Series on Biomechanics, Vol.29, No. 2-3 (2015), 31-38
Computational fluid dynamics for the nordic combined skiing jump N. Gardana, J. Laheurte b, E. Gouy c, N. Dey d, E. Abdi e, U. Asgher f, M. A Choukou g, A. Schneider h, R. Taiar i a
CAD and CAE R&D Center DINCCS of Micado;
[email protected] Fédération Française de Ski ;
[email protected],
[email protected] d Bengal College of Engineering & Technology, India;
[email protected] e E. Orange School District and Montclair State University, USA ;
[email protected] f National University of Science and Technology (NUST), Islamabad, Pakistan;
[email protected] g Laval University, Québec. Department of Family Medicine;
[email protected] h URCA/CReSTIC/NUM3D, France;
[email protected] i Université de Reims Champagne Ardennes ;
[email protected] b,c
Abstract The Nordic Combined is a winter sport which athletes compete in two specific and very different disciplines: the cross-country skiing and the ski jumping. In this study, we focus on the ski jumping and more specifically on the take-off and the stable flight phases of the high level of athlete’s behaviors. This production is performed in partnership with the French team of the Nordic Combined. The aim of our paper is to develop combined methodologies between the experimental data approaching from the kinematical Vicon system and those from the CFD (Computational Fluid Dynamics) methods. The simulation of the body movement will permit to understand and improve the posture of the skier during jumps in addition to distribution of the air pressure on the body of the skier. This will allow to develop and design a specific and dedicated form of underwear. The aim of this study is to present a robust 3D methodology capable of improving the skier jump. The developed model can be used with specific three-dimensional athlete data in two ways: 1) the aerodynamic behavior of the skier in the stable flight phase (study of the aerodynamic drag) and 2) the air pressure influence on the skier. In order to design the specific and dedicated underwear. Keywords: Computational fluid dynamics, ski jumping, methodology, virtual wind tunnel
1. Introduction The Nordic Combined is a winter sport which athletes compete in two specific and very different disciplines: the cross-country skiing and the ski jumping. Performance in ski jumping is characterized by the length of the jump. All parameters of the performance require effective coordination of different jump phases which comprises run, take-off, flight and landing. The latest published literature review illustrated the importance of the take-off phases in high-level performance. Many methods were developed and focused on the quantification of the athletes’ behavior during different phases of the movement. In a study Chardonnensn et al. [2], 2012 developed a system to measure the kinematics during the entire ski jump sequence. A very complete database of measurements based upon the state of the art and results of the project were provided. To analyze an early flight phase, Virmavirta et al. [19], 2005 used two genlocked high-speed video cameras installed at the Winter Olympic Games in Salt Lake City. In order to undertake the kinematic analysis, a model skeleton skier and the equipment were added to videos. Thus, the authors were able to examine a wide area of the early flight dynamic including the jumpers’ velocity and other selected angular parameters. Consuming an industrial wind tunnel, Chowdhury et al. [3], 2011, developed an experimental setup to evaluate the aerodynamic performance and its influence on ski jumping suits. The developed system allows the experimental evaluation of drag and lifts for all types of ski gear and can be expanded as a tool for the airflow visualizations. Various speeds have been tested to analyze the influence on the skier and more particularly on the suit. All tests were carried out at a fixed position: a
31
N. Gardan et al. / Computational fluid dynamics for the nordic combined skiing jump
unique hydrodynamic posture has been tested based on the research of Schmölzer [14], 2005 and Müller, 2005 [12]. Numerous authors expended a wind tunnel to quantify the athlete’s behavior in the ski jumping context (Reisenberger et al. [13], 2004, Seo et al. [15], 2004, and Barelle [1], 2011). The basic idea of mentioned researchers was to start from simple geometric configurations to understand the different influences of sharp or rounded form of “the body edges, thickness, roughness and inclination relative to the airflow”. In regards to contribution of the suit for performance, the international ski federation imposes the utilization of the same material and characteristics such as air permeability and thickness for all high-level athletes. The difference can be found only in the underwear characteristics, which assists athletes to improve performance. The influence of textile surface, especially with the help of a wind tunnel is described by Chowdhury et al. [4], 2009 and 2011. The modeling and simulation of take-off phase in the movement under application of various stresses imposed on athletes will assist in controlling of training. This will occur in order to predict and optimize performance. The aim of this current study is to develop a methodology that allows the utilization of CFD (Computational Fluid Dynamics) to predict the influence of the underwear suit on performance. 2. Methodology The skier’s posture used in the stable flight phase influences the aerodynamic forces in many ways. The result of aerodynamic forces on the skier are divided into two components, i.e.1) the drag force and 2) the lift force (see figure 1). Lift and drag forces are direct result of pressure differentials generated by flowing fluid around the skier’s body. In a very simplistic approach, the drag force acts in a direction that is opposite of the oncoming air flow and the lift force acts in a direction that is perpendicular to the oncoming air flow (Leutwyler and Dalton [11], 2004)(Goff [9], 2013). In the stable flight phase, the skier has to orient the body and skis to minimize drag and the lift while making a V-shape with the skis. The posture of V-shape (see the photo in figure 1) improves the projected frontal area of the skis. This move increases the lift force generated by the air turbulence around the skier. The quality of the flight can be simulated by CFD (computational fluid dynamics) in order to quantify for the turbulent flow. This occurs around the body of the skier and his equipment. The study of a new posture, the test of different angles of attack and results of airflows can be completed with the support of the simulation. The first step in this study’s is the 3D reconstruction of the skier and the material in stable dynamic air flight. The originality of the methodology developed in this study lies in the coupling of 3D point measurements and a parametric design of the skier (see the complete process in figure 2).
Fig. 1.Aerodynamic forces on a skier with the drag force, the lift force, the velocity, skier and the V-shape of the high level athlete
32
the weight of the
N. Gardan et al. / Computational fluid dynamics for the nordic combined skiing jump
Fig. 2. The process developed to obtain the CFD calculus
The 3D model of the skier is then based on anthropological data and 3D points approaching from measures operated with 8 cameras of the Vicon. The Vicon system has existed for the measure of the kinematic variations which allowed for geometric construction. The 3D points of the stable-flight posture (after an interpolation application of different measure, each skier has been realized 5 tests) has been loaded in the 3D skier parametric model and the model has been adapted to the 3D points to have the right posture. This measure is completed with the equipment. The helmet and the skis were scanned with a 3D laser scanner (Zscan 700 CX) and added to the 3D model of the skier. The position of the ski has been processed in regards to the position of 3D measure (with Vicon cameras) of the feet. Every skier was equipped with active marker (about thirty per skier). Figure 3 shows the different steps of modelization.
Fig.3. High level athlete process modeling
After the integration of the equipment, the 3D model of the skier is meshed with triangle elements (3 nodes) of suitable quality. The influence of mesh quality can be important on the grid density (the grid is the cells on which the flow is solved), on the solution accuracy, and on the CPU time required. The surface mesh processing and grid permit us to build our model.
33
N. Gardan et al. / Computational fluid dynamics for the nordic combined skiing jump
The mesh density should be relevant to capture enough flow features and we must direct attentions to the mesh adjacent to the “wall’ (for example in our case the skier body). Actually the boundary mesh should be fine enough to resolve correctly the boundary layer flow. Some methodologies developed byour team to easily link CAD model to CAE model have been very useful (Danesi et al. [5], 2012), (Gardan et al.[7], 2011), (Gardan and Gardan [8], 2003) and (Jaisson et al. [10], 2012) to obtain a good result (see figure 4). The mesh is then loaded in a Virtual Wind Tunnel and a “negative” mesh is ready (see figure 5) to obtain the grid. Kinematical variables obtained in this study were compared to those obtained by Chardonnens et al., 2012. The figure 6 shows the principal values of angles.
Fig.4. Visualization of the body athlete mesh
Fig.5. Virtual Wind Tunnel mesh
34
N. Gardan et al. / Computational fluid dynamics for the nordic combined skiing jump
Fig. 6. Angle value for posture retrieve with Vicon markers
4. Results and discussion The CAD tools which were used are DAZ Studio© from DAZ 3D for the human 3D modeling and the link with physical markers obtained with the Vicon camera, Catia V5© from Dassault Systems for the integration of the equipment (helmet, ski and suit adaptation). Boots and gloves are not digitized as they are represented with a deformation of the meshing. A more precise representation is planned for the future model optimization. The meshing tool is Hypermesh© from Altair and the CFD software is Acusolve© from Altair. More particularly is the use of the Virtual Wind Tunnel (VWT) tool which is a vertical solution to perform wind tunnel simulations of vehicles (adapted by us for skiing application). Transient and steady state studies can be performed accurately and in a short time frame. An intuitive user interface streamlines the process to define parameters for meshing as well as the CFD case setup. VWT and Acusolve© both are used to perform the CFD analysis. The set of the CFD calculus have been done with three kinds of loads and constraints: - the density is adjusted to 1kg/m3 (match to an altitude close to 2000 meters) - the air flow speed is settled to 26 m/s (amount to 93,6 km/h) - the orientation of ski’s air flow is ruled to 30 degrees (see figure 7).
Fig.7. Air flow orientation in the virtual wind tunnel
The resolution was performed on almost 7 million elements on a classical simulation computer (16 Go of RAM). To develop more precise results a partnership agreement was prepared with ROMEO HPC Center (Delévacq et al. [6], 2013), a platform is hosted by the University of Reims Champagne-Ardenne and funded by the Champagne-Ardenne regional council (2600 CPU cores, 6 TB of memory, 220 TB of storage). A transient calculus to verify the air flow time dependent influence was prepared before a classical steady-state analysis.
35
N. Gardan et al. / Computational fluid dynamics for the nordic combined skiing jump
To compute the above aerodynamic coefficients, the following equations are used: -
Drag coefficient:
-
Lift coefficient:
-
Cross coefficient:
-
Drag area:
with
,
and
forces are acting on the body skier in x, y and z directions, ρ the air density, v the air
flow velocity and the frontal projected area of the skier. The results are classified in two parts: - results relative to the improvement of the skier posture (training influence). - results relative to the design of the underwear (the most important one is the air pressure on the body of the skier). The model presented in this study contains the kinematics of the athletes, modelization of the body geometry, geometry of the ski, geometry of the underwear and the air flow. These different steps required precision for preliminary simulation towards developing a realistic model representing the real experimental data. In this study only the angle of 30° was taken and the results were correlated from the simulation and experimentation. In fact, the number of variables is decreased pending the validation of the model. The result that athletes used is, for instance, stabilization of the drag coefficient during time. This information can help them to work on the body balance during the flight phases. In the next step of this study, the inter-action between the drag coefficient (characterizing the body posture) and velocity for different angles of attack can be observed (Taiar et al.[16], 1999), (Taiar et al.[17], 2000). In future work, the solution that is considered rests on the implementation of a data-processing platform on virtual prototyping is studied. In order to simulate all the stages of clothes industry and to lead the most powerful model of textile specification including the various materials’ laws behavior and their air permeability forthcoming research is needed (Taiar [18], 2014). Computational Fluid Dynamics (CFD) and human modeling will be used for the simulation and will be compared to the experimental data for validation. The principal results are shown in figure 8. This concerns the drag coefficient evolution in the transient analysis, the air flow visualization in figure 9 and the pressure influence in figure 10. Those results are currently used to improve the athletes training behaviors and to design the new underwear in aims to decrease drag during take-off phases.
Fig.8. Drag coefficient evolution
36
N. Gardan et al. / Computational fluid dynamics for the nordic combined skiing jump
Fig. 9. Air flow evolution
Fig.10. Body surface pressure influence
4. Conclusion and future trends The aim of this paper was to present a methodology which allows the study of the stable flight in ski jumping. An innovative approach has been settled using the kinematical Vicon variables and a virtual wind tunnel approach with CFD. Modelling is also based on specific data (anthropological one, for example) that are dedicated to five athletes of the Nordic combined French team. All the different steps of the modelization and simulation have been developed for one athlete and will be applied to other four athletes representing the French team Sotchi Olympic Games (2014). The results of the CFD analysis are relevant to the training part (posture analysis, air flow influence, drag/lift coefficient evolution) and the design of specific underwear (body surface pressure retrieves with CFD is very interesting in this case). For the first tip, the results are currently analyzed by skiers. Several areas of improvement were identified for the next steps of our work. The first one is the necessity to re-mesh the body of the skier in case of posture modification for the CFD analysis. Specific methodologies have been developed to automate many aspects of the process. The second area is a complete integration of the equipment such as additional boots and gloves as well as better material modeling of the suit (for instance the suit is understood by the CFD model like a second skin). Finally the last area is the possibility of using a very fine mesh to obtain more precise results. Alternative way of prospecting is a complete automation between virtual posture modification and associative CFD mesh.
37
N. Gardan et al. / Computational fluid dynamics for the nordic combined skiing jump
References [1] Barelle, C., 2011. Sport Aerodynamics: on the Relevance of Aerodynamic Force Modelling versus Wind Tunnel Testing., in: Colman Lerner, J. (Ed.), Wind Tunnels and Experimental Fluid Dynamics Research. InTech. [2] Chardonnens, J., Favre, J., Cuendet, F., Gremion, G., Aminian, K., 2013. A system to measure the kinematics during the entire ski jump sequence using inertial sensors. J. Biomech. 46, 56–62. [3] Chowdhury, H., Alam, F., Mainwaring, D., 2011. Aerodynamic study of ski jumping suits. Procedia Eng. 13, 376–381. [4] Chowdhury, H., Alam, F., Mainwaring, D., Subic, A., Tate, M., Forster, D., Beneyto-Ferre, J., 2009.Design and methodology for evaluating aerodynamic characteristics of sports textiles. Sports Technol. 2, 81–86. [5] Danesi, F., Gardan, N., Kwassi, E., Reimeringer, M., 2012. Design Process Automation and Optimization using industrial knowledge based Simulation and Design, in: 23rd IASTED International Conference on Modelling and Simulation. Presented at the MS 2012, Banff, Canada, 823–827. [6] Delévacq, A., Delisle, P., Gravel, M., Krajecki, M., 2013. Parallel Ant Colony Optimization on Graphics Processing Units. J. ParallelDistrib. Comput. 73, 52–61. [7] Gardan, N., Gardan, Y., 2003. An application of knowledge based modelling using scripts. Expert Syst. Appl. 25, 555–568. [8] Gardan, Y., Gardan, N., Kwassi, E., Lerouge, A., 2011. A methodology to optimize design: fundamentals and case study, in: Proceedings of the 13th WSEAS International Conference on Mathematical and Computational Methods in Science and Engineering. Presented at the MACMESE’11, Catane, Italie, 30–35. [9] Goff, J.E., 2013. A review of recent research into aerodynamics of sport projectiles. Sports Eng. 16, 137–154. [10] Jaisson, M., Lestriez, P., Taiar, R., Debray, K., 2012. Finite element modeling of TMJ joint disc behavior. Int. Orthod. 10, 66–84. [11] Leutwyler, Z., Dalton, C., 2004. A CFD Study to Analyze the Aerodynamic Torque, Lift, and Drag Forces for a Butterfly Valve in the Mid-Stroke Position 41–49. [12] Müller, W., 2005. The physics of ski jumping, in: European School of High-Energy Physics. 269–278. [13] Reisenberger, E., Meile, W., Brenn, G., Müller, W., 2004. Aerodynamic behaviour of prismatic bodies with sharp and rounded edges. Exp. Fluids 37, 547–558. [14] Schmölzer, B., Müller, W., 2005. Individual flight styles in ski jumping: results obtained during Olympic Games competitions. J. Biomech. 38, 1055–1065. [15] Seo, K., Murakami, M., Yoshida, K., 2004.Optimal flight technique for V-style ski jumping. Sports Eng. 7, 97– 103. [16] Taïar, R, Sagnes P, Henry C, Dufour AB and Rouard AH., 1999. Hydrodynamics optimization in butterfly swimming: position, drag coefficient and performance. J. Biomech, 32(1999): 803-810. [17] Taïar, R, Sagnes P, Henry C, and Rouard AH., 2000. Hydrodynamics optimization in butterfly swimming. J. Biomech, 33:4:509-510. [18] Taiar, R., 2014. Impact of the Smart Textile on the High Level Sport Performance and Patient Behavior in Medicine. Global Journal of Researches in Engineering: A Mechanical and Mechanics Engineering, 14, 8, 22-28. [19] Virmavirta, M., Isolehto, J., Komi, P., Brüggemann, G.-P., Müller, E., Schwameder, H., 2005. Characteristics of the early flight phase in the Olympic ski jumping competition.J.Biomech.38, 2157–2163.
38
