Estimation of kinetic parameters for glycerol/alcohol dehydration reaction systems with incomplete chromatography data sets

• Published in: Journal of the Taiwan Institute of Chemical Engineers Vol. 60; pp. 185 - 198
• Main Authors: Chang, Jyh-Shyong, Deng, Jin-Han, Wang, Gow-Bin
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.03.2016
• Subjects: Dehydration reaction systems; Incomplete chromatography data sets; Ordinary differential equation models; Parameter estimation; Parameter estimation; Incomplete chromatography data sets; Ordinary differential equation models; Dehydration reaction systems
• ISSN: 1876-1070; 1876-1089
• DOI: 10.1016/j.jtice.2015.11.011

•Estimation of kinetic parameters for reaction systems with incomplete chromatography data sets was developed.•Parameters were estimated through the two-stage (TS) method augmented with carbon–hydrogen–oxygen (CHO) balance optimization.•One simulation example and one experimental example were examined, and the results confirmed the utility of the developed method. There is a shortage of pure ethers available for quantizing the gas chromatograph (GC) peaks that appear during the dehydration reactions of glycerol with alcohols. This problem can be overcome computationally by embedding the carbon–hydrogen–oxygen (CHO) balance optimization in the parameters estimation of the reaction system. A two-stage (TS) parameter estimation method was used to identify ordinary differential equation (ODE) models. This approach minimizes the squared residuals between the derivatives of the fitted values of available measurements and the rate changes of the states, used to estimate the unknown parameters. In this way, the problem of ill-conditioning, encountered in integrating the system equations, can be overcome. The resulting estimates are used as initial guesses for the next stage, in which the final estimated parameters are obtained by traditional minimization of the squared residuals between the estimated states and the measurements. Two examples were examined whose results confirmed the utility of the developed TS parameter estimation method and showed it to be easily implemented with a low requirement for computation time. The first example is a simulation (glycerol ether synthesis from glycerol with TBA) and the second example examines glycerol ether synthesis from glycerol with tert-butyl alcohol at elevated temperatures. Schematic of TS method for parameter estimation. Result of TS augmented CHO balance optimization method for ODE parameter estimation. [Display omitted]