ESTIMATION OF PARAMETERS OF KINETIC COMPARTMENTAL-MODELS BY USE OF COMPUTATIONAL NEURAL NETWORKS

A novel methodological approach to the estimation of parameters involv ed in multicomponent kinetic determinations from real kinetic data by use of computational or artificial neural networks (ANNs) is proposed. The ANN input data used are also estimates obtained by using the Leve nberg-Marquardt method in the form of an approximate nonlinear functio n that is the sum of the two expressions associated with the pseudo-fi rst-order kinetics of the two mixture components. The performance of t he optimized network architecture, 2:4s:21, was tested at variable rat e constant ratios. The reduced dimensions of the network input space o btained using the Kolmogorov-Sprecher theorem result in improved limit s of precision in estimating parameters at near-unity rate constant ra tios. Experiments with real kinetic data provided a relative standard error of prediction of 2.47% and 4.23% for the two mixture components. These errors are much smaller than those obtained with existing alter native methods, particularly at the low rate constant ratio involved ( 1.37).