EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Residential Energy Demand‐Side Management Using Deep Learning Models: A Steady State Energy Disaggregation and Forecasting of Appliance Consumptions

Elvis Tamakloe, Benjamin Kommey, Jerry John Kponyo, Eric Tutu Tchao, Andrew Selasi Agbemenu

The Journal of Engineering · 2026-01 · 2026 (1)

DOI: 10.1049/tje2.70192 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of Engineering, online erschienen am 25.05.2026. Verfasst von Elvis Tamakloe und 4 weiteren Autor:innen. Die Arbeit nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Poor energy conservation practices in residential settings due to reckless use of energy and faulty appliances account for the excess percentage of energy consumed daily. In the quest to curtail this prevalent situation, several user‐centric methods have been suggested. The goal of this work is directed toward addressing the challenges of employing intrusive load monitoring (ILM) methods by using deep learning (DL) algorithms to optimize ILM non‐intrusively. To solve this challenge, convolutional neural network–weighted recurrence plot (CNN–WRP) energy disaggregation algorithms were proposed for appliance classification and forecasting. Approximately a 70.77% R 2 score with an MAE of 0.1638 was recorded by the proposed model with respect to forecasting both aggregated and appliance power consumption. A classification accuracy, F1‐score and Hamming loss of 98.33%, 99.88% and less than 0.1% were produced on the Almanac of Minutely Power dataset (AMPds) dataset v2, respectively. The performance was afterward evaluated against other state‐of‐the‐art benchmark models. The significance of this work lies in using a steady state WRP feature extraction modality to capture intrinsic load signatures to optimize household consumption. The adoption of this energy disaggregation model importantly enables customers to curb wastage, save on electric bills, understand the health status of their appliances and get recommendations for their energy needs. This redounds to the benefit of utilities that utilize such feedback data to accurately make decisions and effectively plan the power distribution system. Implementing energy disaggregation on a larger scale facilitates the realization of smart and sustainable cities, which is vital to the success of industry 4.0.

Zitieren

APA

Tamakloe, E., Kommey, B., Kponyo, J. J., Tchao, E. T., & Agbemenu, A. S. (2026). Residential Energy Demand‐Side Management Using Deep Learning Models: A Steady State Energy Disaggregation and Forecasting of Appliance Consumptions. The Journal of Engineering, 2026(1). https://doi.org/10.1049/tje2.70192

Vancouver

Tamakloe E, Kommey B, Kponyo JJ, Tchao ET, Agbemenu AS. Residential Energy Demand‐Side Management Using Deep Learning Models: A Steady State Energy Disaggregation and Forecasting of Appliance Consumptions. The Journal of Engineering. 2026;2026(1). doi:10.1049/tje2.70192

BibTeX
@article{Tamakloe2026,
  author = {Tamakloe, Elvis and Kommey, Benjamin and Kponyo, Jerry John and Tchao, Eric Tutu and Agbemenu, Andrew Selasi},
  title = {Residential Energy Demand‐Side Management Using Deep Learning Models: A Steady State Energy Disaggregation and Forecasting of Appliance Consumptions},
  journal = {The Journal of Engineering},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1049/tje2.70192}
}
Kostenlos starten →