94. a computational modeling approach of user behavior for swarm control applications

Department: Computer Science & Engineering
Faculty Advisor(s): Tajana S. Rosing

Primary Student
Name: Dhanesh Girish Pradhan
Email: dpradhan@ucsd.edu
Phone: 858-729-8241
Grad Year: 2017

Abstract
As increasing number of household appliances are connecting to Internet we can leverage their sensing, monitoring and actuation capabilities to predict user behavior. Such contextual user behavior can be utilized to schedule household appliances to minimize energy requirements, maximize cost savings and improve user satisfaction. In this poster we present a system which combines demographic based user behavior models, household appliance energy and indoor user localization data to create an optimized appliance schedule. It is also used to produce various house profiles with different energy demand characteristics in a reproducible manner. Comparison with real data shows that our model captures the power demand differences between different family types and accurately follows the trends seen in real data. For different families, generated power trace matches the peak power times and obtains 38% absolute mean error, with minimum 0.25% error.

Industry Application Area(s)
Control Systems | Energy/Clean technology | Internet, Networking, Systems

« Back to Posters or Search Results