Accurate simulation of fuel properties influence in internal combustion engines performance is a very complex approach and combines many physical and chemical concepts such as combustion phenomena, chemical kinetics, fluid dynamics, turbulence and thermodynamics. The right modelling of that is still a challenge and currently available software packages for engines simulation usually consider standard or surrogate fuels.
The objective of this paper is the prediction of gasolines performance in internal combustion engines as an auxiliary tool in researches and developments of new fuels, reducing experimental timing and costs. It is proposed the use of kriging metamodels based on bench test results of a flexible fuel engine running with distinct blends of iso-octane, n-heptane, toluene and ethanol, to predict performance, energetic efficiency and pollutant emissions in function of fuel properties and operating conditions.