Leakage-Aware Benchmarking of Machine-Learning Models for Residential Appliance Energy Prediction Using a Public Smart-Home Dataset
Progress in Energy and Environment · 12.07.2026 · 32 (2) · 36-51
DOI: 10.37934/progee.32.2.3651 ↗Kurz gefasst
Fachartikel in der Zeitschrift Progress in Energy and Environment, online erschienen am 12.07.2026. Verfasst von Hongzhi Lu und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Accurate residential appliance energy prediction is useful for building energy monitoring, demand-response research, and green-building analytics, but performance estimates can be sensitive to validation design. This study presents a fully computational benchmark using the public UCI Appliances Energy Prediction dataset to compare common machine-learning models under random, chronological, and blocked temporal validation. The workflow validates and preprocesses 10-minute smart-home observations, engineers time features, evaluates linear and tree-based regressors, quantifies uncertainty with bootstrap confidence intervals, and uses permutation importance to interpret predictive associations. Under chronological testing, the best non-baseline model was Ridge regression, with MAE 52.54 Wh, RMSE 84.94 Wh, and R2 0.126. Its RMSE changed by -12.8% relative to the random split, illustrating that random validation can misrepresent performance in time-dependent energy data. Feature ablation and interpretability results indicate that all variables and variables including RH_2, RH_1, RH_3, hour_cos, T3 were most informative for this dataset. The contribution is a reproducible, leakage-aware energy analytics workflow, not a deployment or measured energy-saving study.
Zitieren
APA
Lu, H., Lu, H., & Kassymova, G. K. (2026). Leakage-Aware Benchmarking of Machine-Learning Models for Residential Appliance Energy Prediction Using a Public Smart-Home Dataset. Progress in Energy and Environment, 32(2), 36-51. https://doi.org/10.37934/progee.32.2.3651
Vancouver
Lu H, Lu H, Kassymova GK. Leakage-Aware Benchmarking of Machine-Learning Models for Residential Appliance Energy Prediction Using a Public Smart-Home Dataset. Progress in Energy and Environment. 2026;32(2):36-51. doi:10.37934/progee.32.2.3651
BibTeX
@article{Lu2026,
author = {Lu, Hongzhi and Lu, Hongxue and Kassymova, Gulzhaina K.},
title = {Leakage-Aware Benchmarking of Machine-Learning Models for Residential Appliance Energy Prediction Using a Public Smart-Home Dataset},
journal = {Progress in Energy and Environment},
year = {2026},
volume = {32},
number = {2},
pages = {36--51},
doi = {10.37934/progee.32.2.3651}
}
EUVIMED