Invention Reference Number
Water heaters and heating, ventilation, and air conditioning (HVAC) systems collectively consume about 58% of home energy use. Managing these systems is crucial for managing peak demand, carbon emission, energy consumption, electricity price and integrating distributed renewables into older grid systems. However, model forecasts lack anomaly detection systems which are essential for efficient use of devices as they can allow to study the behavior of the model, improve forecasting load accuracy, and implement error correction schemes. This technology is an autonomous system that can perform data curation, data cleaning, conduct error analysis, and detect anomalies of model forecasts of HVAC and water heater systems.
Description
This technology is software based on machine learning that autonomously identifies anomalies in HVAC and water heater systems in homes and commercial buildings. It is time-consuming and intensive to detect these anomalies manually as a residential neighborhood often contains several homes. The complexity further increases with the increasing time period of analysis. This is an autonomous system that can perform data curation, data cleaning, conduct error analysis, and detect anomalies of model forecasts of HVAC and water heater systems, and can classify anomalies for subgroups such as different floors, fan use, corner units, basement and water usage profiles.
Benefits
- Only method to automatically detect anomalies in predictive control and correct them without human intervention
- Saves time and money
- More efficient
- Can run on laptop or cloud
- Increases occupant comfort
- Can detect anomalies in power grid
Applications and Industries
- Electric utilities
- HVAC/Plumbing
- Construction
Contact
To learn more about this technology, email partnerships@ornl.gov or call 865-574-1051.
