Universidade Federal de Santa Maria

Ci. e Nat., Santa Maria v.42, Special Edition: Micrometeorologia, e35, 2020

DOI:10.5902/2179460X53217

ISSN 2179-460X

Received: 17/08/20  Accepted: 17/08/20  Published: 28/08/20

 

 

Special Edition

 

Atmospheric Flow at Alcântara Launch Center

 

O escoamento atmosférico no Centro de Lançamento de Alcântara

 

Karine Klippel ^I

Elisa Valentim Goulart ^II

Gilberto Fisch ^III

Neyval Costa Reis Junior ^IV

Cayo Prado F. Francisco ^V

 

^I Universidade Federal do Espírito Santo, Vitória, Brazil. E-mail: karineklippel@gmail.com.

^II Universidade Federal do Espírito Santo, Vitória, Brazil. E-mail: elisavalentim@gmail.com.

^III Instituto de Aeronáutica e Espaço, São José dos Campos, Brazil. E-mail: fisch.gilberto@gmail.com.

^IV Universidade Federal do Espírito Santo, Vitória, Brazil. E-mail: neyval@gmail.com.

^V Instituto de Aeronáutica e Espaço, São José dos Campos, Brazil. E-mail: cayo.francisco@gmail.com.

 

 

ABSTRACT

The atmospheric flow at Alcântara Launch Center (CLA) was studied using Computational Fluid Dynamics (CFD) techniques. To characterize the region were considered the coastal cliff and the Integration Mobile Tower, called TMI, both within the launching and preparation area (SPL). In this study, the cliff was represented by a step of 90° with 40 meters of height. The inlet velocity profile was elaborate according to the power law, with exponent of 0.11, freestream velocity of 20 m/s and Reynolds number of 4.3 x 10⁵, adopting neutral atmosphere. Three wind directions were considered, 90º, 125º and 135°. The numerical model used was the Reynolds Stress Model (RSM), based on the Reynolds-Averaged Navier-Stokes (RANS) equations. The solution of the equations was obtained by ANSYS FLUENT 19, which uses the finite volume method. The results showed good agreement with the wind tunnel tests especially for wind direction perpendicular to the cliff. The incident wind direction strongly influences the flow dynamics in the SPL forming a helicoidal vortex over the coastal cliff the higher the wind slope.

Keywords: Flow around an obstacle; Computational Fluid Dynamics (CFD); Atmospheric turbulence; Rocket launch.

 

 

RESUMO

O escoamento atmosférico no Centro de Lançamento de Alcântara (CLA) foi estudado utilizando técnicas de Dinâmica dos Fluidos Computacional (CFD). Para caracterizar a região foi considerada a presença da falésia e da Torre Móvel de Integração (TMI), ambas dentro do Setor de Preparação e Lançamento (SPL). Nesse estudo, a falésia de 40 metros de altura foi representada por um degrau de 90°. O perfil de velocidade de entrada foi elaborado de acordo com a lei de potência, com expoente de 0,11, velocidade média de 20 m/s e número de Reynolds de 4,3 x 10⁵, considerando a atmosfera com estabilidade neutra. Três direções de vento foram testadas, 90º, 125º e 135°. O modelo numérico utilizado foi o Reynolds Stress Model (RSM), baseado nas equações de Reynolds-Averaged Navier-Stokes (RANS). A solução das equações foi obtida pelo software ANSYS FLUENT 19, que utiliza o método dos volumes finitos. Os resultados apresentaram boa concordância com os experimentos de túnel de vento especialmente para direção do vento perpendicular à falésia. A direção predominante do vento influencia fortemente a dinâmica do escoamento atmosférico no SPL, formando um vórtice helicoidal sobre a falésia associado à rotação do vento.

Palavras-chave: Escoamento ao redor de um obstáculo; Dinâmica dos Fluidos Computacional (CFD); Turbulência atmosférica; Lançamento de foguetes.

 

 

1 INTRODUCTION

The Alcântara Launch Center (CLA) is the main Brazil´s gateway to space, located in the north part of northeast Brazil. The Launch Pad Area (SPL) is positioned 150 meters from the coastline, on an irregular cliff. Therefore, some important points for the atmospheric flow passing over the SPL are observed: abrupt change of roughness (smooth surface of the ocean to the rough surface of the continent) and topographic variation caused by the cliff. Furthermore, the launching platform has some buildings nearby necessary for rocket integration as the Mobile Integration Tower (TMI) with an exit tower coupled (Figure 1). The space vehicle is prepared and assembled inside the TMI, which is then displaced by approximately 55 meters to start the rocket launching process. In this situation the vehicle is subject to the wind and atmospheric turbulence action coming from the ocean.

 

Figure 1 – Launch Pad Area (SPL)

Source: AEB, 2019.

It is well known that the flow around obstacles is dominated by shear-generated mechanical turbulence. In this way, the presence of buildings modifies the flow pattern and forms some characteristic regions like separation and recirculation zones and turbulence wake behind the obstacle (ARYA, 1999). Thus, understanding of the flow pattern around the launch pad buildings and how the flow changes due the geography of the region are important to establishing the conditions in which the CLA is exposed and providing the necessary basis for development of rockets as well as the safety procedures during the launchings. Another important environmental issue is associated with the pollutant dispersion emitted by the launches, such as HCl, and to the wind loads on the rockets.

In this context, wind tunnel (WT) tests and numerical simulations are important tools and they have been used previously to understand better the atmospheric flow in the CLA region. Roballo, Fisch and Girardi (2009) were pioneers in these analyzes. The authors conducted wind tunnel tests considering the presence of the cliff, represented by a step with two different inclinations: 70º and 90º. In addition, some tests were performed with elements that represented the surface roughness of the region. The best results were obtained for the cliff of 90º and without the presence of the roughness elements. It was observed typical flow characteristics after the cliff: detachment, recirculation zone and reattachment.

Pires et al. (2010) and Pires et al. (2011) analyzed the atmospheric flow in a coastal cliff of different configurations using WT experiments and 2D numerical simulations, respectively. Pires et al. (2010) also considered the presence of a rectangular building representing the TMI. The results showed a modification of the atmospheric flow near the surface when reaching the coastal cliff and the formation of a recirculation region. The vorticity generated downwind of the coastal cliff has a strong turbulence. The presence of the TMI results in a downwind turbulence like the generated by coastal cliff. For coastal cliffs of the same height and different inclination angles, the intensity of the maximum vorticity remains the same, but with increasing angles, the inner boundary layer (IBL) heights suffer a small decrease.

Additionally, Souza, Fisch and Goulart (2015) conducted 3D numerical simulations using the Computational Fluid Dynamics (CFD) technique considering the coastal cliff (90º) and the TMI. The results showed vortex structures next to cliff and downstream the TMI. A reduction of the flow velocity was also observed as it passed by the cliff.

Recently, Faria, Avelar and Fisch (2019) performed WT tests using a scaled model of the SPL with different cliff inclinations and incident wind intensity and direction. The vortexes pattern behind the TMI or the exit tower is strongly influenced by the wind incidence angles. These vortexes may induce vibrations or load excess on these buildings structures if they remain there for a long time.

However, few investigations of the atmospheric flow in the CLA were performed by computational methods so far, especially in microscale aspects. Furthermore, the complementation the information already acquired by WT experiments is important to promote new analyses. Thus, the objective of this study was to simulate the atmospheric flow in the CLA region using CFD techniques, inserting the buildings and the topography of the site.

 

 

2 Material and Methods

The test cases are based on experiment conducted by Faria, Avelar and Fisch (2019) in a boundary-layer wind tunnel and their results were used for validation of the numerical simulations. It was considered a freestream flow with velocity of U_∞ = 20 m/s, Reynolds number of 4.3 x 10⁵ and turbulence intensity ranging from 0.05 to 0.10.

Three obstacles (representing the TMI, exit tower and engine room) and a step (representing the cliff) were considered in geometry, 1:120 scale. The TMI with 270 x 85 mm dimensions. The exit tower, the engine room and the cliff with 210 mm, 35 mm and 330 mm of height, respectively. The domain was sufficiently far from the obstacle to avoid any influence on the results, with height of 10h_cliff, width of 10h_TMI and length behind TMI of 10h_TMI (Figure 2).

 

Figure 2 – Computational domain

 

Three incident wind directions α (90º, 125° and 135°, corresponding to cases 1, 2 and 3, respectively) were considered. The wind direction variation was obtained from the rotation of the geometry in the computational domain.

The Reynolds Averaged Navier-Stokes simulation was used to simulate incompressible flow with ρ = 1.225 kg/m3 and μ = 1.7894 x 10−5 kg/m.s. The turbulence model used to solve the conservations equations was the second-order Reynolds Stress Model (RSM) model, ω–baseline, which obtained a better performance for this case when compared to κ-ω SST first order model. This model considers the anisotropy of turbulent viscosity and closes the system calculating a transport equation for each component of the Reynolds stress tensor.

The incident turbulent velocity profile was approximated using a power law at the inlet boundary, according the wind tunnel experiments:

(1)

in which  is the mean velocity at height , is the mean velocity at reference height  (10 meters according World Meteorological Organization - WMO) and,  is 0.11, determined according to Hsu, Meindl and Gilhousen (1994), who performed measurements on the sea surface considering near neutral stability. The wind profile at CLA is so strong that the mechanical turbulence produces a neutral atmosphere.

At the outlet boundary, a zero-gauge pressure condition is adopted. Symmetry conditions are applied to the lateral boundaries. At the top, a free-slip condition is applied. No-slip boundary condition is applied to all solid surfaces.

The numerical simulations were performed using the commercial software ANSYS Fluent version 19, which employs the finite volume method to discretize the conservation equations. The algorithm SIMPLE, proposed by Patankar (1980), was used for pressure-velocity coupling. A second order scheme was chosen to spatial discretization of momentum, specific dissipation energy rate and Reynolds stresses variables. The structured meshes used in the cases have around 7 million of elements and nodes.

 

 

3 Results and Discussion

The influence of the wind direction on the atmospheric flow in SPL was analyzed for three cases. The Figure 3 presents the streamlines in the central XZ plane for these. It can be observed that for all wind directions there are typical regions like separation and recirculation zones. A separation zone is formed downstream the coastal cliff, with high velocities. For case 1 (90º), there is a recirculation zone below the separation zone, with low velocities in the opposite direction of the flow (Figure 3a). For cases 2 and 3 (125º and 135º), a disturbance of the flow is observed and a helicoidal vortex on the cliff the greater the wind slope (Figures 3b and 3c). This vortex is due to cliff inclination, which creates a velocity component parallel to its topography, and could be seen in Figures 4b and 4c.

 

Figure 3 – Streamlines in the central XZ plane for wind direction of: a) 90º; b) 125º and c) 135º. The arrow indicates the flow direction

 

From Figure 3 it is possible to notice that the engine room presented small influence on the flow. But, the presence of the other buildings (TMI and the exit tower) generates a new separation zone. For case 1 (90º) the flow detached due to the cliff has not yet reached the reattachment point when it arrives the TMI and, therefore, this building further delays its reattachment (Figure 3a). For all cases, there are formation of a recirculation zone behind the TMI, with the formation of large low-velocity vortex.

Saeedi, LePoudre and Wang (2014) analyzed the turbulent wake behind a building with high aspect ratio and Reynolds number in order of 10⁴ in developing boundary layer through direct numerical simulation. They verified that due of the relatively high Reynolds number and the high aspect ratio of the building the wake is widely spread behind the obstacle and exhibits complex and energetic vortex motions. This result was also observed in this study.

 

Figure 4 – 3D streamlines in the domain for wind direction of: a) 90º; b) 125º and c) 135º. The arrow indicates the flow direction

 

Luo et al. (2011) studied models of cuboid obstacles to characterize the three-dimensional responses of airflow behind obstacles to variations in the incident flow in wind tunnel tests. As well in the present study, the flow patterns behind obstacles were complicated by changes in the incidence angles. The flow separated both horizontally and vertically, creating reverse flow zones downwind of the obstacles. The horizontal plane is characterized by two asymmetrical around the TMI, opposing reverse vortices just at flow incidence angle 90º. The vertical plane presents a single vortex behind TMI for all angles (Figure 3).

The Figures 5, 6 and 7 show the comparison between numerical simulations and wind tunnel tests of Faria, Avelar and Fisch (2019) for wind directions of 90º, 125º and 135°, respectively. The main flow characteristics (recirculation and separation zones) were observed in both results. For wind direction of 90º (Figure 5) there is a recirculation zone just downstream the cliff that extends to the TMI. Behind the TMI the velocities are near zero. Considering the vorticity field, the higher vorticities (close to 1000 s^-1) were found at flow separation zone in the cliff and the top of TMI. It was observed a lower vorticity intensity above TMI compared to the other cases.

For wind direction of 125º and 135º (Figures 6 and 7), a strong recirculation zone is observed behind the TMI for both results. For numerical simulations, there is a stronger recirculation zone downstream the cliff with larger vorticity than wind tunnel results. The higher vorticities (close to 1000 s^-1) were found at flow separation and recirculation zones in the cliff and TMI.

 

Figure 5 – Comparison between (a) numerical results and (b) wind tunnel tests of Faria, Avelar and Fisch (2019), for wind direction of 90º. The figures show the velocity vectors, the streamlines and the vorticity XZ plane. The black regions are the buildings (TMI and engine room). The gray regions in figure (b) are the Faraday shield towers, part of the WT model. The red arrow indicates the coastal cliff

 

Figure 6 – Comparison between (a) numerical results and (b) wind tunnel tests of Faria, Avelar and Fisch (2019), for wind direction of 125º. The figures show the velocity vectors, the streamlines and the vorticity XZ plane. The black region is the TMI. The gray regions in figure (b) are the Faraday shield towers, part of the WT model. The red arrow indicates the coastal cliff

 

Figure 7 – Comparison between (a) numerical results and (b) wind tunnel tests of Faria, Avelar and Fisch (2019), for wind direction of 135º. The figures show the velocity vectors, the streamlines and the vorticity XZ plane. The black region is the TMI. The gray regions in figure (b) are the Faraday shield towers, part of the WT model. The red arrow indicates the coastal cliff

 

 

4 Conclusion

Comparing numerical simulations and wind tunnel tests, the main flow characteristics like separation and recirculation zones were observed in both cases. Also, the vorticity field had similar characteristics. Recirculation zones were observed downstream the cliff and behind the TMI. It was observed that the coastal cliff generates a flow separation zone and the presence of the TMI generates a new separation of the boundary-layer. It can be concluded, therefore, that the presence of the cliff and of the buildings/obstacles change the flow dynamics in the CLA region, generating typical flow regions. The predominant wind direction has an important role in the atmospheric flow, causing changes in the characteristics of these formed regions.

 

 

Acknowledgments

The authors acknowledge the financial support of the National Council for Scientific and Technological Development (CNPq) throughout the Project Estudo do escoamento atmosférico no Centro de Lançamento de Alcântara e sua influência para o lançamento de veículos espaciais (n° 403899/2016-8). Also, the first author would like to thank CNPq for her Master of Science scholarship (nº 133209/2018-0).

 

 

References

ARYA, P. S. Introduction to micrometeorology. [S.l.]: Elsevier, 1999. v. 79.

FARIA, A. F.; AVELAR, A. C.; FISCH, G. Wind Tunnel Investigation of the Wind Patterns in the Launching Pad Area of the Brazilian Alcântara Launch Center. Journal of Aerospace Technology and Management, v. 11, 2019.

HSU, S.; MEINDL, E. A.; GILHOUSEN, D. B. Determining the power-law wind-profile exponent under near-neutral stability conditions at sea. Journal of Applied Meteorology, v. 33, n. 6, p. 757-765, 1994.

LUO, W.; DONG, Z.; QIAN, G.; LU, J. Wind tunnel simulation of the three-dimensional airflow patterns behind cuboid obstacles at different angles of wind incidence, and their significance for the formation of sand shadows. Geomorphology, v. 139, p. 258-270, 2012.

PATANKAR, S. Numerical heat transfer and fluid flow. CRC press, 1980.

PIRES, L. B. M.; ROBALLO, S. T.; FISCH, G.; AVELAR, A. C.; GIRARDI, R. M.; GIELOW, R. Atmospheric flow measurements using the PIV and HWA techniques. Journal of Aerospace Technology and Management, v. 2, n. 2, p. 127-136, 2010.

PIRES, L. B. M..; SOUZA, L. F.; FISCH, G.; GIELOW, R. Numerical study of the atmospheric flow over a coastal cliff. International Journal for Numerical Methods in Fluids, v. 67, n. 5, p. 599-608, 2011.

ROBALLO, S. T.; FISCH, G.; GIRARDI, R. d. M. Escoamento Atmosférico no Centro de Lançamento de Alcântara (CLA): PARTE II – Ensaios no Túnel de Vento. Revista Brasileira de Meteorologia, v. 24, n. 01, 2009.

SAEEDI, M.; LEPOUDRE, P. P.; WANG, B. C. Direct numerical simulation of turbulent wake behind a surface-mounted square cylinder. Journal of Fluids and Structures, v. 51, p. 20-39, 2014.

SOUZA, B. H. de; FISCH, G. F.; GOULART, E. V. Simulação do escoamento atmosférico no Centro de Lançamento de Alcântara (CLA) utilizando a técnica de CFD. In: WORKSHOP BRASILEIRO DE MICROMETEOROLOGIA, 9., 2015, Santa Maria - RS. Annals... Santa Maria - RS: UFSM, 2015.