Chemometric analysis of Raman spectroscopic data for process control applications

Two examples are given demonstrating the use of multivariate modeling in Raman process control applications. In one example, principal component analysis (PCA) and principal component regression (PCR) are used to model the curing of a high performance thermoset. The PCA results are found to give more accurate results when compared to univariate methods. In a second example, the octane number of gasoline is accurately modeled using partial least squares (PLS) regression analysis. For both examples, methods of normalization are considered in an effort to overcome the limitations of the single beam nature of Raman spectra.