Alireza Moradi, a Ph.D. student at Georgia Tech and research assistant with the NSF AI Institute for Advances in Optimization (AI4OPT), recently presented AI4OPT research at the Power Systems Engineering Research Center (PSERC) Industrial Advisory Board Meeting hosted by the Colorado School of Mines.

Moradi presented Realistic Injection Time Series for Node-Breaker Representations of Transmission Systems, a project focused on developing realistic synthetic nodal load and generation time series for the RTE7000 transmission network, a large-scale network topology dataset from RTE. The work aims to provide high-resolution, reproducible, and openly accessible data that supports research in machine learning, optimization, and power systems while preserving the confidentiality of real operational data.

To reconstruct nodal injections, the research combines publicly available network snapshots with regional time-series data, market information, and generator output data. The resulting dataset captures realistic temporal patterns and spatio-temporal correlations across loads and generators, enabling researchers to develop and evaluate new methods using representative system behavior.

A key outcome of the project is TS4PS, a high-resolution time-series dataset and reproducible codebase designed to support benchmarking, algorithm development, and open research in power systems.

The project is the result of collaborative work by Mathieu Tanneau, Reza Zandehshahvar, and Pascal Van Hentenryck, with Moradi contributing to the research and presenting its findings to industry and academic leaders at the PSERC meeting.

Resources

• Code repository: https://github.com/AI4OPT/TS4PS
• TS4PS dataset: https://huggingface.co/datasets/AI4OPT/TS4PS