Multi-appliance recognition system with hybrid SVM/GMM classifier in ubiquitous smart home

Lai, Ying-Xun; Lai, Chin‐Feng; Huang, Yueh‐Min; Chao, Han‐Chieh

doi:10.1016/j.ins.2012.10.002

Cited by 85 publications

(42 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…By applying event detection to classify the appliances, [15] get good performance on high power consumption appliances such as refrigerators. Hybrid support vector machine/Gaussian mixture model classification model [11] helps to process fuzzy power information. In [12], an expectation maximization (EM) based on multi-label classification method is applied in NILM.…”

Section: Related Workmentioning

confidence: 99%

A Framework of Energy Disaggregation Based on Adaptive Association Rules Mining

Long¹,

Chen²,

Li³

2017

dtetr

View full text Add to dashboard Cite

Non-intrusive load monitoring (NILM) process intends to deduce the individual energy consumption of appliances used in the house, which is mainly based on measuring the aggregate power consumption with a single smart meter. Comprehensive device consumption information can facilitate the improvements of the energy usage habits of the users. This work presents a novel framework on NILM via adaptive association rules mining, which aims to address the following issues. 1) The failure of detection for the appliances or group of appliances consuming the same or similar power. 2) Small power appliances' signals get overwhelmed by noise in the aggregate power data. To evaluate the effectiveness of our proposed framework, we implemented experiments based on the public datasets in production environment. By comparing with the state-of-art algorithms in general evaluation metrics, our framework gets around 10% gain on average in the performance of NILM task.

show abstract

Section: Related Workmentioning

confidence: 99%

A Framework of Energy Disaggregation Based on Adaptive Association Rules Mining

Long¹,

Chen²,

Li³

2017

dtetr

View full text Add to dashboard Cite

show abstract

“…Besides the prediction of demand, as pointed out in [21], the identification of sources of energy consumption within buildings is another key issue for the automated planning of energy schedules. This leads to the appliance recognition problem [32,42], which is another form of data mining, whose value is more localized and inherent to the energy management within the building.…”

Section: Big Data Analyticsmentioning

confidence: 99%

Smart Grid for the Smart City

Bonetto

Rossi

2016

Designing, Developing, and Facilitating Smart Cities

View full text Add to dashboard Cite

Modern cities are becoming more and more dependent on the reliability and efficiency of the electrical distribution infrastructure. In the past few years, a great effort has been devoted to the creation of an integrated infrastructure that combines a resilient power distribution system, distributed generation devices based on renewables (as, for example, photovoltaic panels and wind turbines), a reliable and secure communication system, and real-time energy pricing policies. The resulting infrastructure is called smart grid and constitutes the backbone of the smart city. As noted in [22], "The smart city is all about how the city "organism" works together as an integrated whole and survives when put under extreme conditions" and, moreover: "the energy infrastructure is arguably the single most important feature in any city. If unavailable for a significant enough period of time, all other functions will eventually cease." Hence, developing and implementing a fully functional smart grid infrastructure is a priority for future smart cities.In order to implement a smart grid, at least the following three technical domains must cooperate, namely: (i) power electronics, (ii) information and communication technology (ICT), and (iii) economics.From a power electronics standpoint, the diffusion of distributed energy resources (DERs) poses a number of interesting challenges that the future power distribution grid must face to evolve into a smart power distribution grid. On the one hand, the presence of grid connected DERs (i.e., DC power generators connected to the main power distribution system through grid-tie inverters or capacitor banks) must be carefully handled to maintain acceptable electrical power quality. In particular, voltage sags (overvoltages) due to the switch off (on) of groups of DERs must be avoided. Moreover, harmonic distortion must be minimized, since it is one of the major factors degrading the overall power quality [18]. On the other hand, if the presence

show abstract

“…After edge detection, features (for example, active power) are extracted in the identified event windows, and then the events are classified into pre-defined categories, each corresponding to one known appliance. Different state-of-the-art classification tools have been used, including SVM (for example, in [22], [23], [24]), neural networks (for example, in [25], [26]), and decision trees [27], [4]. However, the performance of these event-based NALM approaches, is limited by the event detection tool employed.…”

Section: B Low-rate Nalmmentioning

confidence: 99%

A graph-based signal processing approach for low-rate energy disaggregation

Stanković

Liao

2014

2014 IEEE Symposium on Computational Intelligence for Engineering Solutions (CIES)

View full text Add to dashboard Cite

This version is available at https://strathprints.strath.ac.uk/51401/ Strathprints is designed to allow users to access the research output of the University of Strathclyde. Unless otherwise explicitly stated on the manuscript, Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Please check the manuscript for details of any other licences that may have been applied. You may not engage in further distribution of the material for any profitmaking activities or any commercial gain. You may freely distribute both the url (https://strathprints.strath.ac.uk/) and the content of this paper for research or private study, educational, or not-for-profit purposes without prior permission or charge.Any correspondence concerning this service should be sent to the Strathprints administrator: strathprints@strath.ac.ukThe Strathprints institutional repository (https://strathprints.strath.ac.uk) is a digital archive of University of Strathclyde research outputs. It has been developed to disseminate open access research outputs, expose data about those outputs, and enable the management and persistent access to Strathclyde's intellectual output.A graph-based signal processing approach for low-rate energy disaggregation Abstract-Graph-based signal processing (GSP) is an emerging field that is based on representing a dataset using a discrete signal indexed by a graph. Inspired by the recent success of GSP in image processing and signal filtering, in this paper, we demonstrate how GSP can be applied to non-intrusive appliance load monitoring (NALM) due to smoothness of appliance load signatures. NALM refers to disaggregating total energy consumption in the house down to individual appliances used. At low sampling rates, in the order of minutes, NALM is a difficult problem, due to significant random noise, unknown base load, many household appliances that have similar power signatures, and the fact that most domestic appliances (for example, microwave, toaster), have usual operation of just over a minute. In this paper, we proposed a different NALM approach to more traditional approaches, by representing the dataset of active power signatures using a graph signal. We develop a regularization on graph approach where by maximizing smoothness of the underlying graph signal, we are able to perform disaggregation. Simulation results using publicly available REDD dataset demonstrate potential of the GSP for energy disaggregation and competitive performance with respect to more complex Hidden Markov Model-based approaches.

show abstract

Multi-appliance recognition system with hybrid SVM/GMM classifier in ubiquitous smart home

Cited by 85 publications

References 18 publications

A Framework of Energy Disaggregation Based on Adaptive Association Rules Mining

A Framework of Energy Disaggregation Based on Adaptive Association Rules Mining

Smart Grid for the Smart City

A graph-based signal processing approach for low-rate energy disaggregation

Contact Info

Product

Resources

About