In this paper, author proposed a multi-sensor system for power monitoring. This system can provide the aggregated power of applications and sensor values that are related to individual application power. e.g. Light intensity and noise intensity. An data processing and calibrating method is also presented to estimate the power of individual application using previous sensor values. In the workshop, I will introduce the system and calibrating algorithm.
Title: Unsupervised Disaggregation of Low Frequency Power Measurements (SDM2011)
Fear of increasing prices and concern about climate change are motivating residential power conservation efforts. We investigate the effectiveness of several unsupervised disaggregation methods on low frequency power measurements collected in real homes. Specifically, we consider variants of the factorial hidden Markov model. Our results indicate that a conditional factorial hidden semi-Markov model, which integrates additional features related to when and how appliances are used in the home and more accurately represents the power use of individual appliances, outperforms the other unsupervised disaggregation methods. Our results show that unsupervised techniques can provide perappliance power usage information in a non-invasive manner, which is ideal for enabling power conservation efforts.
Title: REDD: A Public Data Set for Energy Disaggregation Research (KDDsust2011)
In this paper, the authors present the REDD, a freely available data set containing detailed power usage information from several homes, which is aimed at furthering research on energy disaggregation. The paper discuss past approaches to disaggregation and how they have influenced authors' design choices in collecting data, and describe hardware and software setups for data collection, and present initial benchmark disaggregation results using a well-known FHMM(Factorial Hidden Markov Model) technique.
Title: Using Hidden Markov Models for Iterative Non-intrusive Appliance Monitoring (KDDsust2011)
The paper proposes an approach by which individual appliances are iteratively separated from the aggregate load. In this paper, prior models of general appliance types are tuned to specific appliance instances using only signatures extracted from the aggregate load. The paper evaluate approach using the REDD data set, and show that it can disaggregate 35% of a typical household's total energy consumption to an accuracy of 83% by only disaggregation three of its highest energy consuming appliances.
Title: Modelling electricity consumption in office buildings: An agent based approach (Energy and Building, 2010)