Performance and emission analysis of a dual-fuel engine using biogas and algal biodiesel: Machine learning prediction and response surface optimization

This study investigates the dual-fuel operation of a compression ignition engine fueled with biogas as the primary fuel and algal biodiesel as the pilot fuel. Machine learning (ML) models and response surface methodology (RSM) were utilized to evaluate and optimize engine performance and emissions. Engine parameters, including load and compression ratio (CR), were varied to assess their impact on brake thermal efficiency (BTE), combustion characteristics, and emissions of NOx, CO, and unburnt hydrocarbons. Optimization revealed that a load of 82.71 % and a CR of 18.5 achieved a BTE of 16.98 %, a Pmax of 48 bar, and reduced emissions of CO 2 (3.16 %), HC (112 ppm), NOx (306 ppm), and CO (172 ppm). ML modeling provided accurate predictions, with XGBoost outperforming RF and LR in emission models. XGBoost achieved the lowest MSE (0.16 for BTE, 0.02 for CO 2 ), highest R2 (0.9944 for BTE), and smallest MAPE (1.63 % for BTE). The study highlights the feasibility of biogas and algal biodiesel as sustainable energy sources for dual-fuel engines and demonstrates the effectiveness of advanced computational techniques like RSM and ML in optimizing renewable energy applications, paving the way for efficient and eco-friendly engine technologies.