Search

Abstract: This study investigates the use of ANN (artificial neural networks) modelling to predict BSFC (break specific fuel consumption), exhaust emissions that are CO (carbon monoxide) and HC (unburned hydrocarbon), and AFR (air–fuel ratio) of a spark ignition engine which operates with methanol and gasoline. To obtain training and testing data, a number of experiments were performed with a four-cylinder, four-stroke test engine operated at different engine speeds and torques. The experimental results reveal that the methanol improved the emission characteristics compared with the gasoline. For the ANN modelling, the standard back-propagation algorithm was found to be the optimum choice for training the model. In the building of the network structure, four different learning algorithms were used such as BFGS (Quasi-Newton back propagation), LM (Levenberg–Marquardt learning algorithm). It was found that the ANN model is able to predict the engine performance and exhaust emissions with a correlation coefficient of 0.998621, 0.977654, 0.998382 and 0.996075 for the BSFC, CO, HC and AFR for testing data, respectively. It was obvious that the developed ANN model is fairly powerful for predicting the brake specific fuel consumption and exhaust emissions of internal combustion engines. [Copyright &y& Elsevier]

• Taguchi method was applied to investigate factors affecting performance, emissions. • The factors selected are different biodiesel blends, EGR rates, loads and speeds. • The L16 orthogonal array was found for factors and levels. • Studies were done by Taguchi method, saving 70% in terms of time and cost. Biodiesel obtained from different vegetable or animal fats is a renewable fuel that can be used without any modification in diesel engines. And using biodiesel gives close performance data compared to diesel fuel. However, it increases the NOx (Nitrogen oxide) emissions, which is a major problem for diesel engines. Therefore, considering both the amount of biodiesel in blend fuel and NOx emissions reduction measures can be much more effective. If various factors are used together, it is necessary to determine the optimum conditions and utilization amounts considering performance and emissions. In this study, using different proportions of corn oil methyl ester (COME) blends (B0, B10, B20 and B50) and EGR (EGR0, EGR10, EGR15 and EGR20) on a direct injection (DI) diesel engine at variable loads (40%, 60%, 80% and 100%) and speed (1600 and 2400 rpm) conditions, the optimum factor levels in terms of performance and emission characteristics were determined by Taguchi method. As a result of the experimental design, L16 orthogonal array (OA) was found to be suitable and the studies were performed according to the combinations given in this array. The effect levels of factors were analyzed by ANOVA analysis. According to the results of the analysis, optimum levels of all factors were determined in terms of engine performance and emission characteristics. [ABSTRACT FROM AUTHOR]