466075 Identification Methodology for Stirred and Plug-Flow Reactors

Thursday, November 17, 2016: 8:30 AM
Carmel II (Hotel Nikko San Francisco)
Alfredo Bermúdez1, Noemí Esteban2, José L. Ferrín1 and José F. Rodríguez-Calo3, (1)Matemática Aplicada, Universidad de Santiago de Compostela, Santiago de Compostela, Spain, (2)Instituto Tecnológico de Matemática Industrial (ITMATI), Santiago de Compostela, Spain, (3)Centro de Tecnología de Repsol, Móstoles (Madrid), Spain

Nowadays, optimal operation of chemical reactors is a critical element in the economics of several processes of the chemical industry as, for instance, oil refining and petrochemistry. This optimal operation can be achieved if an accurate model of the chemical kinetics is available allowing us to simulate the reactor operation. For this purpose a major issue is the identification of the reaction mechanisms and their kinetics among different suitable functional expressions, by using experimental measurements and inverse problems methodology.

The present work deals with the identification problem of kinetic models and their corresponding parameters which better fit a set of experimental data, when the reaction system is supposed to be known. As said above, the optimal kinetic models are selected from a set of allowed functional expressions. For solving this inverse problem, several techniques can be considered, such as the so-called differential, integral and incremental methods (see [1]). When dealing with a stirred tank reactor (STR) our choice is the combination of two methods in cascade: the incremental method and the integral one (see [2] or [3]). Whereas in the case of plug flow reactors (PFR) we propose a multi-start strategy and then the integral method combined with the variable neighborhood search (VNS) to move towards a better solution (see [4] or [5]). In all cases, the adjoint method is used to compute the gradient of the objective function.

Often, information on temperature is not available and then the energy equation has to be considered in the model. Moreover, when experimental data of temperature are provided, differences between calculated values and experimental ones can be included into the objective function to be minimized. In our work, we consider both situations: either the temperature is given as a data and does not change inside the reactor or it is computed from the energy equation.

Finally, the applicability and performance of the proposed methodology will be shown through different numerical simulations for STRs and for PFRs.


Part of this research was developed as an activity in the Joint Research Unit Repsol-ITMATI (code file: IN853A 2014/03) which is funded by FEDER, the Galician Agency for Innovation (GAIN) and the Ministry of Economy and Competitiveness in the framework of the Spanish Strategy for Innovation in Galicia. It was also supported by the Spanish MICINN project MTM2013-43745-R and by the Xunta de Galicia under research project GRC13-014.


[1] A. Bermúez, E. Carrizosa and N. Esteban. A two steps identification process in stirred tank reactors. The incremental and integral methods. In preparation.

[2] A. Bermúdez, N. Esteban, J.L. Ferrín, J.F. Rodríguez-Calo, and M.R. Sillero-Denamiel, Identification problem in plug-flow chemical reactors using the adjoint method. In preparation.

[3] N. Bhatt, N. Kerimoglu, M. Amrhein, W. Marquardt, and D. Bonvin. Incremental identification of reaction systems-A comparison between rate-based and extent-based approaches. Chemical Engineering Science, 84: 24-38, 2012.

[4] M. Brendel, D. Bonvin and W. Marquardt. Incremental identification of kinetic models for homogeneous reaction systems. Chemical Engineering Science, 61:5404-5420, 2006.

[5] D. Rodrigues, J. Billeter and D. Bonvin. Incremental Model Identification of Distributed Two-phase Reaction Systems. IFAC-PapersOnLine, 48(8):266-271, 2015.

Extended Abstract: File Not Uploaded
See more of this Session: Process Modeling and Identification
See more of this Group/Topical: Computing and Systems Technology Division