To efficiently reduce energy usage in buildings, it is necessary to understand how energy is consumed today. Non-intrusive load monitoring (NILM) is a promising approach where appliance level load profiles can be extracted from an agglomerated single-point measurement using statistical or machine-learning methodology. Moving NILM to the edge of the network holds many advantages like reduced operation cost and decreased power consumption while minimizing privacy concerns. In this paper, we present a NILM hardware that can apply real-time NILM on the edge of the network on an ultra-low power AI-optimized microcontroller.
Wohrl, H., Brunelli, D. (2020). Non-intrusive Load Monitoring on the Edge of the Network: A Smart Measurement Node. Cham, Switzerland : Springer [10.1007/978-3-030-37277-4_55].
Non-intrusive Load Monitoring on the Edge of the Network: A Smart Measurement Node
Brunelli D.Supervision
2020
Abstract
To efficiently reduce energy usage in buildings, it is necessary to understand how energy is consumed today. Non-intrusive load monitoring (NILM) is a promising approach where appliance level load profiles can be extracted from an agglomerated single-point measurement using statistical or machine-learning methodology. Moving NILM to the edge of the network holds many advantages like reduced operation cost and decreased power consumption while minimizing privacy concerns. In this paper, we present a NILM hardware that can apply real-time NILM on the edge of the network on an ultra-low power AI-optimized microcontroller.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
