The subject of unsteady aerofoils has received considerable interest over the past century. In the early 1930s, such flows were of interest to aerodynamicists who were primarily concerned with gust response and
flutter analysis of aircraft. However, with Micro Aviation Vehicles (MAVs) becoming a distinct possibility in the near future, researchers have revisited this field in the hope of shedding some light on the mystery of insect flight. From an aerodynamics perspective, the ultimate goal would be to understand and effectively apply the techniques employed by these tiny winged creatures to augment lift and thrust production.
In our research in the Mechanical Engineering Department, we have been conducting three-dimensional experiments to simulate the wing motion of flying insects to better understand the highly unsteady aerodynamics of flapping flight, as part of a MAV research program spearheaded by the Department.
The 3-D flapping mechanism consists of two wings (see Figure 1), each of which is driven by three computer-controlled stepper motors through gearboxes, timing belts and coaxial drive shafts. The wings can be individually activated if necessary to generate non-symmetric manoeuvres such as those involved in turning flight. To measure aerodynamic forces, a force/moment sensor is installed at the base of one of the wings. The sensor is able to detect forces along two axes and moments along three axes. Experiments were conducted with a pair of scaled Hawkmoth-shaped wings mimicking the hovering wing motion of insects. A record of the force history for simple harmonic flapping in the horizontal stroke plane is displayed in Figure 2.
In parallel with and in support of the experimental programme, we have also been developing a computational model based on a hybrid Cartesian-meshfree A rbitrary Lagrangian-Eulerian ( ALE ) methodology. The model is designed for simulatin g for flows driven by multiple independent moving bodies, including propulsion by deforming bodies as in swimming. The scheme does not require remeshing of the computational domain as do conventional finite-element and finite-volume schemes. It is also expected to have better boundary resolution, due to use of body-fitting grid, than the typical Cartesian-based immersed boundary method. Preliminary simulations of fore and hind wing interaction are shown in Figure 3. The code will be useful in situations where experimental measurements cannot be easily made.
The research team comprises Assoc Prof TT Lim and KS Yeo, and Dr KB Lua .
