The large-scale uptake of heat pumps to decarbonise the GB residential heat sector is expected to significantly impact the magnitude and shape of the electricity demand profiles at different spatial scales. Aggregate electricity demand profiles of heat pumps for regions are different due to varying characteristics of the housing stock that affect number and size of heat pumps that can be installed. This will lead to different level of network reinforcement needs at different Bulk Supply Points (BSPs).
Key objectives of this projects are:
- To estimate half-hourly electricity demand profiles for heat pumps at different spatial scales such as LA and national, for selected future scenarios.
- To quantify technically available flexibility from electrified residential heat
Benefits
- Improved temporally and spatially resolved data on electricity demand needed for heating residential buildings in future years. This will be used to inform the ESO Future Energy Scenarios (FES).
- Flexibility that can be exploited due to thermal inertia of buildings will be quantified. We don’t currently have the ability to model this.
- More robust heat pump demand profiles, informed from observed trial data. Our current heat pump profiles are derived from gas boiler thermal demand profiles, evidence suggests these are not optimal.
- Improved analysis and insights of peak demands and flexibility.
Learnings
Outcomes
At this stage of the project, we have gained early insights into the effects of heat pump profile operation on peak demands, whereby continuous profiles can deliver significant reduction in peaks. However, it is too early in the delivery to be able to incorporate this into our modelling quantitatively, as this will require the model outputs, which have not been fully developed yet. This will be further developed as part of work packages 3 and 4.
Lessons Learnt
From the literature review and initial analysis of estimating annual demands it was shown that there exists a relatively large amount of uncertainty in estimating annual heating demand. This was deemed to be due to factors such as limited data on housing stock, heat pump performance and occupant behaviour. Factors such as the recent cost of energy crisis and increased working from home since COVID-19 have also added to this uncertainty.
Assumptions were identified for the model to define long term scenarios and sensitivity analyses. These include housing data, EPC data (including data on potential energy efficiency), indoor heating temperatures, efficiency of heat pumps, weather data and heating demand profile patterns.
Two methods for estimating annual heat demand were performed and compared against government ECUK data. Both methods, one using heat demand directly from EPC data, and one using the project’s lumped parameter RC model for half hourly consumption, tended to overestimate heating demand compared to the ECUK data. For validation, there will be a focus on comparing the model outputs to external sources.
Different scenarios for heat demand profiles were explored by distributing the aggregated demand among pre-defined heat pump operation profiles, and the effect of different profiles on output demands was analysed. The pre-defined profiles were based on real world heat pump monitored data. The pre-defined operation profiles included were bimodal (similar to a gas boiler profile), daytime and continuous. The scenarios which assumed bimodal and daytime operation profiles for heat pumps resulted in the highest peaks, with the bimodal pattern giving the highest. The continuous profile gave a lower peak and lower total consumption.
More analysis will be done on the sensitivity of these profiles to the assumptions identified above and the effect of flexibility and to determine what profile distribution most heat pump operations fall into. More detailed analysis will also be performed to determine the impact of this geospatially and considering flexibility through pre-heating.
