Modeling of muskmelon seed oil extraction under supercritical carbon dioxide condition: article / Sakinah Khaidzir

From a secondary supercritical extraction data of muskmelon seed which was obtained from the work done by J. Prakash Maran and B. Priya, a suitable mathematical model was found for the oil yield extracted. The reason this study was done is to obtain a model for the process which will be able to predict oil yield of future experiments on supercritical carbon dioxide extraction of muskmelon seed oil. An application of Artificial Neural Network model was used where it was trained until desired output was obtained. A simple mathematical model was derived by applying the equation of straight line to the regression plot of the model. The regression value of the plot is 0.97. The average absolute deviation obtained for the comparison of actual and model oil yield is 2.63%. The actual oil yield and model values show good agreement with each other thus making the model obtained reliable to be used in the future.