ENERGYTICS

In autumn 2018, a master thesis written by a student at University of Oslo and associated with ENERGYTICS was submitted. The author, Bjørn Ingeberg Fesche, has done analysis on smart meter data to see if it is possible to detect electric vehicles in electricity consumption data from smart meters. To train neural network models, publicly available consumption data from USA was used, in which it is known when electric vehicles are charging and in addition it contains measurements of the electric vehicle chargings. The models were then run on Norwegian data, and the results are encouraging. The next step in the demonstrator on detecting electric vehicles will be to further assess the performance of the models on Norwegian data and train the models on Norwegian consumption data annotated with charging events and/or measurements of chargings of electric vehicles which are common in Norway.

Unfortunately, the master thesis is not available online.

Demonstrator: Detection of electric vehicles and other flexible appliances