A gradient-based optimization method with functional principal component analysis for efficient structural topology optimization

摘要: Structural topology optimization (STO) is usually treated as a constrained minimization problem, which iteratively addressed by solving the equilibrium equations for problem under consideration. To reduce computational effort, several reduced basis approaches that solve in space have been proposed. In this work, we apply functional principal component analysis (FPCA) to generate basis, and couple FPCA with gradient-based method first time literature. The proposed algorithm has tested on large STO 4.8 million degrees of freedom. Results show achieves significant savings negligible loss accuracy. Indeed, density maps obtained capture larger features without but significantly lower times, are associated similar values minimized compliance.