Adaptive neuro-fuzzy inference system (ANFIS) and response surface methodology (RSM) prediction of biodiesel dynamic viscosity at 313 K

The purpose of this work was to investigate the applicability of adaptive neuro-fuzzy inference system (ANFIS) and response surface methodology (RSM) approaches for modeling the biodiesel blends property including dynamic viscosity at various volume fractions of biodiesel, kinematic viscosity and density of biodiesel blends at 313K. An experimental database of dynamic viscosity of biodiesel blends (biodiesel blend with petro-diesel fuel) was used for developing of models, where the input variables in the network were volume fractions of biodiesel, kinematic viscosity and density of biodiesel blends. The model results were compared with experimental ones for determining the accuracy of the ANFIS and RSM predictions. The developed models produced idealized results and were found to be useful for predicting the dynamic viscosity of biodiesel blends with a limited number of available data.