Prediction of some thermo-physical properties of biodiesel using ANFIS and ANN cum sensitivity analysis

Abstract

The prediction of kinematic viscosity (KV) and cetane number (CN) of African pear seed oil methyl ester (APOME) using adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN) models were carried out. The solvent extraction was used to extract oil from African pear seed and characterized with Association of Analytical Chemist (AOAC) methods. The catalysts were synthesized by thermal and chemical activation methods. The activated clays were characterized and used to convert African pear seed oil (APO) to biodiesel. The physiochemical and thermo-physical properties of the biodiesel were determined using American Society for Testing Materials ASTM D6571 and some of the thermo-physical properties (KV and CN) were modeled using adaptive neuro-fuzzy inference system. The APO yielded 52% oil which is equivalent to the yields obtained from other non-edible seeds. The APSOME produced from the APO had qualities that were within the standard limits for biodiesel. The thermo-physical properties of the biodiesel (APSOME) were predicted accurately by ANFIS and ANN with good performance parameters. The double-input-double-output (DIDO) ANN architecture model best predicted the properties with lower root mean square error of 0.03423 and higher coefficient of determination of 0.9999. The biodiesel fraction has 14% more relative contribution than temperature on the thermo-physical properties. Therefore, ANFIS and ANN can be used to predict KV and CN.