Please use this identifier to cite or link to this item: http://dspace.cityu.edu.hk/handle/2031/7087
Title: Appliance Signature and Energy Auditing for Sustainability - CityCom Technology Limited
Other Titles: Appliance signature identification solution using K-means clustering
Appliance signature: multi-modes electric appliances
Authors: Chui, Kwok Tai (徐國泰)
Department: Department of Electronic Engineering
Issue Date: 2013
Course: EE4181 Project
Programme: Bachelor of Engineering (Honours) in Electronic and Communication Engineering
Supervisor: Supervisor: Dr. Tsang, K F; Assessor: Dr. Chan, W S
Description: Nominated as OAPS (Outstanding Academic Papers by Students) paper by Department in 2013-14.
This paper is under embargo at City University of Hong Kong until March 2015.
Two relevant conference papers have been published: 1. Chui, K. T., Tsang, K. F., Chung, S. H., & Yeung, L. F. (2013). Appliance signature identification solution using K-means clustering. IECON 2013 - 39th Annual Conference of the IEEE Industrial Electronics Society, 8420-8425. doi: 10.1109/IECON.2013.6700545. 2. Chui, K. T., Hung, F. H., Li, B. Y. S., Tsang, K. F., and Chung, H. S. -H. (2014). Appliance signature: Multi-modes electric appliances. IEEE The International Conference on Consumer Electronics-China (ICCE-China), 1-3. doi: 10.1109/ICCE-China.2014.7029896.
Citation: Chui, K. T. (2013). Appliance signature and energy auditing for sustainability: CityCom Technology Limited (Outstanding Academic Papers by Students (OAPS)). Retrieved from City University of Hong Kong, CityU Institutional Repository.
Abstract: Sustainability, energy conservation and demand response have become an inevitable concern around the world. This project proposes a sustainable and all-round system which consists of three partitions, advanced metering infrastructure, energy auditing and appliance signature. Advanced metering infrastructure in smart grid composes of hundreds of smart meter which collect total energy consumption of each apartment to the utilities through wireless communication. It is first of its kind that a WiFi and ZigBee building area network of advanced metering infrastructure is designed to overcome the signal obstruction and heavy traffic loading inside the complex structure of building. An understanding of the specific load profile will intensively increase the energy saving awareness and strategy of end users. Thus, the proposed system has incorporated the sub-metering for energy auditing. For the development of energy auditing, ZigBee is utilized to cater for communication between the central server and automated sockets or light switches. The measurement of data is achieved by current transformers. The central server analyses the energy consumption of electric appliances (together with the household working hours) and generates the load profile. Local users can reduce the time of use of appliances with high electricity footprint in a judicious manner based on the load profile. Appliance signature is accompanied by energy auditing which classifies the type of end-user appliances without a prior knowledge. Multi-class support vector machine is newly employed for appliance signature with characteristics of a maximum margin classification with accuracy of 95%. It has improvement accuracy by 40.5% compared to traditional K-means algorithm. This provides an evidence for utilities in energy management according to the energy usage pattern. Based on this project, one journal paper has been published and two conference papers have been accepted. Also, the 2-Runners Up was awarded in the paper award competition organised by the HKIE CAD.
Appears in Collections:Electronic Engineering - Undergraduate Final Year Projects
OAPS - Dept. of Electronic Engineering

Files in This Item:
File Description SizeFormat 
conference_paper_1.html167 BHTMLView/Open
conference_paper_2.html167 BHTMLView/Open


Items in Digital CityU Collections are protected by copyright, with all rights reserved, unless otherwise indicated.