Multi-Criteria Optimization and Prediction of Diesel-Fusel Oil Dual-Fuel Engine Performance Using ANN, RSM, and MCDM Approaches

With growing interest in sustainable combustion technologies, this study investigates the performance and emission characteristics of a diesel–fusel oil dual-fuel engine using experimental and computational approaches. Six fuel blends (D100, F5, F10, F20, F30, F37) were tested across engine speeds from 1000 to 3250 RPM. Key parameters such as BSFC, engine power, torque, emissions (CO₂, CO, HC, NOx, smoke opacity, PM), and engine noise were evaluated. The optimal blend D70F30 at 2750 RPM offered the best balance between efficiency and emissions. ANN models achieved high prediction accuracy (R² > 0.93 for all metrics), while RSM revealed that increasing fusel content raised BSFC and HC but reduced NOx. Fuzzy logic identified D73.57F26.43 at 1000 RPM as optimal for minimum NOx (84.91 ppm) and BSFC (331.81 g/kWh), while D100 achieved peak power and torque. TOPSIS and VIKOR methods ranked D80F20 at 2000 RPM and D70F30 at 3250 RPM as top blends. This study confirms fusel oil’s potential to reduce emissions without engine modification, offering insights for cleaner, efficient dual-fuel applications.