Sense has partnered with Fraunhofer USA and Boston University to deliver on a Department of Energy (DOE) pilot program. This project is part of the DOE’s nearly $83 million in funding to 44 projects that aim to lower Americans’ energy bills by investing in new energy-efficient building technologies, construction practices, and the U.S. buildings-sector workforce.
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 solar), to determine an optimal dispatch schedule that reduces utility grid load without compromising occupant comfort, and potentially achieving cost-savings for the end user.
By collaborating on the project, Sense aims to improve its algorithms and ability to help its users reduce their carbon footprint. Potentially helping customers save money – while maintaining their comfort and having a minimal impact on their day.
The Why and the How
Sense is working to advance device detection methods which will help improve energy science modeling. The goal is to improve energy management programs, significantly contributing to a brighter, cleaner future!
The program will show how the incorporation of data from multiple sources can be used to model household net load and increase the accuracy of generation and load forecasts, as well as improve Sense’s device detection capabilities. Some of the sources include communicating thermostats & EVs, smart appliances, weather forecasts and predictive formulas developed by our partner Fraunhofer.
Sense will analyze the data from registered devices, ensuring first and foremost that all data is isolated and anonymized. Our goal is to gather valuable insights into energy usage patterns, enhance thermostat and EV detection, and contribute to the advancement of energy management technologies while prioritizing customer preferences and comfort. Through micro-actuation setpoints and charging actions, Sense will improve its ability to give you customized reporting on Energy Savings, Appliance performance, and potential fault detections.
Through this pilot, Sense aims to improve its existing features through:
- Enhancing current device recognition capabilities & data disaggregation
- Assessing household preferences on load-flexibility
- Monitoring performance of HVAC & other major appliances to detect potential abnormalities that could lead to device failure and risks
- Evaluating the potential of actuating controllable devices
What’s next
With preliminary research now complete, the program has recruited a select group of active Sense users to get unparalleled insights into their energy usage, contribute to cutting-edge research, and pave the way for a greener, more sustainable energy future.
The pilot program will run from January 2024-December 2024, with device actuation happening up to 7 individual times between May-September of that year. Sense will slightly adjust participants’ thermostats or modify EV charge settings for a brief period to measure the impact of the adjustments on energy consumption. Sense will ensure that these changes do not impact participants’ day-to-day life or comfort.
Pilot participants will always receive advance notice and will always have the option to opt out or override the proposed settings.
Who Are the Partners?
The Fraunhofer USA – Center for Manufacturing Innovation is a non-profit applied R&D organization located in Boston. CMI’s Energy Systems activity works with industry to research, develop, and demonstrate new and emerging technologies to help enable and accelerate the transition to a sustainable energy future.
Sense is supporting the project with our residential device detection technology and platform as well as engaging Sense customers in the pilot. Sense has engaged Enode, a tech firm that enables integrations across multiple devices, making the program seamless for participants. Additional research is being provided by the Boston University Center for Information and Systems Engineering.