A hybrid gradient-based/metaheuristic method for Eurocode-compliant size, shape and topology optimization of steel structures

The production and processing of building materials is responsible for a significant share global greenhouse gas emissions. Nevertheless, in current practice, lot material wasted, because load-bearing structures are often grossly overdimensioned. Numerical optimization tools have the potential to reduce consumption structural materials, but lack customized algorithms prohibits their use daily design practice. This real-life problems complex, involving both discrete continuous variables as well large numbers code constraints. In literature, such usually solved using metaheuristic (often genetic) algorithms, although gradient-based better suited variables. this article, new hybrid gradient-based/metaheuristic algorithm proposed. It combines methods nested approach, exploiting strengths both: ability handle fast convergence terms applicability usefulness demonstrated two realistic case studies minimize weight structure, taking into account all relevant rules Eurocode 3. Within identical computing time, obtains savings compared conventional genetic algorithm.