Numerical And Experimental Investigation Of Flow And Heat Transfer On The Iced Airfoil

Ice accretion on airfoil can cause changes in geometric characteristics of the airfoil.Especially ice horn can lead to a flow separation,drop in the stall angle and increase of drag consequently.Thus,it is important to predict local ice growth to prevent degrade of aerodynamic performance.This study focuses on the solution of heat transfer coefficient and velocity field for a 2D iced airfoil.It has been found out that Detached Eddy Simulation(DES)is effective for predicting the aerodynamic flow variables.All of the computations have been carried out by means of the commercial FLUENT solver.To utilize DES model properly,grid spacing and time step are required to be selected considering the size of eddies resolved by LES modes.The grid is determined for each model after mesh independency tests.A new time stepping method is used to solve the problem due to small domain.The DES methodology is first validated with M 5-6 airfoil.While the differences with experimental data are generally lower than the results for previous simulation,these error values are acceptable in the validation.The wind tunnel tests are carried out to make sure that results from the simulations are reasonable.The measurement method is devised from the Newton’s cooling law.There are a few assumptions that ice do not further grow at the moment and the effect of conduction and radiation are ignored.The thermocouples are fixed on the airfoil surface and the surface is then covered with heater sheet.The electrical energy is then supplied to the different airfoils made of acrylic.The experiments are repeated with the three different air speed conditions for the higher accuracy.When flow reaches steady state,the values of voltage,current and initial & final temperatures on the surface are recorded and the values of heat transfer coefficient are calculated using these result values.The comparison of experimental data is made between the numerical results for RANS and DES.As for clean airfoil,the results for both DES and RANS are identical with each other.The experimental result seems to make sense in its tendency and quantity.The faster the air flow velocity,the greater heat transfer coefficient observed over the whole experiments.The results for mixed ice show it is still smaller than the experimental result and a little disagreement is seen due to ambiguity of its geometric shape on the pressure surface.For some glaze ice with low horn angle,there is less difference in DES and RANS besides the lower surface,where the values of RANS are far smaller than DES.In regard of airfoils with large horn in front of leading edge,DES fits better with the experiment regarding tendency.Both the DES and RANS predict the highest value on the tip of ice horn,which is larger than other points on the surface by roughly 50%.On the other hand,a sharp decrease appears in heat transfer coefficient along the pressure surface in RANS.The flow pattern and turbulence layer are investigated to look into what makes DES and RANS have different results.Comparison of lift and drag coefficients are also attained based on horn angle and height.The results for RANS do not reflect the effect of angle and reach the normal range of values.The DES shows that horns over a certain length greatly affect back flow.Finally,the cause of discrepancy of experiment is analyzed following a method of Error Analysis.However,fundamental problem exists in a basic concept: turbulence is a natural event in three-dimensional space.Thus,advanced flow analysis needs to be conducted in 3D to predict accurately heat transfer on iced airfoil.This study will be followed up in a subsequent research.