Self-adjusted convex approximation method for structural optimization

This study presents an approximation method called self-adjusted convex approximation for optimum design of structures. The proposed method is a generalization of the convex approximation method. The order of the approximation for each constraint is automatically adjusted in the optimization process. This self-adjusted capability makes the approximate constraint values conservative enough to maintain the optimum point of the approximate problem in the feasible region. This feature can prevent oscillation of the optimum design point and increase the convergence rate of the redesign process. Results obtained from three examples demonstrate the effectiveness of the proposed method.