ESTIMATING DIFFUSION-COEFFICIENTS OF A MICELLAR SYSTEM USING AN ARTIFICIAL NEURAL-NETWORK

A three-layer feedforward artificial neural network (ANN) model, train ed using the error-back-propagation algorithm has been developed to es timate the diffusion coefficient of sodium dodecyl sulfate (SDS) micel lar system over a wide range of operating parameters such as temperatu re and concentrations of SDS and NaCl. The network model validates the experimentally observed qualitative and quantitative trends. The opti mal model parameters in terms of network weights have been estimated a nd can be used for computing diffusion coefficients over wide-ranging experimental conditions. (C) 1995 Academic Press, Inc.