Momentum method powered by swarm approaches for topology optimization

Abstract

This paper presents the momentum algorithm for topology optimization problems. Since topology optimization problems are multimodal, the momentum may be trapped into a local optimum. To achieve a better solution, this approach is combined with an enhanced particle swarm optimization (PSOG). The constrained problems are converted into unconstrained ones by an external penalty function. To illustrate the effectiveness of the proposed method, five examples with static, multiple and self-weight loadings are examined. The problems are investigated by the proposed method and the results are compared with optimality criteria algorithm (OC), method of moving asymptotes (MMA), sequential linear programming (SLP), momentum, PSO and PSOG. The numerical results show the superiority of the proposed approach.