Title: Numerical analysis of parameter identifiability for a mathematical model of a chemical reaction
Authors: Liana Safiullina; Irek Gubaydullin
Addresses: Bashkir State University, 32 Zaki Validi Street, Ufa, 450076, Russia; Ufa State Petroleum Technological University, 1 Kosmonavtov Street, 450062, Ufa, Russia ' Ufa State Petroleum Technological University, 1 Kosmonavtov Street, 450062, Ufa, Russia; Institute of Petrochemistry and Catalysis of the UFRC RAS, 141 Prospekt Oktyabria, 450075, Ufa, Russia
Abstract: A numerical approach to the analysis of the parametric identifiability of a mathematical model of a chemical reaction, as well as the procedures aimed at elimination of non-identifiability and reduction of the mathematical model, are developed. The primary focus is on the reaction of low-temperature propane pyrolysis with the detailed scheme containing 157 reactions. The purpose of this work is to determine kinetic parameters of the model to be identified and to determine the parameters of the model which can be considered insignificant. The detailed propane pyrolysis scheme is reduced to a 30-reaction scheme. A kinetic model of low-temperature propane pyrolysis is proposed. The authors present the results of numerical simulation of three-dimensional dynamics of gas flow of propane pyrolysis in the reactor using the ANSYS Fluent software package and a reduced kinetic model.
Keywords: sensitivity analysis; mathematical model; Monte Carlo modelling; Sobol's method; parametric identifiability; numerical analysis; low-temperature propane pyrolysis; reaction mechanism; reduction of mechanism; chemical kinetics.
DOI: 10.1504/IJESMS.2020.111269
International Journal of Engineering Systems Modelling and Simulation, 2020 Vol.11 No.4, pp.207 - 213
Received: 01 Oct 2019
Accepted: 03 Mar 2020
Published online: 17 Nov 2020 *