earticle

영어

Exhaust contaminant of gasoline vehicles is a crucial aspect to measure the vehicle performances and the air pollutions. According to the feature of vehicles, the emission of exhaust contamination of a vehicle is different as time goes by, which shows an increase tendency in most of the cases. Measuring the changes of a vehicle's exhaust contaminant emission is of great importance in the field of vehicle engineering. However, it is hard to determine and find out the regulations of the emission, needing a long time for regular determination and advanced relevant machines. In this article, we aim at providing two novel methods for the prediction of exhaust contaminant of gasoline vehicles, using grey model GM (1,1) and artificial neural networks (ANNs) models respectively. Results show that both the GM (1,1) model and ANN models are comparatively precise for the prediction. The GM (1,1) model can quickly obtain the predicted values of exhaust contaminant, but it is less precise than ANN models. However, ANN models need more time for the training process, compared to GM (1,1) mo