TY - JOUR
T1 - Evaluating building-level parameters for lower-temperature heating readiness
T2 - A sampling-based approach to addressing the heterogeneity of Dutch housing stock
AU - Wahi, Prateek
AU - Konstantinou, Thaleia
AU - Visscher, Henk
AU - Tenpierik, Martin J.
PY - 2024
Y1 - 2024
N2 - The Dutch government aims to eliminate natural gas for residential heating in 1.5 million homes by 2030. One strategy is connecting existing dwellings to lower-temperature district heating (DH) systems, although these dwellings might require energy renovations. The heterogeneous dwelling stock causes varying renovation needs that complicate the energy transition. The present study addresses this issue by assessing the building-level parameters affecting the readiness of the Dutch terraced-intermediate and apartment types for lower-temperature heating (LTH) supplied by DH systems. A sampling-based approach was employed to capture variability within these dwelling types, addressing the limitations of archetype-based methods. The findings suggest a sample size of 1300 to represent the variations in these dwelling types. Parametric simulations and machine learning methods were used to identify significant building-level parameters for medium-temperature (MT: 70/50 °C) and low-temperature (LT: 55/35 °C) supply levels. These include heating setpoints (desired indoor temperature) and ventilation-related parameters (ventilation system type and air infiltration rate), followed by fabric-related parameters (roof, glazing, wall, ground, and door insulation) and geometric properties (orientation, compactness ratio, and window-to-wall ratio). Additionally, radiator oversizing also impacts LTH readiness. These results broadly apply to the studied dwelling types, although feature importance varies by supply temperature and dwelling type. The findings can guide stakeholders in assessing current conditions and prioritising renovation measures, aiding the development of targeted renovation solutions. Encompassing the representative variations within studied dwelling types enhances the robustness of the results. However, incorporating more refined data could improve the accuracy of the findings, better supporting the energy transition of these dwellings.
AB - The Dutch government aims to eliminate natural gas for residential heating in 1.5 million homes by 2030. One strategy is connecting existing dwellings to lower-temperature district heating (DH) systems, although these dwellings might require energy renovations. The heterogeneous dwelling stock causes varying renovation needs that complicate the energy transition. The present study addresses this issue by assessing the building-level parameters affecting the readiness of the Dutch terraced-intermediate and apartment types for lower-temperature heating (LTH) supplied by DH systems. A sampling-based approach was employed to capture variability within these dwelling types, addressing the limitations of archetype-based methods. The findings suggest a sample size of 1300 to represent the variations in these dwelling types. Parametric simulations and machine learning methods were used to identify significant building-level parameters for medium-temperature (MT: 70/50 °C) and low-temperature (LT: 55/35 °C) supply levels. These include heating setpoints (desired indoor temperature) and ventilation-related parameters (ventilation system type and air infiltration rate), followed by fabric-related parameters (roof, glazing, wall, ground, and door insulation) and geometric properties (orientation, compactness ratio, and window-to-wall ratio). Additionally, radiator oversizing also impacts LTH readiness. These results broadly apply to the studied dwelling types, although feature importance varies by supply temperature and dwelling type. The findings can guide stakeholders in assessing current conditions and prioritising renovation measures, aiding the development of targeted renovation solutions. Encompassing the representative variations within studied dwelling types enhances the robustness of the results. However, incorporating more refined data could improve the accuracy of the findings, better supporting the energy transition of these dwellings.
KW - Energy transition
KW - Heating decarbonisation
KW - Energy renovations
KW - Parametric simulations
KW - Machine learning
UR - http://www.scopus.com/inward/record.url?scp=85202343181&partnerID=8YFLogxK
U2 - 10.1016/j.enbuild.2024.114703
DO - 10.1016/j.enbuild.2024.114703
M3 - Article
SN - 0378-7788
VL - 322
JO - Energy and Buildings
JF - Energy and Buildings
M1 - 114703
ER -