A Simple Method for the Estimation of the Axial Dispersion Coefficient in Gas Flow
A simple method which is suitable for determining with reasonable precision the parameters of gas flow system has been proposed. An inverse boundary-value problem is considered. The model of gas flow with the Danckwert’s boundary conditions in a real measurement system has been analyzed and solved....
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Isfahan University of Technology
2018-01-01
|
| Series: | Journal of Applied Fluid Mechanics |
| Subjects: | |
| Online Access: | http://jafmonline.net/JournalArchive/download?file_ID=45813&issue_ID=249 |
| Summary: | A simple method which is suitable for determining with reasonable precision the parameters of gas flow system has been proposed. An inverse boundary-value problem is considered. The model of gas flow with the Danckwert’s boundary conditions in a real measurement system has been analyzed and solved. The tracer technique was applied to determine axial dispersion coefficient of gas phase and Pèclet number. These parameters are commonly used to characterize the flow behavior of fluids. Axial dispersion coefficients were estimated by comparing model solution with recorded TCD signal (an inverse problem as a method for model parameter estimation) employing the Laplace transform technique. The Gaver-Stehfest algorithm for the solution of the mathematical model has been applied. The proposed model of gas show a good agreement with the experimental data. The obtained results show that under operation conditions in the studied system the flow behaviour is neither plug flow nor perfect mixing. The described method is very fast in both experimental and computational part. Simple and errorless derivation of sophisticated model formulas has been possible by application of the Computer Algebra System-type program. The program also simplifies computations. Mathematical manipulations and computations were performed using program Maple®. |
|---|---|
| ISSN: | 1735-3572 |