Context-aware and user-centric residential energy management

Abstract

The Internet of Things (IoT) has brought increased sensing, monitoring and actuation capabilities to several domains including residential buildings. Residential energy management methods can leverage these capabilities and devise smarter solutions. This requires processing and reasoning data constantly generated by various IoT devices. In this paper, we use a hierarchical system model for IoT-based residential energy management, that includes a general purpose functional unit to drive data processing and reasoning. We apply this hierarchy to represent the electricity delivery structure from the utilities to individual residences. Our system captures additional data generated by various devices as user context and uses this context to determine user flexibility towards energy management. Our experiments show that modeling user context brings over 14% improvement in energy flexibility prediction accuracy and 12% reduction in annual grid energy cost.