Electricity demand in electrified UK homes: the role of heat pumps, seasons, and property type

• A novel method combines SERL and EoH datasets for robust demand modelling. • Housing type is a key driver; detached homes contribute most to peak demand. • Heat pump adoption sharply increases evening peaks in detached/semi homes. • Full heat pump adoption can increase winter evening peak demand by over 100 %. • Scalable framework supports DNO planning, scenarios, and demand flexibility. The decarbonisation of domestic heating is central to achieving the UK’s netzero targets, with heat pumps expected to play a leading role in replacing fossilfuel boilers. However, widespread deployment risks overloading lowvoltage (LV) networks. This paper develops and validates a methodology to generate synthetic halfhourly electricity demand profiles for UK households by integrating smart meter data from the Smart Energy Research Lab (SERL; 2019–2024, 12,000 households) with monitored heat pump consumption from the Electrification of Heat (EoH) project (742 households, 2021–2022). Data quality was enhanced using a hybrid anomaly detection approach (Local Outlier Factor with statistical checks and manual review), preserving realistic peaks while filtering errors. Profiles were segmented by housing type and seasons and validated against Scottish and Southern Electricity Network (SSEN) data from the Courtland Road substation in Oxford. The model achieved strong accuracy: substationlevel MAPE of 7.5% and feederlevel MAPE of 18.2%, with RMSE of 4.32 kWh and 0.41 kWh, respectively. Scenario analysis shows evening peak demand rising from 68 kWh (baseline) to 76 kWh with 50% of detached homes adopting heat pumps, 87.6 kWh with 30% penetration in semidetached houses, and over 100 kWh at 50% adoption in terraced homes. Full adoption more than doubles the peak to 149.7 kWh. Detached and semidetached homes disproportionately drive earlystage peak growth, while terraced homes dominate later stages. The results highlight critical implications for LV reinforcement, showing the value of targeting areas with vulnerable housing stock. This is the first UK study to combine SERL and EoH datasets for validated projections of heat pump adoption impacts, offering a scalable framework for network planning under netzero transitions.