Experimental and machine learning‐based investigation on the engine parameters of the reactivity controlled compression ignition engine fueled with microalgae biodiesel blends, n‐butanol and hydrogen gas

Abstract The study aims to investigate the combustion, performance and emission behavior of an reactivity controlled compression ignition (RCCI) engine using Chlorella vulgaris microalgae biodiesel (CMBD) as high reactivity fuel, n‐butanol and hydrogen gas as a low reactivity and supplementary fuels. The CMBD fuel is prepared with the aid of transesterification process and characterized by Fourier Transform Infrared Spectroscopy analysis. The engine tests were performed at different loads and using various fuel samples on the RCCI engine then results were analyzed by using machine learning (ML)‐based (random forest [RF] and support vector machine) algorithms for prediction and optimization of engine parameters. The CMBD + But + 8 LPM HG fuel achieved the highest Brake thermal efficiency of 33.2% which is 2.5% and 11.4% higher than diesel and CMBD fuel, while Brake specific energy consumption (BSEC) decreased by 26.5% and 37.5% accordingly. The addition of hydrogen gas and oxygenated fuel enhances the premixed combustion, resulting in faster heat release, higher peak cylinder pressure and improved overall combustion efficiency. The reduced CO (0.30% Vol), HC (52 ppm) along with smoke (46.6 HSU) compared to diesel due to the improved combustion, whereas NOx emissions increases to 1280 ppm owing to higher in‐cylinder temperature for the CMBD + But + 8 LPM HG fuel. Among the ML models, SVM model predicted Brake thermal efficiency with higher accuracy ( R 2 of 0.9975) while RF model shows superior prediction for BSEC and NOx ( R 2 = 0.9879 and 0.9896). Overall, the study demonstrates that hydrogen‐assisted oxygenated (n‐butanol) fuel blends significantly enhances the engine performance and provides an effective pathway for ML‐based optimization for RCCI engines.