Strategic optimization of engine performance and emissions with bio-hydrogenated diesel and biodiesel: A RVEA-GRNNs framework
Niti Klinkaew, Attasit Wiangkham, Atthaphon Ariyarit, Prasert Aengchuan, Anupap Pumpuang, Sarunporn Sripratum, Atthaphon Maneedaeng, Jiraphon Srisertpol, Ekarong Sukjit
Abstract
This study investigates the effects of blending bio-hydrogenated diesel and biodiesel on engine performance and emissions across various engine speeds (2000-3000 rpm) and loads (25-90%), addressing the growing demand for sustainable transportation fuels. Using a single-cylinder diesel engine from POLAWAT ENGINE Company Limited, we evaluated different bio-hydrogenated diesel and biodiesel blends, optimizing their composition through the Reference Vector Guided Evolutionary Algorithm with a surrogate objective function via Generalized Regression Neural Networks. Results indicate that engine performance is closely linked to combustion temperature, which significantly affects fuel consumption and thermal efficiency. Bio-hydrogenated diesel demonstrated superior performance compared to conventional diesel, with lower fuel consumption and higher thermal efficiency, attributed to its higher cetane index (78 vs. 56) and heating value (47.02 MJ/kg vs. 43.48 MJ/kg). However, increasing biodiesel content in blends tended to increase fuel consumption and decrease efficiency due to biodiesel's lower heating value (39.62 MJ/kg) and higher viscosity (5.16 cSt vs. 2.58 cSt for bio-hydrogenated diesel). Emission analysis revealed that bio-hydrogenated diesel generally produced lower hydrocarbon, carbon monoxide, and smoke emissions than conventional diesel, though nitrogen oxide emissions were slightly higher. Biodiesel blends often showed increased emissions, particularly at higher blend ratios, due to poor fuel atomization. The Generalized Regression Neural Networks model demonstrated high accuracy in predicting engine performance and emissions, with coefficient of determination (R 2 ) values exceeding 0.98 and mean absolute percentage errors below 5%. Optimization results indicated that bio-hydrogenated diesel ratios centered around 90% provided the best balance of performance and emissions. This research bridges critical gaps in combined bio-hydrogenated diesel and biodiesel usage and artificial intelligence-driven biofuel optimization, providing valuable insights for practical implementation in Thailand's agricultural sector. • RVEA optimization via GRNNs improved engine analysis. • BHD outperforms diesel while blending biodiesel reduces efficiency. • Higher engine loads increase in-cylinder pressure and energy release. • BHD reduces emissions while biodiesel blends amplify. • GRNNs accurately predict engine behavior, validating model effectiveness.