Enhanced production of methane enriched biogas through intensified co-digestion process and its effective utilization in a biodiesel/biohydrogen fueled engine with duel injection strategies: ML-RSM based an efficient optimization approach.

Due to stringent emission regulations, vehicle manufacturers must control diesel engine emissions. To address these regulations, this study introduces a novel approach for generating methane-enriched biogas through ultrasonic pretreatment and using it in dual fuel-mode diesel engines with a Hiptage biodiesel blend and biohydrogen. The study employs an RSM-based central composite design (CCD) approach to develop the L 54 orthogonal array for five factors and five levels of parameters. Additionally, seven different machine learning algorithms (DTR, RFR, ETR, GBR, LR, CBR, and XGBR) are used to develop a prognostic model based on the acquired experimental data. These models are used to estimate key performance indicators such as brake thermal efficiency (BTE), brake-specific fuel consumption (BSFC), carbon monoxide (CO), hydrocarbon (HC), and nitrogen oxides (NO x). Based on the evaluation of performance indicators, including R2 values (0.9021–0.97344), RMSE (0.00203–26.5539), MAE (0.00114–23.5944), and MAPE (0.0215–0.04438), the GBR model establishes the most accurate predictions. Furthermore, the RSM-based desirability approach is used to evaluate the trade-off between performance and emissions, thereby identifying the optimal engine operating condition. This condition includes an engine load of 83%, a compression ratio of 18:1, a fuel injection pressure (FIP) of 243 bar, a fuel injection timing (FIT) of 26° crank angle advance, and a biogas flow rate of 1.2 kg/h. Experimental investigation validates the RSM analysis findings, revealing an error margin of 7%. Overall, the introduction of methane-enriched biogas into a dual-fuel diesel engine improves combustion, resulting in increased engine performance and reduced emissions. [Display omitted] • Examine the effect of ultrasound on methane yield in the co-digestion of waste. • Study the behavior of a dual-fuel diesel engine with biodiesel, biogas, and biohydrogen. • Apply ML-RSM for prediction and optimization of engine characteristics. • The GB Regressor model demonstrated the most accurate predictions for key performance. • Experimental validation shows a maximum error of 7%. [ABSTRACT FROM AUTHOR]