| Title |
Structural Topology Optimization Algorithm using Accelerating Method of Design Variables |
| Authors |
Lee Dong-Kyu ; Shin Soo-Mi |
| Keywords |
Topology Optimization Algorithm ; Convergence ; Design Variable ; Accelerating Method ; Moved and Regularized Heaviside Function ; Density Distribution Method |
| Abstract |
This study proposes an accelerating method of design variables in order to improve convergence of conventional topology optimization algorithm. The accelerating method utilizes a ‘moved and regularized Heaviside function’ and improves moving velocities of the design variable values. This function is defined as the combination of a concave and convex function in domain between 0 and 1. Therefore design variable values less than 0.5 move fast toward the value of 0 and those more than 0.5 become rapidly close to the value of 1. The accelerating effects can be various according to the nature of the concave and convex of the function. In addition, the design variable values which become renewed by the accelerating method are able to be speeded up by repetitions of the accelerating effect. The accelerating effects of design variables are verified through numerical examples of linear elastostatic structures in topology optimization algorithm of a density distribution method. |