Energy analyses of dwelling stocks combine a stock model and an energy model. The stock model describes the stock size, composition and renovation status, whereas the energy model describes the average energy intensities of the various segments of the stock and assumed energy savings obtained when dwellings are renovated. Dwelling stock models that include the renovation status of the dwelling stock enable energy analyses of the stock to inform policy.
The all-encompassing disaggregated thermophysical input data required to effectively inform residential stock energy consumption models are computationally intensive and have relied traditionally on laborious manual data analysis. Since it has been impractical to model every single building, it is normal to define a set of reference dwellings (RDs) that are representative of typical national or regional building. RDs are used to produce overall energy saving extrapolations.
Driven by policy to reduce domestic energy use, a wide range of bottom-up building stock models representing dwelling stocks have been developed.
Policy that seeks to reduce domestic energy use is driving rapid change in the sector, TU Dublin has established that significant retrofits have taken place in the sector (see Figure 1 and Table 1), resulting in approximately 60% of the existing Irish housing stock being well insulated in 2014 (see Table 2). The level of retrofit is significantly higher than is assumed by policy. This is attributed to energy stock models lagging the renovation wave with stock models often being created using RD’s characterised on as-built or default values classified by construction period leading to stock models lacking validity.
The rate and number of thermal retrofits, ‘the state of a stock’ is captured contemporaneously within national Energy Performance Certificate (EPC) databases. A rapid, robust and automated process to extract information from these datasets is necessary to keep pace with the rate of renovation and to track the effectiveness of policy.
Current research in TU Dublin is defining the conditions for a validated EPC dataset and automating methods for cleaning an EPC dataset to an acceptable level before using machine learning and unsupervised clustering methods to derive objectively, contemporaneous RDs characterising a national stock model. The generalisable methodology created allows for the rapid creation of contemporaneous stock models informing real-time decision-making, providing new insight to match the pace of renovation while also allowing quick ‘what if’ analyses.
This research, in the context of Irelands residential climate action targets will:
- Model energy demand and supply of RD’s using dynamic simulation programmes to develop new knowledge to inform policy
- Specifically, model RD’s in order to facilitate;
- the identification of sensitive parameters important to overall performance,
- through changing such parameters, forecasting the consequences of specific scenarios or policy-interventions,
- policy-makers in preparing substantive arguments for particular retrofit interventions and contemporaneous insight-driven policies.
- Through first using the dynamic models created for use at RD or end-use level and then aggregating upwards, to observe, analyse and inform energy use characteristics at stock level.
- Advise on realisable targets along with specific and tailored policy interventions for the Irish Housing stock.
A rapid, robust and automated process to extract information from EPC datasets to create a real-time stock model for Irish housing is being developed by TU Dublin. This research proposes to use this information to energy model Irelands housing stock to develop new knowledge to inform policy. Analysis will be undertaken in the context of Ireland’s climate action plan targets. The doctoral candidate will be working closely with other researchers in this area.
Minimum of a 2.1 honours degree (level 8) in a relevant discipline
Scholarship not available. Fees & Materials to be paid by the student. Materials costs may be very significant
If you are interested in submitting an application for this project, please complete an Expression of Interest.
Applications submitted without an EOI form will not be considered.