I-Corps: Residential Energy Management and Analytics

点击翻译按钮获取中文摘要

The broader impact/commercial potential of this I-Corps project derives from its promise to positively affect not only consumers, but also enhance efficiencies in the electric energy marketplace as a whole. The basic functionality of matching users to optimal electric plans will result in both higher consumer satisfaction, and result in Retail Electric Providers (REPs) offering more competitive rate packages. Identifying the actual usage profile of appliances over time and providing actionable information to consumers will enable consumers to take informed decisions on appliance purchases, as well as manufacturers to optimize designs with real-world inputs. The platform will enable the solution of system-wide problems faced by utility companies like peak period demand surges to be countered through demand response initiatives such as incentivizing customers to modify their usage patterns to smoothen the load curve. Each function of the proposed system is geared toward addressing a specific source of friction in the electric energy marketplace, and consequently also possesses commercial potential. Thus, the overall economic impact will be on consumers, REPs, utility companies and appliance manufactures, while promoting a greater knowledge and engagement among the electricity consumers.This I-Corps project explores the value of creating a bundled energy management system aimed at residential users. The system uses machine-learning-based analytics of the customer's daily, weekly, monthly, and seasonal energy usage trends to offer potential savings through (1) recommending the best retail energy service provider plan matching the customer's usage patterns, (2) incentivizing customers to enable full integration with smart home devices, including smart thermostats, (3) identifying and predicting the electricity consumption of different appliances and providing actionable information on their optimal usage and maintenance, and (4) services to allow customers to navigate through the process of switching plans. A smartphone app available for the iOS and Android platforms forms the customer interface to the system. The key novelties of this project lie in the development and integration of machine learning tools and behavioral economics ideas into the domain of residential energy management. The project is founded on research into the design, development and validation of such tools in contexts such as predicting residential energy usage over time, disaggregating usage on a per-appliance basis, and experimentation on how best to motivate users to engage in energy usage behavior that induces efficient grid operation.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.