PRECISE CONTROL OF MINIMUM LENGTH SCALE IN TOPOLOGY OPTIMIZATION BASED ON SKELETON EXTRACTION

Due to the constraints of manufacturability, current topology optimization technologies are mostly used only in the conceptual design of structures. Therefore, it is necessary to research the method of topology optimization that is directly facing manufacturing. Based on the heuristic BESO (Bi-directional Evolutionary Structural Optimization) algorithm, this paper proposes an efficient topology optimization method that can precisely control the minimum structural size. Through sensitivity interpolations and refinement of boundary elements, current BESO algorithm is improved and the problem of non-smooth boundary is solved. The skeleton configuration of the structural topology is extracted by using the topology refinement method. On this basis, the structural members whose size violate the minimum size constraint can be determined, and thusly the minimum length scale of the topology structure is precisely controlled based on the improved BESO algorithm. Additionally, to avoid the premature problem, a relaxed constraint method is adopted, in which the minimum size constraint is gradually strengthened during the optimization process. Numerical examples demonstrate the effectiveness of the proposed topology optimization method.