Small logo of ETH main building ETH Zurich : Computer Science : Pervasive Computing : Distributed Systems : Education : Student Projects : Abstract

Integrating Submeters to Individually Monitor Appliances (L)

Status: Abgeschlossen


The electricity consumption in private households is a substantial part of the overall energy usage of our society. Many consumers, especially the residents in a house, can’t estimate how much electricity a certain appliance consumes, which makes it difficult for them to make smart decisions regarding saving energy. Utilities often charge their customers on a yearly or half-yearly basis. Residents only get a aggregated sum of their consumption what makes it even harder for them to figure out which devices use more than others. To improve the awareness of residents on their electricity consumption and to be able to make a supportive energy guidance, it is necessary to measure not only the total consumption of a household with the help of smart meters, but as well to measure critical devices such as refrigerators, washing machines or heaters, individually.

Scope of this Lab Project

This lab project explores measuring individual devices with the help of smart power outlets. Such an outlet is a small adapter installed between the power outlet and the appliance. It measures directly one single device and sends the results periodically via IEEE 802.11, Bluetooth or Zigbee to a central entity in the household, which collects measurements from many such outlets. This central device then provides a RESTful interface enabling other applications to get the data or it sends directly to a database. These measurements can be helpful in checking the quality of new algorithms for disaggregating electricity consumption data by providing some ground truth about important devices. Or they can support an existing measuring infrastructure system (e.g. ETH’s eMeter system) with additional data to improve the overall accuracy. In this lab, different kinds of smart power outlets are evaluated and tested. Sever-al criteria are taken into account, for example measured values (i.e. real-, reactive-, apparent power and phase angle), sample intervals, software- and hardware interfaces, reliability, accuracy and the price. The most suitable device is then the basis to implement a prototype of such a system, which will be deployed in a field study with a utility company.

Student/Bearbeitet von: Manuel Kläy
Contact/Ansprechpartner: Markus Weiss, Christian Beckel

ETH ZurichDistributed Systems Group
Last updated December 14 2012 10:46:45 AM MET mw