synthetic data

high-resolution electricity consumption in open data

We publicly release a synthetic dataset of 10k one-year residential electric load curves, with their local temperatures, contracted power and Time-of-Use plans (Nabil et al., 2025). This open dataset generated by conditional latent diffusion captures the correlation between cold temperature and electric consumption that can be found in certain countries, e.g. in France.

Dataset available on Zenodo

Architecture for time series generation: latent diffusion with decoder conditioning on exogenous variables.

References

2025

  1. synthetic_data.png
    A synthetic dataset of French electric load curves with temperature conditioning
    Tahar Nabil, Ghislain Agoua, Pierre Cauchois, and 2 more authors
    In ICLR 2025 Workshop on Tackling Climate Change with Machine Learning, 2025