Achieving Household Energy Breakdown at Scale

Energy breakdown: the process of breaking down a home’s energy consumption into constituent appliances, is considered an important step towards energy saving in the residential setting. Most current energy breakdown techniques exploit IoT hardware to be installed in each home and are thus not scalable. This is specifically true in the Indian context where penetration of smart meters is close to zero. In this paper, I will first discuss how data at different temporal and device resolution can address different analytical approaches to achieve energy efficiency. I will then discuss in detail our recent research work wherein we propose an approach that does not require any additional hardware installation and can be scaled across a large number of homes to provide accurate monthly per-appliance energy consumption feedback. Our approach exploits the already available datasets with per-appliance energy consumption for a small set of homes through direct instrumentation. It intelligently transforms the energy consumption from such small set of homes even to regions with different weather conditions, thus further eliminating the need to instrument homes across all different types of regions. We demonstrate how our approach can easily scale to provide per appliance energy consumption feedback to all homes in India.

Date:
Speakers:
Amarjeet Singh
Affiliation:
Zenatix