ENERGY CONSUMPTION MONITORING

The disclosure notably relates to a computer-implemented method of machine-learning for non-intrusive monitoring of energy consumption of households with respect to a plurality of appliance classes. The method comprises obtaining a bidirectional variational auto-encoder architecture for a neural net...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Oublal, Khalid, Le Borgne, Emmanuel, Benhaiem, David
Format:	Patent
Sprache:	eng ; fre ; ger
Schlagworte:	ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVEREDIN A SINGLE OTHER SUBCLASS CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING MEASURING MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR PHYSICS TARIFF METERING APPARATUS TESTING
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	The disclosure notably relates to a computer-implemented method of machine-learning for non-intrusive monitoring of energy consumption of households with respect to a plurality of appliance classes. The method comprises obtaining a bidirectional variational auto-encoder architecture for a neural network, and training the neural network. The bidirectional variational auto-encoder architecture includes a temporal attention layer. The neural network is configured to take as an input a signal representing an aggregate electrical load of a household over time, and to output a plurality of signals each representing an electrical load of a respective appliance class over time.