Dayne Tompson
energyFit’s home energy models are validating at exceptional absolute accuracy against metered consumption, and are now being tested across six additional climate zones. This article sets out the evidence, and issues an open call to academics who would like to be involved in the research.
The gap between modelled and metered energy use is a stubborn problem in residential building science. Design-stage predictions routinely diverge from what a household actually pays, and closing that divergence is what allows retrofit and upgrade advice to be relied upon. energyFit, developed in the ACT since 2016, was built to close it.
A Digital Twin of the Actual Home
energyFit constructs a calibrated digital twin of the specific dwelling, combining LiDAR capture with EnergyPlus simulation. EnergyPlus is embedded as the thermal core, with energyFit’s engine running over the top to model occupant behaviour, generation and storage, and the full appliance load, thermal and non-thermal alike. The output is a dollar-denominated upgrade roadmap: what each intervention costs, what it saves, and in what order to do it where household budgets cannot action all advantageous changes promptly.
Validated Accuracy
Against metered household consumption in the ACT, energyFit has validated at 93% absolute accuracy for general electricity use, 96.4% for gas, and 98.3% for hot water. These are absolute figures using standard climate files – higher accuracy is expected using Real-Time Year actual weather data directly coinciding in time with the metered consumption data and is currently being tested. Overestimates and underestimates are not netted against one another, so the results are not flattered by errors in opposite directions cancelling out. For work that has to hold up against a real electricity and gas bill, that is an exceptional result.
Six NCC Climate Zones
Accuracy in a single climate is necessary but not sufficient. The method is now being tested across all six NCC climate zones in NSW, to establish whether ACT-level accuracy holds across a wider range of thermal conditions, building stock, and household behaviour. That trial is the mechanism through which the evidence is being built.
Reducing the Cost of a Low-Bill Home
Delivering upgrades at scale has generally meant standardised packages: the same insulation, the same heat pump, the same solar specification across many homes. That approach is administratively efficient, but the marginal saving from any measure depends on the specific building. A retrofit that pays for itself in one dwelling can return very little in the one next door.
Because energyFit models each home individually, it can find the least-cost path to a target bill for that home, and identify which measures will do the work. The cost of delivering a low-bill home falls, not because ambition is lowered, but because capital goes where the model shows it will earn a higher return. For a scheme administrator, that means more low-bill homes per dollar of subsidy. The same applies even if the household is targeting a Net Zero Emissions outcome.
Call for Collaboration
The six-zone trial will generate an unusual quantity of matched modelled-and-metered data across diverse conditions. Researchers who would like to be involved are invited to make contact.
Collaboration would be especially valuable in thermal simulation validation, the economics of least-cost retrofit optimisation, behavioural questions around why households act on cost information, and the evaluation of information-barrier versus price-barrier segmentation in energy efficiency schemes.
Expressions of interest can be directed to Dayne Thompson at dayne@sjtconsulting.com.au
Disclousre of Interest
Exemplary Investments has made a small hybrid-debt investment in this work; and Exemplary Energy will conduct a randomly selected blind re-simulation of a sample of homes to establish any enhancement of accuracy achieved by applying Real-Time Year weather data to the energyFit method.
