Martínez Pérez, Cristian A. (2021) Análisis del rango de validez entre modelos fluidodinámicos y flujos moleculares / Analysis of the validity ranger between fluid dynamics and molecular flows. Proyecto Integrador Ingeniería Nuclear, Universidad Nacional de Cuyo, Instituto Balseiro.
![[img]](/style/images/fileicons/application_pdf.png)
| PDF (Tesis) Español 5Mb |
Resumen en español
Se utilizan dos métodos computacionales completamente distintos para resolver el flujo en una tobera de expansión axial simétrica. Los métodos utilizados son Dinámica Computacional de Fluidos (Computational Fluid Dynamics CFD) y Simulación Directa de Monte Carlo (Direct Simulation Monte Carlo DSMC). Cada uno posee sus respectivas hipótesis y utiliza metodologías completamente diferentes. Mientras CFD está basado en la resolución de las ecuaciones diferenciales de Navier-Stokes, DSMC se basa en la simulación directa del comportamiento de las partículas que componen el flujo. Debido a que ambos metidos son distintos, se analiza la manera de imponer condiciones de borde equivalentes, a la vez que se discute sobre las hipótesis que garantizan la validez de ambos métodos. Además se discute sobre diversas formas de reducir los tiempos de cálculo para simulaciones realizadas con DSMC, y la influencia que posee la malla en dichos tiempos. En este trabajo se diseñó una tobera de expansión supersónica que permitiera comparar los resultados obtenidos mediante los dos métodos propuestos, se estudió el flujo de un gas simple (nitrógeno) en expansión desde una cámara a alta presión (252 kPa), realizando un análisis de las condiciones de validez y los errores involucrados en los cálculos. En el diseño de la tobera y el planteo del problema se halló la forma de colocar condiciones de borde equivalentes en ambos métodos, lo que permitió realizar comparaciones confiables entre sus resultados. De dichas comparaciones se pudo obtener resultados que permiten vericar la validez del método CFD en todas las zonas de flujo analizadas, en particular en la zona de la garganta de la tobera donde los cambios ocurren rápidamente. Los resultados obtenidos muestran que ambos métodos describen de manera similar el problema, tanto de manera cualitativa como cuantitativa, con algunas diferencias que exceden el error estadístico del método DSMC. Por ultimo, se implementó un método de resolución combinado, donde la parte de alta presión y bajas velocidades se resuelve mediante CFD y la zona de la garganta de la tobera y expansión libre se resuelven utilizando DSMC. Utilizando esta metodología se obtuvieron resultados similares pero con un costo computacional un orden menor al requerido para resolver el problema completo con DSMC.
Resumen en inglés
This thesis is aimed to the study and comparison between two differents methods that are used to solve the fluid dynamics of an axially symmetric nozzle. Those methods are Computational Fluid Dynamics (CFD) and Direct Simulation Monte Carlo (DSMC). Each one has different hypothesis regarding the validity of their solution and uses different methodologies. Whereas CFD is based on solving the Navier-Stokes' equations, DSMC is based on the direct simulation of the behaviour of the particles that form the fluid. Due to the fact that both methods are different, it is discussed the way of setting equivalent boundary conditions and the validity of their solutions. In this work a supersonic expansion nozzle was designed in order to study and compare the results of fluid dynamics obtained by the two proposed methods. The flow of a simple gas (nitrogen) in expansion from a chamber at high pressure (252 kPa) was investigated, performing an analysis of the conditions of validity and errors involved in the calculations. In order to compare correctly the results obtained with the two theoretical methods, in addition to the nozzle design, a methodology to translate the set of boundary conditions between the two methods was developed. From the comparisons of the calculated properties of the fluid it was possible to verify the validity of the CFD method in all the flow zones in the nozzle. In particular in the area of the nozzle throat where the changes occur quickly, wew found to be correctly described by both methods. The obtained results show that both methods describe the problem in a similar way, both qualitatively and quantitatively, with some differences that exceed the statistical error of the DSMC method. Finally, a combined resolution method was developed and implemented, where the high-pressure and flow speed region is solved by CFD and the nozzle-throat area and free-expansion are solved using DSMC. By using this methodology, results in good agreement with the individual methods were obtained, but with a ten-fold reduction in computational times when compared with the pure DSMC calculations.
| Tipo de objeto: | Tesis (Proyecto Integrador Ingeniería Nuclear) |
|---|---|
| Palabras Clave: | Supersonic flow; Flujo supersónico; Computational fluid dynamics; Dinámica de fluidos computacional; [Direct simulation Monte Carlo; Simulación directa de Monte Carlo, Continuum; Continuo; Validity] |
| Referencias: | [1] K. Anderko, G. O., K. R. Kummerer. General and Special Engineering Materials Science Vol. II. 1983. [2] Katopodes, N. D. Chapter 1 - basic concepts. En: N. D. Katopodes (ed.) Free-Surface Flow, pags. 2-98. Butterworth-Heinemann, 2019. URL https://www.sciencedirect.com/science/article/pii/B9780128154892000010. [3] Bird, G. A. Molecular gas dynamics and the direct simulation of gas flows. No 42 en Oxford Engineering Science Series, 2a editon. Clarendon Press, 1994. [4] World-Nuclear ansys-fluent | artemis | training center. http://artem.co.in/courses/ansys-fluent. Accessed: 2021-05-30. [5] Sandia National Lab sparta documentation. https://sparta.sandia.gov/. Accessed: 2021-05-1. [6] World-Nuclear uranium enrichment. https://world-nuclear.org/information-library/nuclear-fuel-cycle/conversion-enrichment-and-fabrication/uranium-enrichment.aspx. Accessed: 2021-05-31. [7] U.S.NRC uranium enrichment. https://www.nrc.gov/materials/fuel-cycle-fac/flur-enrichment.html#enrich. Accessed: 2021-05-31. [8] Feinberg, R. M., Hargrove, R. S. Overview of uranium atomic vapor laser isotope separation, 8 1993. doi:10.2172/10102839. URL https://www.osti.gov/biblio/10102839. [9] Shen, C. Rareed gas dynamics: fundamentals, simulations and micro flows. Springer Science & Business Media, 2006. [10] OSTI graeme a. bird. https://www.osti.gov/pages/servlets/purl/1595027. [11] Versteeg, H. K., Malalasekera, W. An introduction to computational fluid dynamics - the nite volume method. Addison-Wesley-Longman, 1995. [12] Ansys Fluentfluent documentation. https://www.afs.enea.it/project/neptunius/docs/fluent/index.htm. Accessed: 2021-05-1. [13] COMSOL Blog fem vs. fvm. https://www.comsol.com/blogs/fem-vs-fvm/. [14] Rader, D. J., Gallis, M. A., Torczynski, J. R., Wagner, W. Direct simulation monte carlo convergence behavior of the hard-sphere-gas thermal conductivity for fourier heat flow. Physics of Fluids, 18 (7), 077102, 2006. doi:10.1063/1.2213640. URL https://doi.org/10.1063/1.2213640. [15] Plimpton, S., Moore, S., Borner, A., Stagg, A., Koehler, T., Torczynski, J., et al. Direct simulation monte carlo on peta flop supercomputers and beyond. Physics of Fluids, 31, 086101, 08 2019. doi:10.1063/1.5108534. [16] Fang, Y., Liou, W. Micro fluid flow computations using a parallel dsmc code., 2002 2002. [17] GONZALEZ, J. Comparaciones numéricas entre métodos montecarlo y navier stokes. Inf. tec., SATL-CNEA, 2019. [18] Olson, S. E., Christlieb, A. J. Gridless dsmc. Journal of Computational Physics, 227 (17), 8035-8064, 2008. doi:https://doi.org/10.1016/j.jcp.2008.04.038. URL https://www.sciencedirect.com/science/article/pii/S002199910800257X. [19] W.W. Liou, Y. F. Implicit boundary conditions for direct simulation monte carlo method in mems flow predictions. Computer Modeling in Engineering & Sciences, 1 (4), 119-128, 2000. doi:10.3970/cmes.2000.001.571. URL http://www.techscience.com/CMES/v1n4/24713. [20] Fang, Y., Liou, W. W. Computations of the flow and heat transfer in microdevices using dsmc with implicit boundary conditions. J. Heat Transfer, 124 (2), 338-345, 2002. [21] Kundu, P., Cohen, I. Fluid Mechanics. Elsevier Science, 2010. URL https://books.google. com.ar/books?id=d9B5NElxUKwC. [22] NEURUS departamento de computación de alta prestación. https://neurus.dcap.cnea.gov. ar/wordpress/. Accessed: 2021-06-25. [23] Scholes, C., Kentish, S., Stevens, G. Carbon dioxide separation through polymeric membrane systems for flue gas applications. Recent Patents on Chemical Engineering, 1, 01 2010. doi:10.2174/1874478810801010052. |
| Materias: | Ingeniería nuclear |
| Divisiones: | Gcia. de área de Investigación y aplicaciones no nucleares > Gcia. LASIE |
| Código ID: | 1028 |
| Depositado Por: | Tamara Cárcamo |
| Depositado En: | 28 Abr 2022 14:44 |
| Última Modificación: | 28 Abr 2022 14:44 |
Personal del repositorio solamente: página de control del documento
