Home heating flexibility – the key to affordable heat electrification or a performance gap in the making?

By Steven Heath, Technical Director
April 13, 2023

Ours is an era of software displacing hardware; 0s &1s replacing physical things. AirBnB created empty rooms from a digital platform to disrupt the hotel industry, while stepping into a bank is a distant memory for most. The electrification of everything, especially home heating, is heavily reliant on that same trick. Software & machine learning must displace the need to build power stations - or at least allow us to build fewer – if we are to deliver an affordable transition.

We wrote previously on how flipping the Energy Performance Certificate, and the policy tools that underpin it, allows us to stop talking about physical kit – heat pump or insulation installed - and start measuring the desired outcomes ‘high fabric and heat pump efficiency’. That piece ended where this kicks off – home heating flexibility.

Heat demand in buildings represents our thorniest ‘electrification’ problem...our Everest. Why? Because it is the biggest chunk of UK building energy demand driving the uppermost peaks in annual and daily energy demand. And it is the uppermost peaks that you must design your electricity system to service. You can’t buy electricity to meet unusually high peak demand if you don’t have the capacity to generate it. The UK Energy Research Centre’s[1] look at median vs peak energy demand (gas & electricity) when the “Beast from the East’ cold snap hit makes the case. At its peak, gas demand outstripped electricity demand fourfold.


Britain’s local gas demand and electrical system supply - median and maximum demand weeks. 2017-18 heating season.


Designing the electricity grid to service this peak would require huge additional capacity for relatively infrequent events.  We can reduce this need by first shrinking, and then shifting, heat demand. The broken record bit - ‘shrinking’ demand comes from flipping the EPC, and linked new build and retrofit policy, to allow measured fabric and heating system efficiency to be plugged in rather than estimated figures. Measured performance allows the EPC to recognise, & policy to reward, the desired outcome - more efficient homes and heat pumps rather than the ‘kit’ (insulation and heat pumps) as a proxy for these outcomes. Beast from the East 2, 3 & 4 will not care what the efficiency of home fabric or heat pumps ‘should be’ but what it ‘is’. This must happen at the promised EPC review & Future Homes Standard (FHS) consultation. We must ensure the 200,000 new homes built each year become low-demand highly-flexible ‘heat stores’ if we aren’t to undermine Government’s 15% domestic energy demand reduction target or saddle people with avoidably high energy bills.

If building measured efficiency into the EPC and the FHS can support heat demand ‘shrinking’, what of ‘shifting’ demand peaks. A Government sponsored review into the SAP software underpinning the Energy Performance Certificate had this to say:

‘Energy flexibility. All experts agree that new homes need to be better integrated in the wider energy system. They should have a reduced peak demand, and an increased ability to use energy when clean energy is available’

A sentiment that works for all existing homes plugged in to the grid too. Three principles can inform our home ‘shrink and shift’ strategy:

  1. Heat demand – DOWN
  2. Heat pump efficiency - UP
  3. Heat load – FLEXIBLE

While ‘real’, or measured, performance metrics exist for the first two (again cold snaps don’t care about estimates) what of flexibility? The Government expert review suggested a new module for ‘flexibility’ in SAP & the EPC. But how?

What will ‘smart energy flexibility’ look like?

Building a digital understanding of an individual home’s heat-up and cool-down periods in different weather (external temperatures, wind speeds, cloud cover) is complex. It will be the domain of machine learning rather than a desk top calculated metric. Once a home’s response to different weather is ‘learnt’, this knowledge is only useful if it then informs an automated ‘pre-heating’ strategy in a single home that delivers occupants’ desired comfort settings but prioritises the cheapest – non-peak time – electricity every half hour of every day. Especially in winter, and especially in cold periods in winter. The impact grows as homes are aggregated into that strategy, networking pre-heating across 100,000s of homes.

But beware regulating directly for this outcome. The market must be free to innovate & iterate its metrics to refine the optimal strategy offering ever better ‘electricity shifting’ services. As the grid’s ability to reward shifting grows, so must the services’ ability to respond to that reward.

What role for regulation?

Government & the Future Home Standard must set a flexibility ‘backstop’. Government is exploring its decarbonisation pathway in terms of power generation mix. It is the ultimate buyer on the capacity market & it is designing a role for a new ‘Future System Operator’. It is their role to set a ‘backstop’ or minimum flexibility requirement each electrically heated home must have.

It must consider the 200,000 homes to be built each year, or at least those to be electrically heated, and say they must come under this threshold or ‘backstop’. The SAP expert review agreed:

“The inclusion of an output related to peak demand and/or demand management (e.g. Smart Readiness Indicator, energy storage capability, peak demand) would allow SAP to value strategies aimed at reducing peak demand and at shifting demand for system flexibility. These would in turn support policies for the electricity grid to become lower carbon at a smaller cost.”

Where might this backstop be set? Begin with the problem that must be managed. Regularity and severity of cold snaps is recorded. Fairly new to us is the impact a cold snap accompanied by low wind & loss of wind generation recently seen in Texas. The regulator must consider how flexible it would like all electrically heated new homes to be whether that is a cold snap similar to the original ‘Beast from the East’ or something worse.

Setting a backstop

Flexibility in our home heating scenario is concerned with the ‘rate of temperature decay’. This is our backstop figure. Must new homes be able to be heated to 21 degrees 2 hours from peak time and lose 1 degree in that 4-hour period at an internal vs external temperature difference of 20 degrees? Or should we more ambitious? Whatever the figure we can see the backstop ‘currency’:

  1. Temperature loss rate ≤ 0.25° degree per hour at a 20° internal vs external temperature difference

Flex the numbers but the principle remains. Whole energy system modelling should collapse around a plausible ‘cold snap’ risk scenario and set the backstop accordingly. But how can we measure this backstop for building regulations compliance? Asking a house builder to wait for an in-situ measured performance pegged to an extreme weather event is a non-starter. We must focus on a ‘theoretical’ rate of decay which is informed by two home properties: thermal mass & heat transfer coefficient (HTC). The Heat Transfer Coefficient describes the heat transferred per unit area per kelvin. Thermal mass can be considered ‘how much energy a home releases when it becomes 1 degree colder relative to outdoor temperature’.

Theoretical rate of decay =

Thermal mass / Heat Transfer Coefficient * 20 degrees / 1 hour

How might such a ‘regulated’ backstop be measured? A theoretical Heat Transfer Coefficient (HTC) is already set in the design SAP underpinning the EPC while the ‘real performance’ measurement tools described can establish whether that design target HTC has been met once a home is built. The builder and occupant might have to wait a few weeks – but not have to wait for a 1 in 100 year cold snap.

The ‘thermal mass’ of a home – the numerator in our equation - can be straightforwardly approximated from the materials used to build it – concrete has high thermal mass; timber, low thermal mass. For Building Regulations purposes in setting a ‘flexibility’ backstop, it is likely enough to measure the HTC post-completion initially. If a house builder fails to deliver the promised design HTC there is a strong indication home insulation (and thermal mass) is being bypassed and / or air tightness targets have been missed increasing the rate of temperature decay.

The HTC denominator ‘switches on’ the thermal mass numerator; but only if the HTC’s been delivered in practice. Consideration in the backstop should be given to embodied carbon too – there may need to be backstops dependent on lightweight or heavyweight construction type.  

What of the smart software & machine learning home energy services that will avoid the need for power stations? These services will operate once a home is built or retrofitted at a level of granularity beyond this backstop. Heat loss at different wind speeds and temperature differences will be measured and the homes ‘thermal mass response’ understood.

Regulation must ensure homes ‘can be’ flexed in line with need in a modelled ‘worst case’ – the backstop. The smart stuff will do the flexing for the benefit of occupants and grid alike. Over time they will also straightforwardly back calculate whether the completed home has come close to hitting its backstop flexibility target.


Digital innovation across so many sectors has allowed us alternative, and often superior, routes to achieve our desired outcome. Refusing the seduction of ‘business as usual’ – the focus on ‘things’ not outcomes – will be a leap of faith for regulators. Yet, Government’s recent Green Day announcements offered some green shoots. The Energy Security Act will require all energy suppliers to offer time-of-use tariffs while all Energy Smart Devices, including heat pumps, will need to be interoperable with controls made by different manufacturers.

Grid storage, country interconnectors and batteries will play their role, but it is digital tools – software & machine learning – that will manage a smart ‘shrink and shift’ heat demand strategy. Digital tools will tell you your heat pump is hitting its target efficiency and your home fabric is slowing heat loss as intended. Digital will heat your home in the early hours away from expensive peak electricity knowing your home can retain that heat through the peak period. Digital will help us climb our heat electrification Everest… or its absence will get us stuck at base camp. Which, if any, of the: EPC review, the impending Future Homes Standard consultation or the Energy Efficiency Taskforce will set us on the right path?


[1] Heat decarbonisation challenges: local gas vs electricity supply | UKERC | The UK Energy Research Centre