Code and data to replicate all results in the paper.
Requires Gurobi and MOSEK licenses (free for academic use).
uv syncuv run python run.py # Run everything
uv run python run.py --train # Train forecasters only
uv run python run.py --experiments # Run experiments only
uv run python run.py --figures # Generate figures onlyIndividual experiments:
uv run python run.py --baseline # Baseline policies
uv run python run.py --prescient # Prescient optimization
uv run python run.py --mpc # MPC policies
uv run python run.py --capacity # Capacity sweep
uv run python run.py --sensitivity # Sensitivity analysisResults are saved to results/, figures to figures/.
@misc{perezpineiro2026home,
title = {Home Energy Management under Tiered Peak Power Charges},
author = {David P\'erez-Pi\~neiro and Sigurd Skogestad and Stephen Boyd},
year = {2026},
eprint = {2307.07580},
archivePrefix = {arXiv},
primaryClass = {math.OC}
}