Residential Electricity-Use Profiling for Demand-Side Management Using a Convolutional Attention Variational Autoencoder and Adaptive GWO-K-Means Clustering

Wang, Jing; Chen, Meng; Peng, Mengfei; Cui, Xue; Yang, Huangyi; Dang, Xuehan

doi:10.3390/en19184430

Open AccessArticle

Residential Electricity-Use Profiling for Demand-Side Management Using a Convolutional Attention Variational Autoencoder and Adaptive GWO-K-Means Clustering

by

Jing Wang

¹,

Meng Chen

¹,

Mengfei Peng

¹,

Xue Cui

^2,*,

Huangyi Yang

² and

Xuehan Dang

²

¹

State Grid Hubei Electric Power Trading Center, Wuhan 430040, China

²

School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, China

^*

Author to whom correspondence should be addressed.

Energies 2026, 19(18), 4430; https://doi.org/10.3390/en19184430 (registering DOI)

Submission received: 1 July 2026 / Revised: 8 August 2026 / Accepted: 16 September 2026 / Published: 18 September 2026

Download

Browse Figures

Versions Notes

Abstract

Residential load profiling can support the design of differentiated demand-side management, but clustering methods are sensitive to representation quality and initialization. We developed a profiling workflow that combines a convolutional attention variational autoencoder (CA-VAE), adaptive gray wolf optimizer K-Means (GWO-K-Means) search over candidate cluster numbers and centers, and a clustering-loss refinement stage. On 6390 prepared residential average-day load profiles from the Commission for Energy Regulation dataset, ten independent runs yielded a final silhouette coefficient of 0.5002 (SD 0.0314), a Davies–Bouldin index of 0.8532 (SD 0.0383), and a Calinski–Harabasz index of 3694.3 (SD 260.5). Relative to raw K-Means, the final workflow improved all three internal validity measures (paired Wilcoxon signed-rank test, p = 0.001953 for each metric). GWO selected K = 5 in 6 of 10 runs; this modal solution produced five interpretable load profiles. The profiles provide load-shape-derived candidate groups for future demand-response trials, rather than validated estimates of operational flexibility.

Keywords: residential electricity use; load clustering; variational autoencoder; self-attention; gray wolf optimizer; joint optimization; demand response

Share and Cite

MDPI and ACS Style

Wang, J.; Chen, M.; Peng, M.; Cui, X.; Yang, H.; Dang, X. Residential Electricity-Use Profiling for Demand-Side Management Using a Convolutional Attention Variational Autoencoder and Adaptive GWO-K-Means Clustering. Energies 2026, 19, 4430. https://doi.org/10.3390/en19184430

AMA Style

Wang J, Chen M, Peng M, Cui X, Yang H, Dang X. Residential Electricity-Use Profiling for Demand-Side Management Using a Convolutional Attention Variational Autoencoder and Adaptive GWO-K-Means Clustering. Energies. 2026; 19(18):4430. https://doi.org/10.3390/en19184430

Chicago/Turabian Style

Wang, Jing, Meng Chen, Mengfei Peng, Xue Cui, Huangyi Yang, and Xuehan Dang. 2026. "Residential Electricity-Use Profiling for Demand-Side Management Using a Convolutional Attention Variational Autoencoder and Adaptive GWO-K-Means Clustering" Energies 19, no. 18: 4430. https://doi.org/10.3390/en19184430

APA Style

Wang, J., Chen, M., Peng, M., Cui, X., Yang, H., & Dang, X. (2026). Residential Electricity-Use Profiling for Demand-Side Management Using a Convolutional Attention Variational Autoencoder and Adaptive GWO-K-Means Clustering. Energies, 19(18), 4430. https://doi.org/10.3390/en19184430

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Residential Electricity-Use Profiling for Demand-Side Management Using a Convolutional Attention Variational Autoencoder and Adaptive GWO-K-Means Clustering

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI