Speaker: Daniel (Jack) Kelly
Title
“Smart Meter Disaggregation: estimating the electricity consumption of individual appliances from a whole-house smart meter signal”
Abstract
“Managing electricity consumption as economically as possible is becoming increasingly important due to high energy prices and environmental concerns. A useful first step towards reducing energy consumption is to obtain a quantitative understanding of current usage.
The UK government have mandated that every house in the UK must have a smart meter installed by 2019. These meters will measure whole-house, aggregate electricity consumption once every five seconds. However, evidence suggests that consumers are best able to reduce their consumption when given appliance-by-appliance consumption data rather than aggregate consumption data.
The ultimate aim of my PhD is to build a system capable of disaggregating smart meter data whilst requiring minimal effort on the part of the user. In particular, the system should not require any prior information about which appliances are active within each house, and the majority of users should never have to train the system.
This talk will describe the main challenges, some existing approaches and my research plan. The literature in this area dates back to the mid-1980s. One challenge which has received little research so far is modelling multi-state appliances, so this is likely to be one of my first areas of research.”