MULTICOMPONENT KINETIC DETERMINATIONS USING ARTIFICIAL NEURAL NETWORKS

Neural networks were successfully used for multicomponent kinetic dete rminations of species with rate constant ratios approaching unity with out the aid of spectral discrimination. The ensuing method relies on t wo inputs describing the profile of the kinetic curve for each mixture , which is obtained by preprocessing kinetic data using nonlinear leas t-squares regression. A straightforward network architecture (2:4s:21) was used to resolve mixtures of 2- and 3-chlorophenol; the trained ne twork estimated the concentrations of both components in the mixture w ith a relative standard error of prediction of similar to 5%, which is much lower than that obtained with Kalman filtering. The effect of so me variables such as the rate constant and analyte concentration ratio s on the proposed multicomponent determination is discussed.