Modelling to predict performance and emission parameters of hydrogen, additive and nano-particles blended fuel in a dual-fuel diesel engine through artificial neural network

Abstract

The depleted state of fossil fuels has led to the need to find an alternate solution for the utilization of transportation vehicles. The use of hydrogen, fuel additives, and nanoparticles has been part of the research in the recent past. Additionally, the use of artificial neural network was applied to anticipate the results for the optimization of the work and output, which has also been a study of interest recently. In the current work, different blends of diesel with hydrogen, TGME as a fuel additive, and Al2O3 nanoparticles were utilized to check for the best performance and the least emissions. The best performance was seen with the inclusion of TGME in diesel along with hydrogen, while the use of nanoparticles increased CO formation. An improvement in the BTE was also found with the blends of TGME and nanoparticles in diesel along with hydrogen. The usage of TGME along with H2 performed the best in terms of BTE (15.8%) and while the least emission of NOx, HC, and CO was found with the combination of diesel and H2. The artificial neural network was applied afterwards to predict the results using the obtained data from the experimental results. It was found that the values of regression coefficients for BTE and ITE were close to 1 (0.99 for BTE and 0.98 for ITE). In the above cases, the value of the mean square error was also observed to be least. Furthermore, the results for the regression coefficient of emissions were close to 1 for NOx (0.94), HC (0.89), and CO (0.93) with minimum possible mean square errors in all the cases. The results showed that they were in line with the predicted data as well as the available literature for the anticipation of the obtained results.