Two-phase flows are often observed in the nuclear industry, within the primary and secondary circuits of power plants. The development of a comprehensive two-phase model is of crucial importance as it would enhance the understanding of the dynamics of these flows within the reactor core and steam generators. This would pave the way for the development of a 'digital reactor' enabling the optimization of the safety margins of nuclear facilities and proposing innovative designs.
Direct Numerical Simulation (DNS) offers the possibility to conduct small-scale numerical experiments to analyze simple configurations. With the increase in computational power and the use of interface tracking algorithms, the study of the evolution of bubble flows with great precision is now achievable. However, this approach remains costly and challenging to apply to industrial configurations. Averaged methods, which are less computationally intensive, enable the study of more complex configurations but require significant modeling. Upscaling, a method that allows the transition from DNS scale to averaged models, is not always easy to apply. This complexity is particularly obvious for intermittent phenomena related to wall presence and high gas ratio [1], such as the Taylor bubble.
Averaged models, aiming to reproduce the topologies of 'large interfaces,' use pseudo-interface tracking techniques or interface reconstruction [2]. They manage to qualitatively reproduce these large interfaces and provide satisfactory results for complex flows, away from walls. However, these models lack efficiency for configurations considering the presence of nearby solid structures. A more in-depth modeling approach is necessary to reproduce the behavior of large near-wall interfaces, taking into account various phenomena (turbulence, wall transfers, interfacial transfers, etc.).
The student will focus on simulating a Taylor bubble using an averaged interface reconstruction modeling approach. Reference data from the TrioCFD calculation code will be used. All cases will be simulated using the Neptune CFD calculation code, co-developed by CEA, EDF, FRAMATOME, and IRSN. Depending on the student's progress, the new model can be applied to increasingly complex configurations (impact of a bubble on a wall, on a mixing grid [7]), aiming to study interactions with solid structures. These configurations are of particular interest for industrial issues related to steam generators or fuel assembly bundles.


[1] A.O. Morgado, J.M. Miranda, J.D.P. Araújo, J.B.L.M. Campos, Review on vertical gas–liquid slug flow, International Journal of Multiphase Flow, Volume 85, 2016, Pages 348-368, ISSN 0301-9322. URL https://doi.org/10.1016/j.ijmultiphaseflow.2016.07.002.

[2] A. De Santis, M. Colombo, B.C. Hanson, M. Fairweather, A generalized multiphase modelling approach for multiscale flows, Journal of Computational Physics, Volume 436, 2021,
110321, ISSN 0021-9991. URL https://doi.org/10.1016/j.jcp.2021.110321.

[3] D. Bestion. The difficult challenge of a two-phase cfd modelling for all flow regimes. Nuclear
Engineering and Design, 279:116–125, 2014. ISSN 00295493. doi: 10.1016/j.nucengdes.
2014.04.006. URL http://dx.doi.org/10.1016/j.nucengdes.2014.04.006.

[4] Collins, R., Moraes, F., Davidson, J., & Harrison, D. (1978). The motion of a large gas bubble rising through liquid flowing in a tube. Journal of Fluid Mechanics, 89(3), 497-514. doi: 10.1017/S0022112078002700.

[5] Arijit Majumdar, P.K. Das, Rise of Taylor bubbles through power law fluids – Analytical modelling and numerical simulation, Chemical Engineering Science, Volume 205, 2019, Pages 83-93, ISSN 0009-2509. URL https://doi.org/10.1016/j.ces.2019.04.028.

[6] Brown, R.A.S., 1965. The mechanics of large gas bubbles in tubes: I. bubble velocities in stagnant liquids. Canad. J. Chem. Eng., vol. 43, 5 (10), pp. 217–223.

[7] Shuai Liu, Li Liu, Hanyang Gu, Ke Wang, Experimental study of gas-liquid flow patterns and void fraction in prototype 5 × 5 rod bundle channel using wire-mesh sensor, Annals of Nuclear Energy, Volume 171, 2022, 109022, ISSN 0306-4549. URL https://doi.org/10.1016/j.anucene.2022.109022.