Comparative assessment of response surface methodology and artificial neural networks in forecasting biodiesel yield from waste cooking sunflower oil. (2021)

• Record Type: Journal Article
• Authors: Devaraj Naik, B.; Udayakumar, M.
• Is Part Of: Materials today. Volume 46:Part 19(2021)
• Page Start: 9908
• Page End: 9913
• Publication Date: 2021
• Subjects: Waste sunflower cooking oil -- Response surface methodology -- Artificial neural network -- Biodiesel -- Design of experiments; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2021.02.291
• Languages: English
• ISSNs: 2214-7853

Abstract

Highlights: Bentonite supported sodium methoxide catalyst was used for biodiesel production. Maximum biodiesel yield of 94.3% was observed at optimum condition. The coefficient of determination R 2 value of ANN and RSM were found to be 0.997 and 0.992. ANN model predicted the biodiesel yield with least mean square error than RSM. Abstract: In this work, comparative assessment of predicted biodiesel yield using response surface methodology (RSM) and artificial neural network (ANN) is implemented. The artificial neural network was trained using RSM based experimental data. The importance of each independent variable on the response was evaluated using sensitivity analysis. This study shows that ANN and RSM models closely predicted the yield with an R 2 value of 99.7 and 99.2 respectively. However, the ANN model precisely predicted the biodiesel yield with the least mean square error value of 0.00033, which is significantly lower than the yield predicted using RSM. The ANN simulation results show good agreement with experimental data. It can be inferred that ANN is a better tool compared to RSM and can be used for accurately predicting biodiesel yield that thereby effectively reducing the time-consuming and expensive experimental test.