Search results

The purpose of the paper is to propose a homogenization-based topology optimization method to optimize the design of variable-density cellular structure, in order to achieve lightweight design and overcome some of the manufacturability issues in additive manufacturing.

First, homogenization is performed to capture the effective mechanical properties of cellular structures through the scaling law as a function their relative density. Second, the scaling law is used directly in the topology optimization algorithm to compute the optimal density distribution for the part being optimized. Third, a new technique is presented to reconstruct the computer-aided design (CAD) model of the optimal variable-density cellular structure. The proposed method is validated by comparing the results obtained through homogenized model, full-scale simulation and experimentally testing the optimized parts after being additive manufactured.

The test examples demonstrate that the homogenization-based method is efficient, accurate and is able to produce manufacturable designs.

The optimized designs in our examples also show significant increase in stiffness and strength when compared to the original designs with identical overall weight.

The purpose of this paper is to propose an improved proportional topology optimization (IPTO) algorithm for tackling the stress-constrained minimum volume optimization problem, which can meet the requirements that are to get rid of the problems of numerical derivation and sensitivity calculation involved in the process of obtaining sensitivity information and overcome the drawbacks of the original proportional topology optimization (PTO) algorithm.

The IPTO algorithm is designed by using the new target material volume update scheme and the new density variable update scheme and by introducing the improved density filter (considering the weighting function based on the Gaussian distribution) and Heaviside-type projection operator on the basis of the PTO algorithm. The effectiveness of the IPTO algorithm is demonstrated by solving the stress-constrained minimum volume optimization problems for two numerical examples and being compared with the PTO algorithm.

The results of this paper show that the uses of the proposed strategies contribute to improving the optimized results and the performance (such as the ability to obtain accurate solutions, robustness and convergence speed) of the IPTO algorithm. Compared with the PTO algorithm, the IPTO algorithm has the advantages of fast convergence speed, enhancing the ability to obtain accurate solutions and improving the optimized results.

This paper achieved the author’s intended purpose and provided a new idea for solving the stress-constrained optimization problem under the premise of avoiding obtaining sensitivity information.