Power estimation of multiple two-state loads using a probabilistic non-intrusive approach


Téléchargements par mois depuis la dernière année

Plus de statistiques...

Henao, N., Agbossou, K., Kelouwani, S., Hosseini, S. S. et Fournier, M. (2018). Power estimation of multiple two-state loads using a probabilistic non-intrusive approach. Energies, 11 (1). p. 88. ISSN 1996-1073 DOI 10.3390/en11010088

[thumbnail of KELOUWANI_S_55_ED.pdf]
Télécharger (1MB) | Prévisualisation


This paper investigates a non-intrusive approach of retrieving electric space heater (ESH) power profiles from a residential aggregated signal. In cold-climate regions with heating appliances controlled by electronic thermostats, an accurate non-intrusive recognition of power profiles is a challenging task. Accordingly, a robust disaggregation approach based on the difference factorial hidden Markov model (DFHMM) and the Kronecker operation is contributed. The proposed method aims to uncover the underlying stochastic tow-state models of ESHs using their common prior knowledge. The major advantage of the developed load-monitoring architecture consists of modeling simplicity and inference as well as load-detection efficacy in the presence of perturbations from other unknown loads. The experimental results prove the effectiveness of the method in manipulating the challenging case of multiple two-state loads with a high event overlapping probability.

Type de document: Article
Mots-clés libres: Building energy consumption Non-intrusive load monitoring Smart grids Hidden Markov models
Date de dépôt: 12 avr. 2023 13:19
Dernière modification: 12 avr. 2023 13:19
Version du document déposé: Version officielle de l'éditeur
URI: https://depot-e.uqtr.ca/id/eprint/10636

Actions (administrateurs uniquement)

Éditer la notice Éditer la notice