Experimental Investigation on Attenuation of Emission with Optimized LPG Jet Induction in a Dual Fuel Diesel Engine and Prediction by ANN Model

Abstract

Environment pollution due to vehicles exhaust emission is still a severe crisis and an international concern has been raised for its control and diminution. Especially, man-kind is experiencing an epidemic of illnesses made worse by air pollution mainly because of increased number of automotive vehicles. The main problem faced by most of the vehicles today is the emission of NOx which can be controlled by different ways such as exhaust gas recirculation, using alternate fuels, turbo charging, and different mode of combustion. On the other hand, diesel engines are the most trustworthy power sources in the transportation. Due to inflexible emission norms and rapid depletion of petroleum resources, there has been a continuous exertion to use alternative fuels. Further researches are being carried out to reduce emission rates and these researches would be explored till reducing the exhaust emission upto zero level. In such a way, the present investigation explores with a series of experiments towards effective combustion of air, LPG and diesel mixture without encompassing major modification in the engine construction. Here, LPG jet is inducted through air inlet to accomplish homogeneous mixture then it is allowed into the combustion chamber of CI engine. The major parameters of the LPG jet injector are optimised through CFD technique by utilizing a commercial CFD code. Changes in the performance of the engine and emission levels with the influences of the jet parameters are observed and analysed experimentally. The optimized jet parameters obtained through CFD technique and the experimental results achieved from the dual fuelled CI engine show considerable improvement in the engine performance and significant reduction in CO2, CO, HC and NOx emission besides the normal engine performance and emission levels of diesel fuel induction. Eventually, an artificial neural network (ANN) model is developed for predicting the emission levels based on the jet parameters, applying load and % of LPG induction by utilizing the experimental results. It is also found that the predicted results provided good agreement with the experimental results.