Inverse problem analysis for identification of reaction kinetics constants in microreactors for biodiesel synthesis

7th European Thermal-Sciences Conference

Pontes, P.C.^1,2 ; Naveira-Cotta, C.P.^1

Dept. of Mechanical Engineering, COPPE, UFRJ, Universidade Federal Do Rio de Janeiro, Cidade Universitaria, Cx. Postal 68503, Rio de Janeiro, RJ

21945-970, Brazil^1

UNIFESSPA, Universidade Federal Do sul e Sudeste Do Pará, Santana do Araguaia, PA

68560-000, Brazil^2

关键词: Computational algorithm;    Computational simulation;    Eigenfunction expansions;    Generalized integral transforms;    Isothermal conditions;    Nonlinear mathematical model;    Reactive mass transfer;    Transesterification reaction;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/745/3/032101/pdf
DOI  :  10.1088/1742-6596/745/3/032101

来源: IOP

【 摘 要 】

The theoretical analysis for the design of microreactors in biodiesel production is a complicated task due to the complex liquid-liquid flow and mass transfer processes, and the transesterification reaction that takes place within these microsystems. Thus, computational simulation is an important tool that AIDS in understanding the physical-chemical phenomenon and, consequently, in determining the suitable conditions that maximize the conversion of triglycerides during the biodiesel synthesis. A diffusive-convective-reactive coupled nonlinear mathematical model, that governs the mass transfer process during the transesterification reaction in parallel plates microreactors, under isothermal conditions, is here described. A hybrid numerical-analytical solution via the Generalized Integral Transform Technique (GITT) for this partial differential system is developed and the eigenfunction expansions convergence rates are extensively analyzed and illustrated. The heuristic method of Particle Swarm Optimization (PSO) is applied in the inverse analysis of the proposed direct problem, to estimate the reaction kinetics constants, which is a critical step in the design of such microsystems. The results present a good agreement with the limited experimental data in the literature, but indicate that the GITT methodology combined with the PSO approach provide a reliable computational algorithm for direct-inverse analysis in such reactive mass transfer problems.