This study investigated biodiesel synthesis prediction alongside transesterification experiments using Karanja oil. It was carried out in two phases: the first involved experimental trials, and the second focused on predicting biodiesel yield using an artificial neural network (ANN). A 4-10-1 topology was utilized for network training, and the Levenberg–Marquardt algorithm was employed to further enhance yield prediction. The input data were divided into three subsets: 70% for training, 15% for testing, and 15% for model validation. The predicted results were then compared with the experimental outcomes. The ANN model achieved an R 2 of 0.9487, indicating strong predictive performance. The analysis demonstrated that the ANN approach was highly effective and produced satisfactory results.
Kumar et al. (2026) studied this question.