Station-Level Power Forecasting and Joint Price and Power Optimization Algorithms

Setup

• To install the necessary packages, navigate to time_series_formulation and data, and run pip install -e ..
• For dataset access, email ecal@berkeley.edu.
• simulations/general_simulator.ipynb contains infrastructure to test each optimizer with a Monte Carlo simulation and is a good starting point for understanding the codebase.

Code

Below we provide a brief description of each subdirectory.

• data: stores utility functions used for data cleaning and feature engineering.
• peak_power_formulation: a very early iteration of this project; we are 99% it will not be useful for future work.
• simulations: contains various optimization algorithms for station-level joint price and power optimization algorithms. Each optimizer is named with the format ________Simulator.py. Every optimizer is a child of BaselineSimulator.py. simulations/general_simulator.ipynb contains infrastructure to test each optimizer with a Monte Carlo simulation.
• time_series_formulation: contains all logic for station-level power forecasting. Run time_series_formulation/src/slrp_ev_ts_forecasting/main.py to train, test, and validate a forecasting algorithm on the SLRP-EV dataset.

Citation

If you use this repository or ideas from it, please cite the following work:

Thibaud Cambronne, Samuel Bobick, Wente Zeng, and Scott Moura. "Joint Price and Power MPC for Peak Power Reduction at Workplace EV Charging Stations." arXiv preprint arXiv:2507.12703 [eess.SY], 2025. https://doi.org/10.48550/arXiv.2507.12703

Data

Data used for this project is collected from SLRP-EV, a human-in-the-loop EV charging experiment conducted on the UC Berkeley campus. For dataset access, email ecal@berkeley.edu. For more information, see the following works:

Teng Zeng, Sangjae Bae, Bertrand Travacca, and Scott J. Moura. "Inducing human behavior to maximize operation performance at PEV charging station." IEEE Transactions on Smart Grid, vol. 12, no. 4, pp. 3353–3363, IEEE, 2021. https://doi.org/10.1109/TSG.2021.3067072

Hassan Obeid, Ayşe Tuğba Öztürk, Wente Zeng, and Scott J. Moura. "Learning and optimizing charging behavior at PEV charging stations: Randomized pricing experiments, and joint power and price optimization." Applied Energy, vol. 351, 121862, Elsevier, 2023. https://doi.org/10.1016/j.apenergy.2023.121862

Ayşe Tuğba Öztürk, Hassan Obeid, Teng Zeng, Wente Zeng, and Scott J. Moura. "Joint price and power optimization experiment for workplace charging stations." Sustainable Cities and Society, 106784, Elsevier, 2025. https://doi.org/10.1016/j.scs.2025.106784.