機械学習技術を用いた超音速翼の大域的トポロジー最適設計

抄録

The supersonic transport (SST) of Concorde has ended its commercial flight at 2003. One of the reasons is that shock waves generated at supersonic flow conditions increase the wave drag acting on SST which results in the deterioration in fuel efficiency. The reduction of the wave drag is required for the next generation SST, so that various conceptual designs have been proposed and investigated. In recent years, topology optimizations have attracted attention as advanced optimization methods with higher degree of freedom to represent various topologies. These methods can automatically propose an innovative/optimal topology which achieves the highest performance. Therefore, it can be an efficient tool to extract innovative design insights. So far, we have proposed a global optimization method for the topology optimization. In this study, we consider the advancement of the search strategy by dynamically changing the parameter of search radius which controls the balance between global/local search, from the latest histories of the objective function. Two analytic functions are minimized by the proposed strategy, and then it can be confirmed that the performance is improved in several conditions.