Abstract
The Non-Intrusive Load Monitoring (NILM) systems measure the energy consumption of individual electrical equipment and appliances connected to a home or building, over time intervals of a few days or weeks. The analysis of the energy consumed by each single device makes it possible to identify the least efficient or malfunctioning ones and to implement the appropriate actions aimed at reducing consumption. NILM systems are also useful when it is necessary to identify the devices in use at a given moment, regardless of the information associated with their consumption. This information is useful in fulfilling the feedback needs of modern smart home, energy management and assisted living systems. In this work an analysis method based on the Sweep Frequency Response Analysis (SFRA) technique, for the identification of loads, is analyzed. SFRA techniques are widely used in diagnostics and mainly for fault finding in transformers and electric motors. More specifically, in this work the SFRA technique has been applied for the identification of the signature of household appliances, for NILM applications, proposing a new system based on the machine learning (ML) technology.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have