Studies on the performance and emission Characteristics of variable Compression ratio diesel engine blended With alternative fuels and a soft Computing approach using rsm and ann

Abstract

Bio-origin sources are the most important alternative sources of fuel like biodiesel, bioethanol, and biogas. Most researchers are keen on understanding the effect of different biofuels on engine performance. Due to the real-life scenario of the machinery, it is evident that the relations between different input and its output in any machinery gives non-linear and complex equations. To operate these types of machinery at good performance with better efficiency a machine learning algorithm needs to be developed. Some of the researchers concentrating more on self-driving automotive, which leads to developing a strong algorithm using various algorithms for CI engine, based on its input and output factors. Among the different methods, Artificial Neural Network (ANN) and Response Surface Methodology (RSM) develops a complex relationship between the input and output. Experiments are done for Neat Diesel (ND), Calophyllum Inophyllum Diesel Blend CIB25, CIB50, CIB75 and CIB100 at various load setting and with 14:1, 15:1, 16:1, 17:1 and 18:1 compression ratios for analyzing the performance and emission characteristics. Out of which CIB25 seems to have better performance than the other blends and it is comparable to ND at higher compression ratios. While comparing the performance parameters, CIB25 had 17% to 36% higher BTE value for CIB25 when compared to CIB50, CIB75 and CIB100 with. Also, 15% to 56% lower SFC seen in CIB25 than the CIB50, CIB75 and CIB100. Even for the mechanical efficiency, 3% to 8% greater for CIB25 when compared to the CIB50, CIB75 and CIB100. For emission parameters, CIB25 had 20%, 20-60% and 12-14% lower emissions of HC, CO and NOx respectively when compared to CIB50, CIB75 and CIB100. Also CIB25 had 3-8% greater CO2 emission when compared to CIB50, CIB75 and CIB100 newline