AI Modeling of Expert Preferences in Freeform Structural Optimization

The building and construction industry is a significant source of greenhouse gas emissions. The quantities of structural materials used globally correspond to approx. 10% of the annual carbon emissions. Reducing the embodied carbon of new construction is needed, especially considering the predicted population growth and urbanization. Recommended strategies include using more environmentally friendly materials and/or structural optimization. Topology optimization is a promising design method in this context since the resulting designs often significantly outperform conventional lowweight designs. It is a computational approach that generates efficient material layouts tailored to a user’s specific design requirements. To take full advantage of its exploratory power, topology optimization leaves the user as a passive observer who initiates the design process and assesses the quality of the design upon completion. The resulting structural designs are typically high-performing and have high levels of geometric complexity. However, ensuring the physical performance is adequately predicted by a fully automated design approach requires including all relevant operating conditions, mechanical behaviors, and fabrication constraints. This talk will discuss different strategies to include relevant considerations for topology-optimized design of low-carbon civil structures, including methods that use AI to leverage human designers’ experiences.