CFD airflow distribution datasets for ANN prediction in naturally ventilated spaces
Simulation195 MB
CFD-generated flow-field variables at probe locations in naturally ventilated spaces for combinations of reference wind speed and direction, sampled with adaptive and Latin hypercube sampling. Useful for training and testing ML surrogates of indoor air distribution.
CC BY 4.0Free to use and share, including commercially, as long as you credit the creators.
Size
195 MB · 9 files
How to cite
Tajuddeen, I., & Grygierek, K. (2026). Airflow Distribution Datasets for ANN Training Using Adaptive Sampling and Latin Hypercube Sampling in Naturally Ventilated Spaces [Data set]. Zenodo. https://doi.org/10.5281/zenodo.22712734
Check the repository for the citation its authors prefer.
193 steady RANS (k–ε, Boussinesq) simulations of one 8.8 × 6.1 × 3.2 m furnished room where only the positions of three supply and three return ceiling vents change. Each case gives cell-centre temperature, velocity and pressure plus boundary-patch data. Useful for training neural-operator surrogates and for studying how diffuser layout affects HVAC airflow.
Processed URANS outputs (regional temperature rise, passive tracer, mean-field sections), surrogate models, reference geometry, input dictionaries and analysis code comparing fan allocations at equal power in a normalised industrial hall. Useful for industrial ventilation design and surrogate modelling.
Input case list for CFD runs, ML inputs and outputs, a trained neural network and an inference script for predicting room ventilation conditions, including humidity. Useful as a ventilation surrogate-modelling example.
A field dataset from heated model enclosures that combine roof radiative cooling with temperature-driven ventilation, about 7 air changes per hour. Includes temperatures, analysis code and background-oriented schlieren flow visualisations. Useful for studying passive stack ventilation and the indoor thermal balance.
Data from CFD simulations of indoor airflow and passive tracer spread on the 8th floor of Pyeongtaek St. Mary's Hospital, in six cases with different outdoor wind directions and mechanical ventilation settings. Useful for studying airborne spread between hospital rooms.
From Zenodo, added 2026-11-15.
Listed there as “Airflow Distribution Datasets for ANN Training Using Adaptive Sampling and Latin Hypercube Sampling in Naturally Ventilated Spaces”. Tags proposed by AI and not yet checked by hand. Spotted a mistake? Tell the portal contacts.