Abstract
Wind tunnel test of full-scale MIRA model was carried out in Aerodynamic and Aero-acoustic Wind Tunnel of Shanghai Automotive Wind Tunnel Center(SAWTC), and aerodynamic forces as well as surface pressure distribution were obtained as the baseline for data comparison with simulation results.The MIRA geometry model is totally identical for test and calculation. Different mesh schemes were compared and the most efficient grid-independent hybrid mesh scheme was determined. With wind tunnel test data, five different turbulence models, standard k-epsilon, RNG k-epsilon, realizable k-epsilon, standard k-Omegaand SST k-Omega, were compared and discussed in terms of drag and lift forces, as well as the surface pressure distribution on MIRA model.
Objective
The objective of this study is, through comparison of test data and simulation results of aerodynamic forces and surface pressure distribution for full-scale MIRA reference car, to determine the suitable application scope of two-equation turbulence model in numerical simulation of external flow field around three-dimensional vehicles.
Methodology
Wind Tunnel test of full-scale MIRA model was carried out inAAWTof SAWTC. In the test, aerodynamic forceswere measuredthrough an exquisite six-component balance with repeatability of 0.01% and accuracy of 0.05%, and surface pressure was obtained through 254 pressure tapping holes distributed all over the MIRA model. In addition, pressure scannerswere employed for surface pressure data collecting. Five different two-equation turbulence models were utilized to calculate the external flow field around MIRA model, andthe simulation results were comparatively analyzed with test datafor both aerodynamic forces and surface pressure distribution.
Results
This work shows that, compared with SKE, RNG, SKW and SST turbulence models,the RKE two-equation approach gives a better solutionof calculating the flow field around MIRA modelfor steady-state analysis, with respect to the efficiency of iteration convergence, the prediction accuracy of aerodynamic forcesand the surface pressure distribution. However, there are some regions of the MIRA surface where the pressure prediction was not in good agreement with test data. Therefore, to achieve a higher-precision prediction of the flow field around ground vehicles there is still much work remains to be done, such as modification of the turbulence model in certain computational regions.
What does the paper offer that is new in the field including in comparison to other work?
Through comparison of surface pressure distribution on MIRA model between the wind tunnel test dataand the numerical simulationresults, the specific regionsof high-accuracy prediction and low-accuracy prediction wereconfirmed, which could enablethe improvement of turbulence model applied in some localized computational domainsto achievehigher-accuracy calculation.
Limitations
Only the hybrid mesh method was considered in this paper, while the tetra mesh and hexa mesh were not.
Conclusions
Full-scale wind tunnel test was carried out for MIRA reference car, and aerodynamic forces and surface pressure distribution were taken as the baselinefor data comparison with simulation results.The most suitable two-equation turbulence model was validated as realizable k-epsilon approach, and the effective and reasonable application scope of two-equation turbulence model in calculation of the external flow around vehicles was also confirmed, which could lay the foundation for the next-step improvement of turbulence model applied in some localized computational domains.
Key words : MIRA model;turbulencemodel;wind tunnel test