トポロジー最適解のSelf-Organizing Mapを用いた探索：第1報：SOM分類結果の検証

Abstract

In recent years, there has been a renewal of interest in the study of supersonic transport (SST). The high speed, economic viability and environmental compatibility are required for the future SST. So, the wing of SST has to be thin as possible in order to minimize the weight and enhance performances. For this reason, the design of lugs between the main wing and body is very important. In structural optimization of lug, topology optimization has been used to determine the optimal structural lay-out. However, topology optimization has the problem that there are several different solutions depending on the analysis data and multiple loads caused by aeroelastic forces on the wing. In this study, in order to search for an optimal solution from topology optimized solutions for several loads, these solutions are classified according to the topological characteristics by using Self-Organizing Map (SOM). In this case, topological characteristics were represented by using model's spatial densities of topology optimized solutions. Through this classification, it is possible to explore to the optimal topology optimized solution within various topology optimized solutions.