Thesis Topic Details

Topic ID:
3474
Title:
Appliance Profiling based on High-Frequency Power Sampling
Supervisor:
Andreas Reinhardt
Research Area:
Wireless sensor networks applications, Machine Learning
Associated Staff
Assessor:
Andreas Reinhardt
Topic Details
Status:
Active
Type:
R & D
Programs:
CS CE
Group Suitable:
No
Industrial:
No
Pre-requisites:
Java and/or Python programming skills
Description:
This project seeks to understand if power readings collected at high sampling rates (in the range of several kHz) can be used to identify the current mode of an appliance's operation. The student will use a power analyser in order to collect power consumption traces of several electrical appliances in different modes of operation. Subsequently, an analysis of patterns in the collected data shall reveal whether characteristics in the power consumption can be unambiguously attributed to different modes of operation. Inferring device operation modes is a vital feature for smart environments that can adapt to a user's current activity and thus enhance user comfort. Prior knowledge in the domains of motif discovery through symbolic approximations [1] is beneficial.

[1] E. Keogh, J. Lin and A. Fu. HOT SAX: Efficiently Finding the Most Unusual Time Series Subsequence. Proceedings of the 5th IEEE International Conference on Data Mining (ICDM), 2005.
Comments:
This is a joint project between the schools of EE&T and CSE, and students from both disciplines are welcome to apply.
Past Student Reports
 
No Reports Available. Contact the supervisor for more information.

Check out all available reports in the CSE Thesis Report Library.

NOTE: only current CSE students can login to view and select reports to download.