On the Relationship Between Design Conditions in Topology Optimization and the Diversity of the Obtained Solutions

Abstract

Designers decide the structure and mechanism of the design target in a rough level, i.e., the design concept, in the conceptual design. Because design targets and demands become more complex and various recently, it becomes difficult to make a decision appropriately in the conceptual design. Thus, our ultimate goal is to propose a methodology for supporting the decision making in the conceptual design. To achieve this goal, it will be promising to generate various and appropriate design solutions using high performance computers and find beneficial design concepts in the generated design solutions using data mining methods. Because topology optimization is suitable to generate various and appropriate design solutions due to its flexibility, we investigate a potential of topology optimization for supporting the conceptual design in this paper. More precisely, we generate various design solutions of the bridge structure with topology optimization while changing design conditions, and find design concepts of the bridge structure in the generated design solutions. Through numerical examples, we demonstrate that beneficial design concepts are obtained using topology optimization.