Prediction of Biodiesel Performance and Exhaust Emissions by Artificial Neural Network

In the present investigation, the use of artificial neural network (ANN) modelling of a diesel engine is explained to predict the brake thermal efficiency (BTE), specific fuel consumption (SFC), exhaust gas temperature (EGT), smoke density, oxygen(O₂), carbon dioxide(CO₂), hydrocarbon (HC), carbon monoxide (CO), oxides of nitrogen (NOx) of the engine. The effect of biodiesel on engine performance is considered. Test was conducted on a single cylinder water cooled diesel engine, at a constant speed of 1500rpm and the maximum power of 5.2 kW. Diesel, B25, B50, B75 and B100 derived from pongamia biodiesel were used to measure the engine parameters and emission. For the ANN modeling, the back-propagation learning algorithm was used and Levenberg-Marquardt function was applied for the variants of the algorithm. It was observed that the ANN model is able to predict the performance and emissions with correlation coefficient (R) of the exhaust gas temperature, smoke density, CO, HC, CO₂, O₂, NO_x, SFC and BTE yield a of 0.999428, 0.998327, 0.982433, 0.998861, 0.999459, 0.999957, 0.999333, 0.997444, and 0.999418 respectively. From the result ANN is seen as the best tool to predict the engine performance and exhaust emissions.