Real-time principal component analysis using parallel Kalman filter networks for peak purity analysis

Abstract

A new approach for pedorming principal component analysls (PCA) during data acquisition is descrlbed. The method is based on a network of multllkmar models whkh are fit to data with the dlscrete Kaiman filter. Appilcatlon to absorbance matrices such as those obtained in chromatography wlth multiwavelength detection is considered. Multivariate data projected into two- and three-dlmenslonal subspaces are fit with linear and planar models, respectively. Model deviations, detected using principles of adaptive Kalman filtering, are used to elucidate the rank of the data set. Principal component eigenvectors are then constructed from the individual models. Results of this initial work using simulated and experimental data demonstrate that extraction of the flrst two principal components is readily accomplished and eigenvectors obtained are in good agreement wlth tradltlonai batch PCA results. Extendon to more principal components should be possible although It wlli increase the number and complexity of models. Advantages of the new algorithm include its recursive implementation, parallel structure, and ability to Indicate model errors as a function of time. The procedure should prove particularly useful for self-modeling curve resolution applications In chromatography.