Skip to content
Technology

Numbers you can check,all the way down.

The standards say what a home should do. Our verification shows what each home actually does. One governed record carries both, on certified data. The product is HeatAssure™; this page is its working, written for the people who check.

Design to operation

The record does not die at handover.

Illustrative: most tools stop at handover. One record carries the promise across it.

Design tools usually stop at the standard. Operational analytics usually start cold, with no design lineage. We hold one record across the boundary: design calculations admissible against EN 12831 at the start, verification carrying the home in operation, each end checkable against the other.

The exhibit below shows what that design half carries: one certified heat pump, one city of homes, modelled three times with only the design flow temperature changed. That single design choice spans the whole gap between the two figures on the certificate.

Exhibit 06One certified heat pump, three radiator temperatures
Modelled data, not a test rig
median in the home, 35 °C flow
5.47
median in the home, 45 °C flow
4.35
median in the home, 55 °C flow
3.60

Each band spans the middle 80% of homes; the tick is the median home. The two certificate figures, measured on the test rig, are marked on every rung.

35 °C radiator water

The fleet beats the certificate: median 5.47 against the certificate figure of 5.0. 9.8% of homes model below it.

middle 80% of homes: 5.01 to 7.41
45 °C radiator water

The certificate figure of 5.0 is out of reach for 75.1% of the fleet: median 4.35 against it.

middle 80% of homes: 3.95 to 5.90
55 °C radiator water

The fleet lands on the certificate's other figure, 3.6: median 3.60, and 90.7% of homes model below the 5.0 figure.

middle 80% of homes: 3.24 to 4.92
Modelled seasonal efficiency in the home (SCOP)
Flow temperature is the temperature of the water a heating system sends to its radiators, fixed when the emitters are chosen at design time. The certificate quotes both ends of it from the test rig: 5.0 at 35 °C and 3.6 at 55 °C. In homes, that one design choice spans the same gap: the median home reads 5.47 at 35 °C, 4.35 at 45 °C and 3.60 at 55 °C. The whole distance between the two figures on the certificate is covered by emitter choice alone.SCOP: Annual space heating delivered to the home divided by the heat pump space-heating electricity input, modelled over a full year at hourly resolution.Modelled: one certified 8.5 kW air source heat pump, installed in 24,931 homes reconstructed from public EPC records, simulated for a full year at hourly resolution, Greater London, London CIBSE test reference year. Three runs of the same fleet, changing only the design flow temperature. Certificate values from the public record.
Aeterno exhibit series: 01 Technology · 02 Home · 04 Housing & Place · 06 Technology
The calculations

Built by us, checked against the government's own reference.

01

SAP 10.3

The UK's standard method for rating home energy, and the assessment standard in force today. We run it in-house, validated end to end against approved SAP software.

02

Home Energy Model

The standard that replaces it. We run that in-house too, checked against the government's official implementation.

03

Beyond the standards

We verify what each home actually does as evidence arrives, with a stated confidence, and say so plainly when the standard and the home part company. That layer is ours alone, and it works alongside the tools you already have.

Owning the whole calculation matters when a figure is questioned. There is no vendor to wait for. We open the working and answer.

Validation
  • Every calculation we stand behind is cross-validated against the government's own reference for its method, from approved SAP software to the official Home Energy Model implementation.
The methods

Built for a world of national methods.

Energy performance is judged by whichever national method is in force where a home stands, and the record is anchored to that method. That is the architecture. Each method we take on earns the same proof the UK calculations above already carry.

One governed record per home
SAP 10.3 · HEMUnited KingdomCross-validated against the government’s own referenceValidated, in production today
DEAPIrelandSEAI’s national method
NTA 8800NetherlandsThe single Dutch method
DIN V 18599GermanyThe GEG method
3CL-DPEFranceThe DPE method
OIB RL 6AustriaThe Energieausweis frame
The method is national; the record is anchored to the one in force where the home stands. Validation is claimed only where it exists.
Scale

Edge cases do not show up in ten runs.

They show up in millions. Running at housing stock scale is how we find the cases a demo never meets.

On the record
  • The Home Energy Model, run at 20 million home scale.
  • The same validated calculations power a reconstruction of UK housing stock covering around 18 million dwellings.

The two figures measure different things. 20 million homes is throughput: the scale we have run the Home Energy Model at. 18 million dwellings is reconstruction coverage: how much of the UK's roughly 30 million homes the stock reconstruction describes.

Governance

Governed by construction.

A number is only worth checking if you can see where it came from. The platform enforces that in three places.

Viewpoint conformance

Every screen resolves against a governed viewpoint. What you read is a checked view of the record, not a copy that can drift.

Validation status on every claim

Every number carries its source and its validation status, including the red ones. Failure stays visible until it is fixed.

Journaled actions

Every outbound action requires an approval token and lands in an append-only journal, so what happened is on the record.

As shipped
  • A live governed platform: every screen resolves against a governed viewpoint, and every claim carries its validation status, including the red ones.

The autonomy policy itself (Assist, Govern and Auto), and the provenance chain behind every recommendation, are set out on Trust & Governance.

The certified data layer

Certified products, not defaults.

A calculation is only as good as the product data underneath it. The exhibit below is the certified library we calculate against: 9,300 certified heat pumps, drawn from the public record.

Exhibit 01The certified heat pump library
Source: public certification record
Each point is one certified heat pump system, plotted from the public record by seasonal efficiency (SCOP at 35 °C) and rated output. 8,887 of the 9,300 systems on record carry both figures; values beyond SCOP 7.5 or 60 kW sit on the axis boundary.Aeterno exhibit series: 01 Technology · 02 Home · 04 Housing & Place · 06 Technology
The library in numbers

Drawn from 9,300 certified systems currently held in the library.

9,300
Certified systems on record
178
Manufacturers represented
4.72SCOP
Median seasonal efficiency at 35 °C
9.8kW
Median rated output at 35 °C
Distribution of SCOP at 35 °C3.17 – 8.75
The reference

We publish our working.

The HEM Technical Reference is Aeterno's independent documentation of the government's Home Energy Model, stage by stage, with the equations. If you want to check the standard at equation level, start there.

The reference will not answer everything. For the questions it does not cover, ask the people who built it.