Estimating Engine Performance and Emission Values Using ANFIS

Abstract

In this study, the effect of methanol mixtures in different proportions to emission and performance of the motor has been estimated using Adaptive Neuro Fuzzy Inference System (ANFIS) model. Training data and test data have been obtained from the results of experiments on a single cylinder, four-stroke engine with direct-injection under different spraying pressures. An ANFIS model has been developed using these experimental data. The estimated performance of the model has been obtained by comparison of the estimated results of the ANFIS model and the experimental results. It is seen that the obtained ANFIS model can make solid estimations about output emission and performance values of an internal combustion engine with different methanol mixtures.