Low Reynolds Number Simulation for Flow Over a Flapping Wing: Comparisons to Measurement Data

Abstract

I. Introduction MANNED aerial vehicle (UAV) is an unpiloted aircraft that can be remotely controlled or fly autonomously based on pre-programmed flight plans or more complex dynamic automation systems. UAVs are currently used in a number of military roles, including reconnaissance and attack. They are also used in a small but growing number of civil applications such as firefighting when a human observer would be at risk, police observation of civil disturbances and crime scenes, and reconnaissance support in natural disasters. In this study, the targeted problem is to understand the low-Reynolds number flow characteristics around small UAVs or Micro-Air Vehicles that mimic the flying motion of birds and insects. Low-Reynolds number flow over a moving and flapping foil is selected as a representation for the model problem. The difficulty in the traditional CFD simulation for this type of fluid-structure interaction problems is to incorporate the motions of the flapping foil. The advent of immersed-boundary (IB) methods enables the simulations of this type of fluid-structure interaction problems with reasonable accuracy and low computational costs. In previous studies using the IB methods, changes of surface force and flow characteristics due to the flapping motion have been investigated. Very low Reynolds number (in the 100s) simulations were often used as the representations for the practical situations with higher Reynolds number in the 1000s (Dong et al. 2006, Web et al., 2008). This was due to the limitation of the IB method used in these studies. With a recently developed IB method (Zhang and Zheng 2007), we are able to simulate flow over flapping foils with a wider range of Reynolds numbers to a few thousands. However, from our preliminary results, we found that Reynolds numbers did have effects on surface force and flow characteristics in some of the flapping wing cases even within the Reynolds number range of 100s-1000s, particularly for the cases of heaving and pitching motion. We therefore performed a detailed study for the effects of Reynolds numbers in simulating flapping wing motion within the low Reynolds regime, which is within the interested Reynolds number range of MAV. Reynolds number effects on heaving and pitching motions will be studied separately and then combined. In this Reynolds number range, transitions to turbulent flow can occur in certain regions; however, detailed flow physics related to turbulence effects are not the focus of the current study, which can be either directly simulated or modeled using LES, URANS, or other models. Rather, two-dimensional simulations using the IB method will be compared with available experimental data at different Reynolds numbers to study the merit of two-dimensional, low-Reynolds number simulation for flapping airfoils.