Adaptive numerical simulations of insect flight using wavelet techniques

We present a wavelet-based adaptive approach to compute the aerodynamics of flapping insects. Dynamically evolving grids using regular Cartesian blocks allow significant reduction of memory and CPU time requirements while monitoring the precision of the computation. Distributing the blocks among MPI processes permits an efficient parallelization on large scale supercomputers. The numerical approximation uses artificial compressibility to avoid solving elliptic problems and volume penalization is applied to impose boundary conditions on the Cartesian grids. A centered 4th-order finite difference discretization is combined with biorthogonal interpolating wavelets as grid refinement indicators. Different validation cases are presented to assess the accuracy and performance of the open access code WABBIT on massively parallel computer architectures (Engels et al., Commun. Comput. Phys., doi:10.4208/cicp.OA-2020-0246). Flow simulations of flapping insects demonstrate its applicability to complex, bio-inspired problems. First computations using realistic fly bodies obtained from mirco-CT scans are likewise presented.