Peak times are when demand for electricity is very high, usually in the mornings and evenings. When electricity demand is high, it’s more likely that fossil fuel power stations need to be switched on to generate it. Using these more expensive sources of generation increases the overall cost of electricity for consumers, and is also bad for the planet.

As more and more homes switch to electric heat pumps the amount of electricity required for heating will increase. But heat pumps could also be part of the solution if they can use energy more flexibly and reduce their demand at peak times. One way this could work is by ‘pre-heating’ a home to a higher temperature before the peak period begins, then reducing the electricity demand from heat pumps when the grid is constrained at times of peak demand.

In the future, households could be paid to reduce their demand when the system needs it. National Grid began trialling this approach in winter 2022. We think flexibility will be more effective if consumers are willing to allow the companies that manage our energy system to control or automate their heating systems remotely.

The technology to use heat pumps flexibly already exists. Nesta and CNZ want to better understand how this flexibility can be delivered whilst maintaining household comfort levels and willingness to let someone else take control of their heating system.

Understanding how much flexibility each home with a heat pump can provide will benefit a range of different stakeholders. Transferring the learnings from HeatFlex UK could have the long term impact of allowing the flexible use of more existing heat pumps, which would bring down the costs of the whole energy system. Flexible demand means lower costs for everyone, and could make buying a heat pump an even more attractive option to individual households.

In winter 2022-23 we piloted a research method that we have used to inform a large-scale trial for the winter of 2023-24. Households participated in “HeatFlex UK events” where we remotely controlled their heat pump via a smart thermostat, to see if we can reduce or shift electricity demand and get feedback on whether their home was still at a comfortable temperature.

The goals of this work are to:

• identify an acceptable approach to remotely controlling heat pumps for flexibility while maintaining comfort and customer satisfaction
• measure the amount of flexibility (kWh reduced during a heat flex event) delivered by the approach
• examine how this impact varies by household and property characteristics.

We know that this trial will be complicated as there are many factors that influence flexibility. These include the type of home and how well insulated it is, the makeup of the household, the heating system itself, as well as people’s preferences and motivations.

Nesta is bringing user-centred design and behavioural science approaches to this project. We want to understand how heat flex events may best fit with users’ homes and lifestyles. We can then ensure the findings of the full-scale trial are rooted in our research questions and are robust and credible.

Over the course of the pilot, we ran 20 events during which we successfully took remote control of our participants’ heat pumps and were able to ‘pre-heat’ their homes before reducing their consumption until the end of the event. Participants completed surveys so that they could tell us about their household’s experience. To gain further insight we conducted a number of interviews at different points in the trial. The electrical consumption and temperature for each property were also recorded across the entirety of the project.

We have launched our trial and have conducted multiple HeatFlex events with our participants. After conducting test events, we were happy that our new method of remotely controlling heat pumps worked as intended and we’ve started to collect data for our mixed methods randomised controlled trial.

However, installing the physical devices into participants’ homes has been challenging, as we found in the pilot study last year. We have been working hard with our partners, CNZ and Octopus Energy, to identify homes where the smart thermostat will work, but we have still encountered difficulties installing the devices. We’ve found that some brands of heat pumps aren’t compatible with the smart thermostat, or that specific home arrangements have meant that we can’t install the smart thermostat (like complex, multi-zoned homes).

As a result of this, along with the usual attrition rates expected in a research trial, the sample size of our randomised controlled trial is smaller than we had hoped at the beginning of this project. Because we could not install smart thermostats in every household, our sample is also less representative of heat pump owners across the UK than we would have hoped. 

Despite these challenges, we are confident that we still have a large enough sample to make meaningful contributions to our original research aims. These additional learnings about the challenges of device interoperability should also be valuable when thinking about unlocking the potential of heat pump flexibility in the future.