In the present work, we propose an optimization framework based on the active learning method, which aims to quickly determine the conditions of tandem flapping wings for optimal performance in terms of thrust or efficiency. Especially, multi-fidelity Gaussian process regression is used to establish the surrogate model correlating the kinematic parameters of tandem flapping wings and their aerodynamic performances. Moreover, the Bayesian optimization algorithm is employed to select new candidate points and update the surrogate model. With this framework, the parameter space can be explored and exploited adaptively. Two optimization tasks of tandem wings are carried out using this surrogate-based framework by optimizing thrust and propulsion efficiency. The response surfaces predicted from the updated surrogate model present the influence of the flapping frequency, phase, and separation distance on thrust and efficiency. It is found that the time-average thrust of the hind flapping wing increases with the frequency. However, the increase in frequency may lead to a decrease in propulsive efficiency in some circumstances.

Active learning of tandem flapping wings at optimizing propulsion performance

Unsteady aerodynamics of a model bristled wing in rapid acceleration motion

Smallest flying insects commonly have bristled wings and use novel aerodynamic mechanisms to provide flight forces, such as the fling mechanism. In the fling motion, the left and right wings are initially parallel to each other, and then the wings rotate around the trailing edge and "open" to form a V shape. Previous studies lacked the detailed flow around bristles, so the interaction mechanism of the two bristled wings in the fling motion was not well understood. In the present study, we obtained the detailed flow around each bristle numerically and revealed the interaction mechanism of two bristled wings. The results are as follows. During the fling, the vertical force of the bristled wings is similar to that of the corresponding flat-plate wings, but the drag of the bristled wings is much smaller. When the initial distance between wings is small, the opening drag of the bristled wings can be one order of magnitude smaller than that of the flat-plate wings. This is due to the different wing-wing interaction mechanisms of the two types of wings: for the flat-plate wings, during the fling motion, a "cavity" is created between the wings, producing a very large drag on the wings; while for the bristled wings, there are gaps between the bristles and Stokes flows move through the gaps, thus the "cavity" effect is much weaker. Very low "opening" drag may be one of the advantages of using bristled wings for the smallest insects.

The wing-wing interaction mechanism of bristled wing pair in fling motion

The smallest insects fly with bristled wings at very low Reynolds numbers (Re) and use the drag of the wings to provide the weight-supporting force and thrust. Previous studies used two-dimensional (2-D) models to study the aerodynamic force and the detailed flow field of the bristled wings, neglecting the three-dimensional (3-D) effect caused by the finite span. At high Re, the 3-D effect is known to decrease the aerodynamic force on a body, compared with the 2-D case. However, the bristled wing operates at very low Re, for which the 3-D effect is unknown. Here, a 3-D model of the bristled wing is constructed to numerically investigate the detailed flow field and the aerodynamic force of the wing. Our findings are as follows: The 3-D effect at low Re increases the drag of the bristled wing compared with that of the corresponding 2-D wing, which is contrary to that of the high-Re case. The drag increase is limited to the tip region of the bristles and could be explained by the increase of the flow velocity around the tip region. The spanwise length of the drag-increasing region (measuring from the wing tip) is about 0.23 chord length and does not vary as the wing aspect ratio increases. The amount of the drag increment in the tip region does not vary as the wing aspect ratio increases either, leading to the decrease of the drag coefficient with increasing aspect ratio. 

/pdf/aerodynamics-and-three-dimensional-effect-of-a-translating-f35vgc0a.pdf

Aerodynamics and three-dimensional effect of a translating bristled wing at low Reynolds numbers

/pdf/aerodynamics-and-three-dimensional-effect-of-a-translating-1yfx6cap.pdf

Yan Peng Liu

Papers

Unsteady aerodynamics of a model bristled wing in rapid acceleration motion

Aerodynamics of two-dimensional bristled wings in low-Reynolds-number flow