Artificial neural networks approach on vibration and noise parameters assessment of flaxseed oil biodiesel fuelled CI engine

The current research dwells the experimental investigation together with artificial neural networks analysis on vibration and noise intensity of VCR diesel engine energized with flaxseed oil biodiesel blends. The tests were carried out at four different loads, namely 25, 50, 75, and 100% and three compression ratios (16.5, 17.5, and 18.5). The vibration levels were measured at different locations of the engine body which accounts for the development of bulk vibration. The RMS acceleration (vibration) and RMS noise were calculated. The flaxseed biodiesel mixes have shown the positive outcome concerning vibration and noise intensities, and FSOME20 (20% blend) was revealed inferior to leftover fuel samples. Besides, the intensity of vibration and noise was seen superior at higher compression ratios and loads. The identical disparity was observed irrespective of the fuel samples. At peak load, the RMS acceleration of FSOME20 was reduced by 12.2, 12.05, and 16.74%, while the RMS noise was brought down by 6.8, 9.6, and 7.23% at CR16.5, CR17.5, and CR18.5, respectively. Finally, the experimental results were compared with artificial neural networks (ANNs) predicted outcomes. The coefficient of correlation (R2) and root mean square error (RMSE) noticed were 0.965 and 0.12% intended for vibration, whereas 0.989 and 0.56% for noise correspondingly.