Appliance consumption signature database and recognition test protocols

Abstract

We report on the creation of a database of appliance consumption signatures and two test protocols to be used for appliance recognition tasks. By means of plug-based lowend sensors measuring the electrical consumption at low frequency, typically every 10 seconds, we made two acquisition sessions of one hour on about 100 home appliances divided into 10 categories: mobile phones (via chargers), coffee machines, computer stations (including monitor), fridges and freezers, Hi-Fi systems (CD players), lamp (CFL), laptops (via chargers), microwave oven, printers, and televisions (LCD or LED). We measured their consumption in terms of real power (W), reactive power (var), RMS current (A) and phase of voltage relative to current (ϕ). We now give free access to this ACS-F1 database. The proposed test protocols will help the scientific community to objectively compare new algorithms.

Automatic identification of electrical appliances using smart plugs

“…We also intend to compare the results of this protocol with those of the second reference protocol proposed in [2], evaluating the ability to generalize on new appliances that are unseen in the training set. Finally, we will evaluate other types of feature extraction and feature reduction methodologies.…”

“…In this direction, the Tracebase database, a collection of about 1'000 single appliance electric signatures has been recorded [7]. Another database, the Appliance Consumption SignatureFribourg 1 or ACS-F1 is also available including two acquisition sessions of one hour for 100 appliances spread uniformly into 10 categories [2]. This database is used for the evaluations carried on in this paper.…”

“…For our evaluation, we use the recently recorded electric signature database ACS-F1 [2] 1 . The database contains two acquisition sessions of one hour on 100 home appliances spread into 10 categories.…”

“…We are reporting here on several improvement of our algorithms and on the use of an evaluation database that was recently released [2]. This database, called Appliance Consumption Signature-Fribourg 1 or ACS-F1, contains two acquisition sessions of one hour for 100 appliances spread uniformly into 10 categories.…”

Toward Web Enhanced Building Automation Systems

“…The target application is the automatic identification of electric appliances so that building residents can have detailed information on their electricity bill. The HMMs are trained using a publicly available database of electric signatures called ACS-F1 for Appliance Consumption Signature Fribourg 1 [11]. The database consists of two acquisition sessions of one hour for 100 appliances spread uniformly into 10 categories.…”