Prediction of Performance and Emission Attributes of Biodiesel Blends in a Single Cylinder Engine Using Artificial Neural Networks

Dr Kiran Kumar Billa

Abstract: The present investigation delivers a comprehensive viewpoint on the current artificial intelligence (AI) meta-modelling in diesel engine system, particularly in the domains of multi-objective optimization. A Lavenberg-Marquardt feed-forward backpropagation learning algorithm was perceived to be good for ANN. The ANN model revealed its adeptness with a higher degree of accuracy between the predicted and experimental datasets and demonstrated a good agreement. The strength of the model is assessed using conventional metrics as well as some sophisticated metrics like MAPE, MSRE and NSE. The AI model showed satisfactory results with MAPE of 0.577-2.01% and acceptable RMSE threshold of 0.0093-0.0324. The special error metrics MSRE was 0.0000951-0.00013, NSE was 0.9967-0.9996, and Theil U2 0.019-0.055.

Keywords: Artificial intelligence, diesel engine optimization, Accuracy metrics, error analysis

How to Cite?: Dr Kiran Kumar Billa, "Prediction of Performance and Emission Attributes of Biodiesel Blends in a Single Cylinder Engine Using Artificial Neural Networks", Volume 13 Issue 3, March 2024, International Journal of Science and Research (IJSR), Pages: 824-828, https://www.ijsr.net/getabstract.php?paperid=SR24311142310, DOI: https://dx.doi.org/10.21275/SR24311142310