A combination of target testing and an iterative determinant analysis is incorporated into a mixture analysis method that achieves data decomposition in the spectral space by successive average orthogonalization (SAO) and the estimation of the pure component spectra by iterative target transformation factor analysis (ITTFA). This mixture analysis method starts with the resolution of component spectra. Hence, the selection of targets for further transformation is critical when a component spectrum contains multiple peaks. Various synthesized spectra, as well as IR spectra collected during the formation of a polyurethane foam, have been analyzed. SAO is shown to produce eigenvectors comparable to those obtained by some well-established methods, such as SVD and NIPALS. It is preferred in this research because it allows the processing of a potentially unlimited number of spectra and linearly dependent data sets. The results of the analysis of both the synthesized spectra and the real IR spectra demonstrate th...
Read full abstract7-days of FREE Audio papers, translation & more with Prime
7-days of FREE Prime access