Modelling of synthesis of waste cooking oil methyl esters by artificial neural network and response surface methodology

Abstract

ABSTRACTThis present study was carried out to investigate the application of artificial neural network (ANN) and response surface methodology (RSM) as modelling tools for predicting the waste cooking oil methyl esters (WCOME) yield obtained from alkali-catalysed methanolysis of waste cooking oil (WCO). The impact of process parameters involved was studied by a central composite rotatable design. A comparison of the two developed models for the methanolysis process was carried out based on pertinent statistical parameters. The calculated values of coefficient of determination (R2) of 0.9950 and the average absolute deviation (AAD) of 0.4930 for the ANN model compared with R2 of 0.9843 and AAD of 0.9376 for the RSM model demonstrated that the ANN model was more accurate than the RSM model. The actual maximum WCOME yield of 94 wt% was obtained at a reaction temperature of 55°C, a catalyst amount of 1 w/v, a reaction time of 70 min and a methanol-to-oil ratio of 6:1.Abbreviations/Nomenclature CV: coefficient of variance; FFA: free-fatty acid; R: correlation coefficient; R2: coefficient of determination