Use of Artificial Neural Networks for Retention Modelling in Ion Chromatography

The aim of this work was to develop an empirical model for retention of inorganic anions (fluoride, chloride, nitrite, sulphate, bromide, nitrate, and phosphate) in suppressed ion chromatography with hydroxide selective stationary phases using artificial neural networks. Three-layer feed-forward neural network trained with a Levenberg-Marquardt batch error back propagation algorithm has been used to model retention mechanisms of inorganic anions with respect to the mobile phase parameters. The number of hidden layer nodes of the neural network and the number of iteration steps were optimized in order to obtain the best possible retention model. This Study shows that an optimized artificial neural network is a very accurate and fast retention modelling tool to model various inherent linear and non-linear relationships of retention behaviour. This has been proven by developing the neural network retention model with average relative errors of 0.88% obtained using only 300 iteration steps