Using of neural network to simulation of the biodiesel production process from waste oil.

Abstract

This study deals with artificial neural network (ANN) modeling of a biodiesel production using waste cooking oil to predict the biodiesel purity. developed ANN was a feed forward back propagation network with one input, one hidden and one output layers. The input parameters for the ANN were methanol/oil molar ratio (3:1–9:1), temperature (45–65â€‌â€‌â€‌â€‌â€‌â€‌â€‌â€‌â€‌‫‪‭ °C), and mixing intensity (200–600 rpm) and the output parameter was biodiesel purity. The biodiesel with best purity was produced at methanol/oil molar ratio, of 6:1 mixing intensity, of 600 rpm and reaction temperature of 65 °C. Different transfer functions and several rules were used to assess the percentage error between the desired and the predicted values. It was observed that the ANN model can predict the biodiesel yield quite well with correlation coefficient (R) 0.99966. We can conclude that ANNs can be used to predict the biodiesel yield from used waste cooking oil.

Keywords

• biodiesel
• waste cooking oils
• artificial neural network
• yield purity