Neural networks for prediction of liquid ternary phase behavior

Artificial neural network was investigated for emulating liquid ternary phase behavior. Reproducing the feed data and expanding the ternary phase pattern makes this approach able to predict the phase behavior beyond the collection of experimental measurements. Well-trained networks even can reproduce outlier data at different compositions and/or temperatures, acceptably.

کلیدواژه ها:

Neural network ، Ternary phase ، Liquid-liquid equilibrium