Experimental Investigation on Aerodynamic Performance of Gliding Butterflies
Ye Hu,Jinjun Wang
DOI: https://doi.org/10.2514/1.45156
2010-01-01
Abstract:R ECENTLY, the development in micro technology has accelerated the research on micro air vehicles (MAVs). In particular, insectlike MAVs have attracted considerable attention. The past investigations demonstrated that a flying insect not only generates sufficient lift, but also controls the flight with great precision [1,2]. According towing-beat frequency and wingspan length, insect flight can be sorted into two categories. The first category is the flight with high wing-beat frequency epitomized by insects with small-scale wingspans such as bees and flies, and the other is the flight with low wing-beat frequency exemplified by insects with large-scale wingspans such as butterflies and hawkmoths. Some dimensionless parameters of bees and butterflies are presented to elucidate the different categories. By contrast with the Reynolds number of 2000 for bees, the Reynolds number of butterflies is nearly 10 based on the definition for a flapping wing and 2 10 for a fixed wing. The span-based reduced frequency of bees is seven times higher than that of butterflies and the aspect ratio is five times higher. Therefore, the investigation on butterfly flight has attracted great attention in the last century, and plentiful kinematic parameters (such as flapping angle and body angle) have been revealed [3–5]. To clarify the high-lift mechanism of butterfly flight, Srygley and Thomas [6] investigated the flow structure over a free-flying Red Admiral butterfly inawind tunnelbysmoke–wireflowvisualizations. The results indicated that some aerodynamic mechanisms including wake capture, two different types of leading-edge vortex, and active and inactiveupstrokeswere applied togeneratemore lift in successive strokes when the butterfly took off, maneuvered, and landed. However, these mechanisms were avoided in a steady forward flight. Senda et al. [7] calculated the aerodynamic force of the butterfly, which was considered as a rigid-multibody system by Lagrangian method.Meanwhile, theyalsomeasured theaerodynamic forcewitha micro load cell to substantiate the validity of their mathematical model. Up to now, the flow structures of butterflies have not been attached such importance to as their flight motions. However, it is found that when a butterfly flies forward in a windless stream, it always flaps wings for several times and then keeps themwith a constant dihedral angle to glide for a distance steadily. The flight combining flapping and glidingmotions improves the flight efficiency greatly. In fact, the glidingmotion does not occur in the flight of a small-scale insect, and it is featured profoundly in a butterfly flight, even more so than the flapping motion. The flow structure of gliding motion may be the primary aerodynamic mechanism for migratory butterflies. Therefore, research on the gliding flight of butterflies may shed some light on the flight mechanism of large-scale flying insects. In addition, the flapping-wingMAVs already succeeding in flying automatically or under control share in the similar Reynolds number with large-scale butterflies. If the gliding motion is applied in the flight of these MAVs, their flight efficiency and endurance will be enhanced greatly. But the glidingmotion of natural butterfliesmay be not epitomized easily into only onewing position, because they glide with different dihedral angles, depending on ambient conditions and their behavioral requirements. Furthermore, the butterflylike MAVs are different from the natural butterflies inwingflexibility, configuration, etc. Therefore, we would like to investigate the gliding flight of butterflies in consideration of attack angle, sweep-forward angle, and dihedral angle, together in the reasonable range of Reynolds numbers. The optimum position of wings generating higher lift and lower drag will be applied in the flight of butterflylike MAVs. However, we used a simplified butterfly model in the preliminary work. Because butterflies take advantage of theglidingmotion in awindless streamwith low turbulence intensity,measurements in a conventional steady water channel or wind tunnel can be used to investigate the flow structure and aerodynamic characteristic of gliding motion. Hu et al. [8] used hydrogen bubble visualizations to observe the flow structures over a simplified model of Papilio ulysses (one species of migratory butterfly) at Re 9 10, and the variation of flow structures with angle of attack was investigated in detail. A new type of leading-edge vortex, shown in Fig. 1a, was observed that might provide high lift for the weight support of butterflies in a gliding flight. Further effort was made to explore the effect of wing shape on the flow structures and then led to the discovery that leading-edge radius could affect the shape and the strength of leading-edge vortex [9]. As shown in Fig. 1, the vortex core of Danaus plexippuswas straighter and more concentrated. The breakdown location of Danaus plexippus was more downstream than that ofPapilio ulysses. These results indicated thatDanaus plexippus had a stronger leading-edge vortex, which should generate higher lift. Moreover, Danaus plexippus has weaker wing-tip vortex than Papilio ulysses, which might lead to lower drag coefficient. In this Note, force measurement and oil flow visualization are carried out in a wind tunnel to study the aerodynamic characteristics and flow structures of two species of migratory butterfly, Papilio ulysses andDanaus plexippus. In nature, Papilio ulysses has a larger wingspan length (150 mm) and a shorter migration distance than Danaus plexippus (100 mm). By comparing the aerodynamic characteristics of the two experimental models of the insects, the lift-todrag ratio ofDanaus plexippus is twice as large asPapilio ulysses at a moderate angle of attack. Furthermore, the relationships between aerodynamic characteristics and flow structures are discussed.