Most relevant characteristics to improve residential energy efficiency

Abstract

Urban areas play a crucial role in global energy demand and in policies to mitigate climate change. Energy use in residential buildings is one of the main sources of greenhouse gas emissions in cities. Spain has more than 25.8 million dwellings, of which around 9.3 million are more than 50 years old. Of these, 16.4% are in a poor or deficient state of conservation (INE, Population and Housing Census 2011). This situation of obsolescence and poor conservation of the building stock requires a major structural, functional and energy refurbishment. As a research hypothesis, it is proposed that the current Spanish building stock has a great potential for energy savings and CO2 reduction and that with an appropriate selection of interventions aimed at refurbishment it is possible to increase the energy efficiency of buildings and make them more sustainable. The main objective proposed is to identify the architectural, design and building system or installation factors that are most relevant for improving the energy efficiency of existing residential buildings in Spain. A large database with geo-referenced observations of housing energy certificates is used for this purpose. The certificates correspond to individual dwellings located in blocks of buildings in the city of Barcelona (C2 climate zone according to the CTE). From these certificates, information on energy consumption and CO₂ emissions, as well as the technical characteristics of the envelope and heating/cooling systems, can be obtained for each dwelling. The dataset contains information on energy consumption and CO₂ emissions with their rating (energy letter); in addition to the characteristics of the buildings such as the surfaces and transmittances of the envelope (opaque and glazed enclosures), the climate zone, type of dwelling, year and building code, heating and air-conditioning installations, percentage of windows according to orientation, etc. Using a linear regression model, the influence of each housing characteristic on energy consumption and CO₂ emissions can be estimated. The interpretation of the regression coefficients allows determining to what extent energy consumption can be reduced, for example by improving the envelope (opaque or glazed), systems/installations, thermal bridges, among other aspects. The linear regression analysis has shown promising results due to its reasonable accuracy for model estimation, and the relative simplicity of its application and interpretation compared to other methods (Fumo et al., 2015).