A robust node-shifting method for shape optimization of irregular gridshell structures

Abstract

Evolutionary node shifting is an effective approach to the shape optimization of spatial structures for excellent mechanical performance. In this paper, a novel shape optimization algorithm is proposed, which is applicable to complex gridshell structures of irregular geometries and non-uniform grids. With the objective of maximizing the structural stiffness, the nodal coordinates are iteratively updated according to the sensitivity information. The perturbation displacements referring to the inverse hanging method are applied to the initial flat model. By analysis, we found that the irregular grids give rise to non-smooth gradient fields, which results in jagged surfaces. To normalize the non-smooth gradient fields and to prevent the optimization process from falling into local optima, a double filter scheme is introduced in the process of optimization. A variety of examples are presented to demonstrate that the proposed algorithm can effectively solve shape optimization problems of general gridshell structures.