Rapid pedestrian-level wind field prediction for early-stage design using Pareto-optimized convolutional neural networks

Abstract

Traditional computational fluid dynamics (CFD) methods used for wind field prediction can be time-consuming, limiting architectural creativity in the early-stage design process. Deep learning models have the potential to significantly speed up wind field prediction. This work introduces a convolutional neural network (CNN) approach based on the U-Net architecture, to rapidly predict wind in simplified urban environments, representative of early-stage design. The process of generating a wind field prediction at pedestrian level is reformulated from a 3D CFD simulation into a 2D image-to-image translation task, using the projected building heights as input. Testing on standard consumer hardware shows that our model can efficiently predict wind velocities in urban settings in less than 1 ms. Further tests on different configurations of the model, combined with a Pareto front analysis, helped identify the trade-off between accuracy and computational efficiency. The fastest configuration is close to seven times faster, while having a relative loss, which is 1.8 times higher than the most accurate configuration. This CNN-based approach provides a fast and efficient method for pedestrian wind comfort (PWC) analysis, potentially aiding in more efficient urban design processes.