Thermal Profiling of Residential Energy Use

This work describes a methodology for informing targeted demand-response (DR) and marketing programs that focus on the temperature-sensitive part of residential electricity demand. Our methodology uses data that is becoming readily available at utility companies-hourly energy consumption readings collected from "smart" electricity meters, as well as hourly temperature readings. To decompose individual consumption into a thermal-sensitive part and a base load (non-thermally-sensitive), we propose a model of temperature response that is based on thermal regimes, i.e., unobserved decisions of consumers to use their heating or cooling appliances. We use this model to extract useful benchmarks that compose thermal profiles of individual users, i.e., terse characterizations of the statistics of these users' temperature-sensitive consumption. We present example profiles generated using our model on real consumers, and show its performance on a large sample of residential users. This knowledge may, in turn, inform the DR program by allowing scarce operational and marketing budgets to be spent on the right users-those whose influencing will yield highest energy reductions-at the right time. We show that such segmentation and targeting of users may offer savings exceeding 100% of a random strategy.