Abstract
Truss-like cellular structures have great potential to be applied in light-weight design applications. However, determining the appropriate designs for these truss-like cellular structures can be a challenging task due to theft geometric complexities and prohibitive computational costs in the design process. In this research, a new design method is proposed which can drastically reduce computational costs and design parameters, while maintaining the performance of the targeted outcome. Furthermore, the proposed method facilitates cellular structure designs that can handle multiple loading conditions. The proposed method utilizes the relative density information obtained from a solid topology optimization to automatically determine the diameter of each individual strut in the structure, which collectively represent the set of design variables This allows the method to produce lattice structures that can perform reliably under multiple loading conditions and also reduce the computational cost associated with the design of these structures. The efficacy of the developed method is compared to existing methods including the size matching and scaling method that combines solid-body analysis and a predefined unit-cell library. (C) 2015 Elsevier Ltd. All rights reserved.