DesignSafe resources, especially the Data Depot publication pipeline, have proven invaluable to wind researchers. Scholars out of the University of Florida and Clarkson University (Tian et al. 2020) were able to reuse experimental data from “Upwind Terrain Effects on Low-Rise Building Pressure Loading Observed in the Boundary Layer Wind Tunnel'' (Fernández-Cabán and Masters 2018) that was published in the Data Depot. The researchers used the data to train a deep neural network (DNN) to predict mean and peak wind pressure coefficients on the surface of a scale model low-rise, gable roof building. Designed to complement existing experimental methods, including those used in the original study, this DNN approach could more efficiently procure wind-related data with less required testing. This research was initiated through a dialogue between the data creators and a colleague with expertise in machine learning/AI. As a result, the data creators discovered an opportunity for collaboration within the NHERI community and found a way to reuse their data in machine learning, a use they had not envisioned when they originally collected the data.
Tian, J., Gurley, K.R., Diaz, M.T., Fernández-Cabán, P.L., Masters, F.J., and Fang, R. (2020) “Low-Rise Gable Roof Buildings Pressure Prediction using Deep Neural Networks.” Journal of Wind Engineering and Industrial Aerodynamics, 196, 104026, https://doi.org/10.1016/j.jweia.2019.104026.
Fernández-Cabán, P. and Masters, F. (2018) “Upwind Terrain Effects on Low-Rise Building Pressure Loading Observed in the Boundary Layer Wind Tunnel [Data set]. Designsafe-CI. https://doi.org/10.17603/DS2W670.