AI WIND: Kunstig intelligens for håndtering av vind i urbane områder og responsiv design

Offentlig prosjektsammendrag

In the rapidly urbanizing world, urban and architectural design pose multifaceted challenges. Traditionally, CFD simulations have been the primary tool for such analysis. However, these methods are computationally intensive, expensive, and time-consuming, hindering rapid design iteration and slowing progress. By overcoming the limitations of current practices, NablaFlow will develop an AI-driven tool based on low cost to serve (50% reduction in computational time and cost) and close to full-service delivery (at least 90% accuracy). This balance will be achieved by developing a Deep Learning-based surrogate model trained on a dataset of CFD simulations.