The project, entitled “High Fidelity Self-Learning Tool for Residential Load and Load Flexibility Forecasting” aims to develop a self-learning computation tool that will continuously monitor and forecast major energy loads in the home (HVAC, water heating, electric vehicles), as well as on-site power generation (PV), to determine an optimal dispatch schedule that minimizes home energy costs, while reducing utility grid load without compromising occupant comfort. This technology is unique in that it does not rely on residential battery storage, which at present is prohibitively expensive.
The project team integrates Fraunhofer USA CMI’s expertise in modeling and machine learning, Sense Corporation’s innovative non-intrusive disaggregation sensing technology, and Boston University’s expertise in scalable optimization methods to characterize, predict and schedule flexible loads in the home.
Learn more at DOE webpage here