Grid-Interactive Efficient Buildings (GEBs) can be a means to improving electricity affordability, enabling distributed energy integration, and improving the comfort of building occupants. The Building Technologies Office (BTO) within the Office of Energy Efficiency and Renewable Energy is researching and innovating in this space by focusing on how and when to use heat pump water heaters.
Over the past three years, the Energy Department's Pacific Northwest National Laboratory (PNNL) has teamed up with Bonneville Power Administration, Portland General Electric and 10 utilities in the Northwest to study the load shifting and energy-efficiency potential of heat pump water heaters relative to electric resistance water heaters in residential buildings. This field validation study included 10 weeks of data, almost 250 water heaters (145 heat pump and 86 electric resistance), and twice daily load shifts. In comparing heat pump to electric resistance, it was found that 90% of evening peak load power can be reduced by switching from uncontrolled electric resistance water heaters to connected heat pump water heaters. The results demonstrate a reduced risk for utilities and reduced costs to consumers.
Across households in America, water heating is the second-largest energy user.1 While typical heat pump water heaters can save 60% of annual energy consumption, there has been little market penetration in the past 10 years. Additionally, residential electric loads fluctuate, costing utilities much more at peak power than at base load power. The thermal characteristics of water heaters provide options to manage peak power requirements, but heat pump water heater load-shifting performance remains unproven in key regions with the greatest energy efficiency and peak demand reduction potential. To address these challenges, this BTO project demonstrated a 24/7 control paradigm for shifting load to align with renewable generation, quantified the load shifting value for utilities and consumers, and evaluated customer acceptance of the control paradigm.
Over 10 weeks during winter, researchers monitored this field validation in the Northwest. The 250 water heaters used the CTA-2045 protocol to shift water heater loads to reduce peak load and enable renewable energy integration. The CTA-2045 is a standard that enables otherwise disconnected devices to be connected. This is the first study that uses such a large number of connected heat pump and electric resistant water heaters from multiple manufacturers, which is significant because of the inherently complicated algorithms.
This is also the first study to implement several regular load-shifting events on a continual basis over the course of a year. In the study, loads were shifted twice daily during peak hours to take advantage of renewable resources. The first was a load-up event, forcing the water heater to turn on and reach set point. The second load shift was a shed event, forcing the water heater to turn off as long as customers still had hot water.
In comparing the heat pump and electric resistant water heaters, the largest peak reduction occurs with connected heat pump water heaters. The study found that when comparing connected heat pump to controlled electric resistant water heaters, 90% of the evening peak load power could be reduced. There is also greater peak load reduction with connected electric resistant than non-connected electric resistant water heaters. However, non-connected heat pump water heaters still had a greater peak load reduction than connected electric resistant water heaters. For benefits to consumers, converting all electric resistant to heat pump water heaters could save American consumers $7.8 billion annually ($182 per household) in water heating operating costs, and cut annual residential source energy consumption for water heating by 0.70 quads.2 With a more regional focus, this technology has the potential to save 1.1 terawatt hours per year in the Northwest.3
This field validation study occurred in the real world, with real consumer habits, which opens the door for consumer dissatisfaction with heat pump and electric resistant operations. Fortunately, most consumers seemed largely unaffected by the demand response program, as the ability to opt-out was only utilized 4% of the time. The largest benefit seems to be for utilities that are greatly constrained and thus very concerned about peak loads. Utilizing load shifting in heat pump water heaters allows them to reduce their costs, as peak costs more than base load, and also become less constrained.
This research has already impacted policy and decision makers in the field. In June 2019, the Regional Technical Forum (RTF) approved a load-shifting methodology for water heaters that was primarily based on this PNNL analysis. The RTF is a technical advisory committee to the Northwest Power and Conservation Council established in 1999 to develop standards to verify and evaluate energy-efficiency savings.
Although the research was focused in the Pacific Northwest, PNNL has developed a methodology to extrapolate results to other regions of the U.S. and is actively helping other utilities across the country understand the benefits of this load-shifting strategy. Due to the high concentration of electric resistance water heaters, this project identified the Southeast as a priority area for further field validation to transfer lessons learned and best practices from the Pacific Northwest studies to a new region. This study is ongoing and will pursue field validation efforts in the Southeast, as well as expand upon the field validations in the Northwest next year by using machine learning to help customize the load shifting controls for each individual water heater to optimize its load-shifting potential.
2 Building America Case Study: Field Performance of Heat Pump Water Heaters in the Northeast. /sites/prod/files/2014/01/f7/case_study_hpwh_northeast.pdf
3 Metzger, et al. 2018. Load Shifting Potential Using Storage Water Heaters in the Pacific Northwest https://www.bpa.gov/EE/Technology/demand-response/Documents/20181118_CTA-2045_Final_Report.pdf