Optimization of Foraha Oil Biodiesel Production throgh Transesterification: A Comparative Study of Response Surface Methodology and Artificial Neural Networks

Biodiesel is an environmental free and substitute for diesel fuel it’s derived from different seeds and vegetable oil it contains long-chain mono alkyl esters. Crude Foraha oil (FO) was assessed in this study as a suitable fuel for the production of biodiesel. At 57.30 mg KOH/g, foraha oil has a high acid value. In order to lower the acid value to 0.85 mg KOH/g, degumming, esterification, and transesterification procedures were carried out. To enhance the yield of biodiesel, the RSM tool optimizes the following parameters: temperature, catalyst amount, reaction time, methanol to oil ratio, and stirrer speed. A quadratic response surface regression model was employed to predict the yield. With a predicted biodiesel yield of 97%, the ideal parameters were found to be a methanol-to-oil ratio of 1:3, catalyst concentration of 2wt% stirring speed of 500 rpm, and reaction duration of 110minutes.Response surface methodology (RSM) and artificial neural networks (ANN) are used to compare the expected biodiesel yield. RSM data was used to train the artificial neural network. Sensitivity analysis was used to assess each in-dependent variable's impact on the reaction. R2values of RSM and ANN is 97 and 98 respectively. Furthermore, the methyl ester properties of the produced biodiesel meet the fuel specifications outlined in the ASTM D6751 and EN 14214 standards.