Application of Artificial Neural Networks in the Optimization of HPLC Mobile-Phase Parameters

Publisher Website
Google Scholar
Abstract
The prediction capability of forward feed neural networks was tested c computer generated capacity factors. The capacity factors were simulate from equations reflecting the contribution of mobile phase changes in pH-organic modifier concentration, and ion-pair concentration. Simulated da1 allows an appropriate experimental design which assures the training of th network does not involve memorization but guarantees the network w generalize. The use of different mathematical forms to calculate th behaviour of capacity factor with changes in pH, methanol concentrator and ion-pair concentration permitted us to explore the capability of neur, networks to fit a variety of curves. Each of the independent variables wer studied separately, and then in combination. The effect of variabl transformation played a very important role in effective training of th network. The neural network output equations were used to formulate nonlinear regression problem and the behaviour of this model was compare to the neural network system. When the neural network systems had onl sufficient processing units needed to solve the problem, nonlinear regressio models and neural networks arrived at identical solutions. When the networ contained excessive neurons, nonlinear regression techniques were unstable, having high intraparameter correlations and showing matrix singularity.