Combined Parallel Computing, Dynamic Mesh and Heuristic Algorithms in Tapered Wings

Abstract

In this study, Onera M6 wing has been optimized by two parameters, the wing section and the taper ratio by combining recent mostly preferable approaches i.e. parallel computing and evolutionary techniques. For the 3-D models developed during the optimization stages, the mesh required generated by dynamic mesh technique. The code developed for this aim is robust and faster than the codes, which are only producing mesh by classical techniques. An Euler flow solver is used to obtain the flow parameters for each member. Because the operating time of the code is very long, on account of our low capacity computer resources, parallel processing has been used. From the results, it is observed that the optimization process is working as expected. During the optimization process the lift coefficient and the thickness ratio are tried to be close to the design values determined at the beginning. The taper ratio is getting smaller while the code is trying to minimize the drag force. But it cannot be reduced to very small values and is kept almost the same at later steps, since the aim is that the drag force should not only be reduced but also the lift force be held close to the design value.