Numerical investigation and prediction of effects of applying Anatase TiO2 nano-particle on performance and emissions of CI engine by using WNN-GA

Abstract

This study analyzed the effects of Anatase titanium dioxide (TiO2) nano-catalysts addition on performance and emissions of a compression ignition (CI) engine. By making a one-dimensional model of the engine in GT-power software and conducting the simulation process by using the model, this has been done. The performance and emissions of the engine in partial loads were investigated in three selected speeds for engine with diesel and again under the same conditions by adding TiO2 nanoparticles of Anatase to fuel. Performance parameters which would be examined for two types fuel, diesel and synthetic, are BSFC, power, torque, NOx, HC and Co, respectively. Then, in regarding to experimental data, a wavelet neural network (WNN) model was developed. We used four different kinds of mother wavelet functions and the best one was chosen in neurons of the network layer to predict the engine performance. Using genetic algorithm (GA), optimization coefficients and WNN parameters were determined. The numerical and prediction results were in a good match with experimental result. Therefore, results show that modeling by GT-power and WNN have very high ability to predict the performance and emissions of the engine, without needing any expensive and time-consuming engine tests.