe Coefficient of Efficiency; DI: Direct Injection; NOx: Nitrogen Oxides; DICI: Direct injection Compression Ignition; TheilU2: TheilU2 uncertainity