Spectral-based kinetic parameter estimation under the influence of colored noise

Because most parameter estimation approaches have been designed based on the assumption that there is only white Gaussian noise in the measured spectroscopic data so far, a kinetic parameter estimation approach under the influence of colored noise was proposed. For spectroscopic data with colored noise of the first-order auto-regressive process, the reaction kinetic parameter estimation was performed by combining observed data difference, collocation method and maximum likelihood principle. The results show that the kinetic parameters can be well estimated from spectroscopic data with colored noise by the proposed approach under the assumption that the colored noise model is completely known. Also, when only the model structure of colored noise is known, both the kinetic parameters and auto-correlation coefficient can be well estimated by the proposed approach simultaneously. The proposed approach has the better estimation precision than the approach derived based on the white Gaussian noise assumption. © 2021, Editorial Board of "Journal of Chemical Engineering of Chinese Universities". All right reserved.