CFD Simulation Results vs. Test data from NASA: Onera M6
The Aerospace industry is one of the most powerful industries worldwide and, as it stands today, supplies 4 basic markets: military aircraft, missiles, space and commercial airliners. Only the commercial aircraft industry, operates more than 20,000 aircrafts at 3,500 airports and is expected to double over the next decade. Considering the global economic state and the more and more strict regulations about flights, aircraft designers have to build more fuel-efficient aircrafts and also make air transportation safer. The only way to accomplish this goal, is by designing planes which generate more lift but produce less drag at the same time. In order to design such aircrafts, it is necessary to have a deep understanding in the aerodynamics of the planes. The most crucial aerodynamic element of an aircraft is the wings. But how do wings actually work?
As an aircraft flies through the sky, the air follows the curvature of the wing surface. Because of this curvature, the air on the top travels faster than the air on the bottom. As a result, the pressure on the top surface is lower (suction surface), and the pressure on the bottom surface is higher (pressure surface). It is this pressure gradient that generates the lift, a force which acts perpendicular to fluid motion, making it possible for aircrafts to fly. Modern wings, are able to generate lift at zero angle of attack, and lift increases as angle of attack increases. The relative motion between air and wing causes friction, and as result a second force acts in a direction which is opposite to the motion of the aircraft. This force is called drag.
In order to find more efficient designs and try different modifications of original wing geometries, aerospace industry has turned to CFD packages with which the fluid flow over a wing can be simulated using computers instead of setting experiments. Computational Fluid Dynamics is the analysis of systems involving fluid flow, by means of computer-based simulation. CFD analysis is a dominant key for aerospace industry, in order to design and optimize an aircraft, and this is because it costs less and gives valuable results faster.
It is much more cost-effective to simulate the flow around an aircraft (or a part of an aircraft such as a wing), in order to get essential engineering data, than using physical experiments. With the rapid development of computers, the cost of simulation is likely to decrease, and with the ability of testing numerous models, and several modifications of the original geometry, the development process of an aircraft can be much more economical.
CFD simulations can be performed in a “short” period of time (depending on the computer available), especially if it is compared with physical experiments, where, in some cases, only setting the experiment can take months. With CFD analysis several models can be tested on a daily basis, making the designing process of an aircraft a lot faster.
CFD analysis is a tool that can actually predict the future. In other words, several “what if” scenarios can be examined and evaluated, judging which design gives the best results, and finally constructing only one prototype, the one with the best performance. By comparing simulation results with experiments, we can be confident that the model represents real physics of the problem, and use it in other cases as well. For instance, the same model can be used, examining different angles of attack and finding the one, which generates the higher lift.
Onera M6 CFD simulation
In this case study, a steady state simulation of the Onera M6 is performed using ANSYS and the results are compared with experimental data obtained from NASA experiment and from a simulation, conducted by NASA. Onera M6 is a well-known wing in aerospace industry, which has been used many times for validating CFD codes. In order to get correct results from the model a mesh with high quality, and the best suitable turbulence model is used. Its simple geometry, can give a deep understanding on how the air flows over a wing, and the pressure distribution on the surfaces.
The experimental data of Pressure Coefficient have been obtained at the seven different measurement positions as it is shown in the picture above. The profile of the wing is symmetrical, which means that a positive angle of attack is necessary in order to generate lift.
The values of interest for this study are the pressure distribution on wing surfaces and calculating the pressure coefficient (Cp). The formula that calculates (Cp) is:
Pressure Coefficient is a dimensionless number that is used a lot from aerospace industry, and the reason is that different wing shapes and geometries can be compared, regardless of their dimensions.
CFD Results vs. Experimental Data
In the figure above, the key areas of interest when examining wing pressure distribution are highlighted. We note that, as the air flows over the top surface, the pressure is decreasing rapidly. This area of wing contributes to 70% of the total lift generated. At a point, the pressure becomes minimum and then starts to recover. Note that on the leading edge Cp equals one.
An overall good agreement between simulation and experimental data is observed. At y/b=0.99 CFD analysis gives less accurate results due to the wingtip vortex that is created in this region of wing. The rotation of the vortices create strong 3-dimensional effects and this is why it is difficult for the model to predict accurately the pressure coefficients in this region. In addition, the diameter of the wingtip vortex is very small making it difficult to be accurately simulated with the present model.
The left picture shows the pressure distribution from a CFD simulation done by NASA for Onera M6. We note that our simulation is at good agreement with NASA. Both analyses predict the lower pressure on the suction surface. Also, the distribution of the pressure, and how it recovers along the wing are very similar.
From the figure below,we note the corellation between pressure and velocity. On the suction surface where the pressure is low, the air is accelarating at values exceeding Mach=1. We also observe on the leading edge, where air is forced to stop moving, that pressure is maximum.
From the last Cp chart (y/b=0.99), we observe the difficulty of the model to predict the pressure coefficient accurately, at the wingtip. In order to get more accurate results, in this region as well, a refinement in this area should be considered.
Generally, a good agreement between experimental data and CFD simulation can be observed. Thus, the air flow over a wing can be simulated with high level of accuracy. From the present analysis we can conclude that, Computational Fluid Dynamics is a tool that can be used with confidence, replacing physical experiments, in order to develop and evaluate new aerodynamics devices, used in aircrafts. Apart from aerospace industry, CFD can be used in other applications as well, such as automotive, hydrodynamics of ships, Internal Combustion Engines, turbomachinery, external and internal environment of buildings: wind loading and heating/ventilation, marine engineering: loads of offshore structures, hydrology and oceanography: flows in rivers and oceans, meteorology: weather prediction…
Written by: Sofoklis Giannaros ([email protected])