Detecting Appliances in Energy Traces
[Examensarbete på avancerad nivå]
The amount of data gathered in the world is increasing every day. More and more energy data is being gathered from households and smart energy meters. To extend the functionality of these energy traces, Non-Intrusive Load Monitoring (NILM) algorithms can be used. These algorithms use training data in the form of appliancespecific energy trace to label different sections of the aggregated energy trace with activity.
In this thesis, we investigate how to create a data set with the goal of using it to investigate NILM algorithms, and to build a platform for future student projects in the area of energy trace data sets. This platform contains suggestions on what methods to use for gathering data, how to store it, and how to analyse it.
Nyckelord: Big Data, Machine Learning, Cyber-Physical Systems, Smart Grid, Advanced Metering Infrastructures
Det verkar som att din webbläsare saknar möjlighet att visa PDF. Klicka Öppna i nytt fönster för att läsa.