Curated Brazilian Energy and Socio-Environmental Dataset
Description
This curated dataset compiles publicly available, multi-domain data capturing Brazil’s energy system and socio-environmental context. The raw data are not owned or produced by the author(s) but are aggregated from the following open sources:
-
Electricity and Grid: Load, solar PV, wind generation, and grid characteristics (line length, capacity, efficiency) from the Open Brazilian Energy Data repository.
-
Weather: Air temperature, pressure, and rainfall from INMET, accessed via Kaggle.
-
Socio-Economic Indicators: State-level GDP and GDP per capita from the Brazilian Institute of Geography and Statistics (IBGE).
-
Population Estimates: State-level population from IBGE.
The dataset was processed and curated specifically for two studies on spatio-temporal graph neural network–based load forecasting. Preprocessing steps tailored to each study are detailed in the following papers:
- Enhanced Load Forecasting with GAT-LSTM: Leveraging Grid and Temporal Features
- Grid-Aware Spatio-Temporal Graph Neural Networks for Multi-Horizon Load Forecasting
Files
grid_df.csv
Additional details
Software
- Repository URL
- https://github.com/ugoorji12/Grid-Aware_STGNN_for_Multi-Horizon_Load_Forecasting
- Programming language
- Python
- Development Status
- Active