Optimizing soybean biofuel blends for sustainable urban medium-duty commercial vehicles in India: an AI-driven approach.

This article presents the outcomes of a research study focused on optimizing the performance of soybean biofuel blends derived from soybean seeds specifically for urban medium-duty commercial vehicles. The study took into consideration elements such as production capacity, economics and assumed engine characteristics. For the purpose of predicting performance, combustion and emission characteristics, an artificial intelligence approach that has been trained using experimental data is used. At full load, the brake thermal efficiency (BTE) dropped as engine speed increased for biofuel and diesel fuel mixes, but brake-specific fuel consumption (BSFC) increased. The BSFC increased by 11.9% when diesel compared to using biofuel with diesel blends. The mixes cut both maximum cylinder pressure and NO x emissions. The biofuel-diesel fuel proved more successful, with maximum reduction of 9.8% and 22.2 at rpm, respectively. The biofuel and diesel blend significantly improved carbon dioxide ( CO 2 ) and smoke emissions. The biofuel blends offer significant advantages by decreeing exhaust pollutants and enhancing engine performance.

Experimental and empirical investigation of a CI engine fuelled with blends of diesel and roselle biodiesel.

The continuous rise in demand, combined with the depletion of the world's fossil fuel reserves, has forced the search for alternative fuels. The biodiesel produced from Roselle is one such indigenous biodiesel with tremendous promise, and its technical ability to operate with compression ignition engines is studied in this work. To characterize the fuel blends, researchers used experimental and empirical approaches while operating at engine loads of 25, 50, 75, and 100%, and with fuel injection timings of 19°, 21°, 23°, 25°, and 27° before top dead center. Results indicate that for 20% blend with the change of injection timing from 19° bTDC to 27° bTDC at full load, brake specific fuel consumption and exhaust gas temperature was increased by 15.84% and 4.60% respectively, while brake thermal efficiency decreases by 4.4%. Also, an 18.89% reduction in smoke, 5.26% increase in CO2, and 12.94% increase in NOx were observed. In addition, an empirical model for full range characterization was created. With an r-squared value of 0.9980 ± 0.0011, the artificial neural network model constructed to characterize all 10 variables was able to predict satisfactorily. Furthermore, substantial correlation among specific variables suggested that empirically reduced models were realistic.