Simplified modeling of butane dehydrogenation: deeper understanding of the system’s dynamics
- 20 June 2024
- journal article
- research article
- Published by IOP Publishing in Physica Scripta
- Vol. 99 (7) , 075250
Abstract
The dehydrogenation of butane is a catalytic process that involves the separation of hydrogen atoms from an organic molecule, and it relies on the presence of a catalyst Z, such as Ni, Pt, Pd, Fe, ZnO, Cr 2 O 3, or Fe 2 O 3. In this paper, we introduce two highly effective mathematical techniques known as Lumping Compartments (LC) and Intrinsic Low Dimensional Manifold (ILDM) to simplify the complexity of the dehydrogenation of butane. Initially, the model consists of seven nonlinear differential equations with eight parameters. To streamline the model, we applied LC to ten different cases, considering various initial states and parameters. As a result, the number of compartments significantly decreased from seven to three. The obtained computational results and total differences reveal a remarkable agreement between the complete model and the simplified version, demonstrating a high degree of predictive accuracy. Additionally, ILDM is employed to enhance the initial estimation of the slow invariant manifold. This enables us to easily distinguish between the fast and slow variations at each point, with the variations acting as chemical kinetics attractors. Fast chemical reactions are drawn towards these attractors, while slow chemical reactions occur within the manifolds.Keywords
This publication has 22 references indexed in Scilit:
- A reduction scheme for coupled Brownian harmonic oscillatorsJournal of Physics A: Mathematical and Theoretical, 2022
- Comparison of invariant surfaces in multi-route reaction mechanismsReaction Kinetics, Mechanisms and Catalysis, 2022
- Entropy production and lumping of species can effectively reduce complex cell signaling pathwaysPhysica Scripta, 2022
- Invariant manifold approach for quantifying the dynamics of highly inertial particles in steady and time-periodic incompressible flowsChaos: An Interdisciplinary Journal of Nonlinear Science, 2022
- Machine learning for integrating combustion chemistry in numerical simulationsEnergy and AI, 2021
- Chemistry reduction using machine learning trained from non-premixed micro-mixing modeling: Application to DNS of a syngas turbulent oxy-flame with side-wall effectsCombustion and Flame, 2020
- Automated proper lumping for simplification of linear physiologically based pharmacokinetic systemsJournal of Pharmacokinetics and Pharmacodynamics, 2019
- Proper lumping in systems biology modelsIET Systems Biology, 2009
- Invariant grids for reaction kineticsPhysica A: Statistical Mechanics and its Applications, 2004
- Method of invariant manifold for chemical kineticsChemical Engineering Science, 2003