Application of artificial neural network in predicting the extraction yield of essential oils of Diplotaenia cachrydifolia by supercritical fluid extraction

In this study, a three-layer artificial neural network (ANN) model was investigated to predict the extraction yield of essential oils from Diplotaenia cachrydifolia by supercritical fluid extraction. A multilayer feed-forward neural network trained with an error back-propagation algorithm was employed for developing a predictive model. The input parameters chosen of the model are pressure, temperature, extraction time and modifier volume while the extraction yield of essential oils is the output. The Levenberg–Marquardt (LM) algorithm was used to train ANN. The results showed that a network with five hidden neurons was highly accurate in predicting the extraction yield of essential oils of D. cachrydifolia. The mean squared error (MSE) and coefficient of determination (R2) between the actual and predicted values were determined as 0.0014 and 0.9983 for training, 0.0184 and 0.9542 for validation and 0.0221 and 0.9703 for testing date sets. The main components that were extracted with SFE were dillapiole (30.2%), limonene (13.7%) and α-calacorene (20.1%).