A method of appliance detection based on features of power waveform

Ito, Minako; Uda, Ryuya; Ichimura, Satoshi; Tago, Kazuya; Hoshi, Tohru; Matsushita, Yasuyuki

doi:10.1109/saint.2004.1266131

Cited by 40 publications

(19 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They use a plug-in sensor to detect electrical events within a home. They leverage the fact that mechanical switches produce electrical noise [21], and that the noise characteristics can vary dramatically by appliance [48]. They apply machine learning techniques to recognize specific devices being turned on or off.…”

Section: Related Workmentioning

confidence: 99%

Unsupervised Disaggregation of Low Frequency Power Measurements

Kim¹,

Marwah²,

Arlitt³

et al. 2011

Proceedings of the 2011 SIAM International Conference on Data Mining

494

345

View full text Add to dashboard Cite

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.

show abstract

Section: Related Workmentioning

confidence: 99%

Unsupervised Disaggregation of Low Frequency Power Measurements

Kim¹,

Marwah²,

Arlitt³

et al. 2011

Proceedings of the 2011 SIAM International Conference on Data Mining

494

345

View full text Add to dashboard Cite

show abstract

“…This is disadvantageous because the provided data makes it difficult to identify per appliance power consumption or generate real-time feedback of energy usage for the end user. To achieve this traceability of energy usage, one field of research is to analyze high-fidelity power traces, measured by smart meters, and thus identify connected appliances and their current operating state [9][10][11] by their load signatures. Other research groups have worked in the field of wireless power metering.…”

Section: Related Workmentioning

confidence: 99%

A low-power wireless system for energy consumption analysis at mains sockets

2016

View full text Add to dashboard Cite

Introduction: Improving energy efficiency and reducing energy wastage is an important topic of our time. But it is quite difficult to figure out how much of our total electricity bill can be mapped to which device or at what time the device used it. We believe energy efficiency of normal households can be improved, if this kind of transparency would be available. In this article, we present a system for energy measurement at mains sockets to gain a transparent view of energy consumption for each device in a household. It consists of several smart energy measuring devices (SEMDs) that use a low-power radio protocol to dynamically build and connect to a radio network to transfer power usage date to a server. At the server, the data is stored and can be accessed via web interface. Results: Our primary goal was to build a back-end system for an energy metering platform with very low energy consumption. This platform can provide data for a variety of services that enables users (the consumers) to understand and improve their energy consumption behavior and increase overall energy efficiency of their households.

show abstract

“…electric switch) of home appliances can be analyzed. Masahito, Ryuya, and Satoshi [13] designed special electric energy parameters, analyzed voltage and current waveforms, and constructed a model for ordinary household electrical products. A current detector is used to collect data, which is combined with the low power-consuming transmission platform on the outlet to calculate the feature parameters, and the result is sent to the master server for data storage.…”

Section: Appliance Recognitionmentioning

confidence: 99%

A Dynamic Power Features Selection Method for Multi-appliance Recognition on Cloud-Based Smart Grid

Lai

Hwang

Chao

et al. 2014

2014 IEEE 17th International Conference on Computational Science and Engineering

View full text Add to dashboard Cite

Most current studies concerning appliance recognition focus on single appliance recognition, but for general home users, it is universal to simultaneously switch on and off multiple electric appliances. Therefore, this study discusses the recognition of a multi-appliance load, and aims to establish recognition sample data, while reducing the packet transmission quantity and computation complexity of the cloud server. This study proposes a dynamic power features selection method for multi-appliance recognition, which uses the electricity information detected by the smart meter to evaluate current operating condition and rate of change in order to dynamically determine the transmission interval time. As power features and electrical waveforms are not completely identical, the recognition architecture proposed in this study is divided approximately into two stages. The first stage of prediction is implemented by the Factorial Hidden Markov Model (FHMM), and in addition to doping out the presently probable load operation combinations, as well as their probabilities, the key point is to obtain all values after power feature standardization of each combination. The larger the value, the better the load condition combination represents the power feature. These values are ordered, and a specific percentage of the power features are selected to estimate the error of the recognition sample data, which is combined with the probability of the first stage as the final forecast result. According to the experimental results, in multi-load conditions, the first 25% of the power features have the maximum recognition of 83.75%. In the case of a single load, the first 75% of the power features have the maximum recognition of 94.59%.

show abstract

A method of appliance detection based on features of power waveform

Cited by 40 publications

References 6 publications

Unsupervised Disaggregation of Low Frequency Power Measurements

Unsupervised Disaggregation of Low Frequency Power Measurements

A low-power wireless system for energy consumption analysis at mains sockets

A Dynamic Power Features Selection Method for Multi-appliance Recognition on Cloud-Based Smart Grid

Contact Info

Product

Resources

About