Picture of smart phone

Open Access research that is better understanding human-computer interaction...

Strathprints makes available scholarly Open Access content by researchers in the Department of Computer & Information Sciences, including those researching information retrieval, information behaviour, user behaviour and ubiquitous computing.

The Department of Computer & Information Sciences hosts The Mobiquitous Lab, which investigates user behaviour on mobile devices and emerging ubiquitous computing paradigms. The Strathclyde iSchool Research Group specialises in understanding how people search for information and explores interactive search tools that support their information seeking and retrieval tasks, this also includes research into information behaviour and engagement.

Explore the Open Access research of The Mobiquitous Lab and the iSchool, or theDepartment of Computer & Information Sciences more generally. Or explore all of Strathclyde's Open Access research...

A low-complexity energy disaggregation method : performance and robustness

Altrabalsi, Hana and Liao, Jing and Stankovic, Lina and Stankovic, Vladimir (2014) A low-complexity energy disaggregation method : performance and robustness. In: 2014 IEEE Symposium on Computational Intelligence Applications in Smart Grid (CIASG). IEEE, Piscataway, NJ., pp. 1-8.

[img] PDF (Altrabalsi-etal-CIASG2014-low-complexity-energy-disaggregation-method)
Altrabalsi_etal_CIASG2014_low_complexity_energy_disaggregation_method.pdf
Accepted Author Manuscript

Download (138kB)

Abstract

Disaggregating total household's energy data down to individual appliances via non-intrusive appliance load monitoring (NALM) has generated renewed interest with ongoing or planned large-scale smart meter deployments worldwide. Of special interest are NALM algorithms that are of low complexity and operate in near real time, supporting emerging applications such as in-home displays, remote appliance scheduling and home automation, and use low sampling rates data from commercial smart meters. NALM methods, based on Hidden Markov Model (HMM) and its variations, have become the state of the art due to their high performance, but suffer from high computational cost. In this paper, we develop an alternative approach based on support vector machine (SVM) and k-means, where k-means is used to reduce the SVM training set size by identifying only the representative subset of the original dataset for the SVM training. The resulting scheme outperforms individual k-means and SVM classifiers and shows competitive performance to the state-of-the-art HMM-based NALM method with up to 45 times lower execution time (including training and testing).