NUMERICAL SIMULATION OF ICE ACCRETION ON AIRFOILS
Yihua Cao,James P. Denier,Qiang Zhang,Matthew Finn,Trent W. Mattner,John Sheridan
2008-01-01
Abstract:Summary An approach to numerically simulate ice accretion on airfoils has been developed and used to characterize ice accretion on a NACA0012 airfoil. In this paper we present the background and motivation for the investigation, the techniques used, and then the main results are presented. The validation of the simulation approach developed in this paper is also discussed. Finally, the conclusions are presented based on our analysis of the results presented. BACKGROUND INTRODUCTION Ice accretion on airplane wings can be a hazard to flight safety. The cause of the accretion is due to supercooled water droplets impinging to the windward face of the wings. Ice accretion can modify the designed aerodynamic shape of airfoils and considerably degrade their aerodynamic performance. Thus, being able to evaluate the mechanisms and consequences of ice accretion is of great importance to anti-icing and de-icing. Different icing conditions form different type of ice accretions. Depending on the icing mechanism, ice accretions can be classified as: rime ice, glaze ice or mixed ice. Rime ice occurs in a low temperature environment and forms because supercooled droplets freeze simultaneously when they impinge onto the airfoil surface. Glaze ice occurs in an environment with relatively high temperature (while still lower than freezing point) under which the droplets freeze partially at the impingement location and then freeze gradually during the flow along the airfoil surface caused by airflow. Mixed ice is defined as a mixture of rime and glaze ice. Glaze ice and mixed ice can corrupt the designed aerodynamic shape of airfoils more significantly than rime ice. These forms of ice accretion can be investigated by several means, including flight test, experimental simulation, engineering method and numerical simulation. Flight test and experimental simulation can obtain exact ice shape but are usually too expensive to be widely adopted. The engineering method uses the typical experimental data and empirical formulae but could hardly analyze the ice accretion process. Therefore, numerical simulation is widely adopted because it is economical and can simulate the icing process and so provide a relatively exact evaluation of ice accretion. Several codes for simulating ice accretion have been developed internationally, such as: LEWICE (USA), ONERA (France), DRA (UK), FENSAP-ICE (Canada), CIRAMIL (Italy). TECHNIQUES DESCRIPTION This paper develops an approach to numerically simulate ice accretion on airfoils which is based on four modules: (1) air flowfield solution, (2) droplets collection efficiency calculation, (3) boundary layer characteristics evaluation, and (4) ice accumulation evaluation via a thermodynamic model. Air flowfield solution The air flowfield can be obtained by using the panel method to solve the Euler equations or directly solving the compressible Navier-Stokes equations. The panel method can calculate the air velocity at any point in the flowfield directly but usually lacks accuracy in computing the complete flowfield. In contrast, the solution of the compressible Navier-Stokes equations can provide a more accurate flowfield computation but is time consuming. In order to synthetically consider the computational precision and efficiency, an Euler flow computation is adopted in this paper. Droplets collection efficiency calculation The droplets’ collection efficiency on the airfoil surface is important in numerically simulating ice accretion and two computational methods are available: Lagrangian and Eulerian two-phase flows methods. The Lagrangian method obtains the collection efficiency by solving the motion equation of droplets to track each droplet’s trajectory in the flowfield. Eulerian two-phase flow methods consider the droplets in the airflow as a form of pseudo fluid which interpenetrates with the air and the collection efficiency is obtained through solving the velocity and apparent density distribution of droplets. There are advantages in using the Eulerian two-phase flow method compared with a Lagrangian approach since the same mesh can be used to solve the governing equations for the airflow and droplets. Also, the droplets’ collection efficiency may be obtained based on the solution of the droplets’ flowfield directly, meaning that particles don’t have to be tracked. For these reasons the Eulerian two-phase flows method has been adopted here.