chapter  5
28 Pages

Substantial and formal approaches to analysis of kinetic data

Analysis of kinetic data is a necessary stage of investigation of processes mechanisms and predicting of practically important properties. Hard approach to such analysis is a traditional inverse problem of chemical kinetics solving with prescribed a priori system of kinetic equations. Alternative is so-called soft approach in which kinetic model is clearly not used. Both approaches have strong and weak sides. In this Chapter we compare these methods, consider methodological questions topical for analysis of kinetic data. Work is illustrated by two examples. In the first one we considered the technique of estimation of parameters of kinetic model. Hard method of Successive Bayesian Estimation (SBE) is used for reception of kinetic information from spectrum data in the case when spectra of "pure" components are unknown. It compared with various soft methods. In the second example the possibility of estimation of antioxidants activity by differential scanning calorimetry (DSC) is studied. Here we used the both hard approach with the help of non-linear regression analysis of data and soft approach based on the method of projection on latent structures in combination with novel method of simple interval calculation. Key words: estimation of kinetic parameters, consistent Bayesian estimation, simple interval calculation, non-linear regression, projection on latent structures.