Room-level data of Simulated Energy consumption and Ventilation dynamics (RSimEV)
Description
This dataset offers simulated data that includes various parameters impacting energy consumption and ventilation across diverse building scenarios. The simulations encompass various room types within buildings of varying shapes and sizes. Comprising a total of 312 CSV files, each file corresponds to simulations conducted in different rooms within buildings with random parameters. Each building undergoes 200 simulations for a one-month period, with the month randomly chosen to account for different weather conditions. Locations are randomly selected from three regions in the north hemisphere: 1) Dusseldorf, North Rhine-Westphalia, Germany; 2) Tehran, Tehran, Iran; and 3) Brockville, Ontario, Canada, representing three climate zones (mixed, warm, and cold). The simulations yield hourly results, resulting in file sizes ranging from 144,000 (representing 200 simulations over 24 hours for 30 days) to 148,800 data rows (for simulations spanning 31 days). Each CSV file is structured with 55 columns, capturing a comprehensive set of attributes relevant to energy consumption and ventilation dynamics. The collective dataset includes 45,562,639 rows, presenting a robust foundation for in-depth analysis and exploration of the intricacies of building performance across many conditions and configurations. It's essential to note that users are accountable for any risks associated with the dataset's utilization, and the creators explicitly disclaim responsibility for specific applications or outcomes. Detailed information on dataset columns and their units is available in the accompanying "readme.txt" file.
Files
Steps to reproduce
The dataset is generated through Honeybee energy simulations and the Grasshopper plugin for Rhino7 software, with input parameters randomly selected from predefined ranges. This deliberate variability allows for robust machine-learning applications.