Image: Michael Erhardsson/Mostphotos
Image: Michael Erhardsson/Mostphotos
Research project The project will contribute to the transition to a sustainable energy system by offering single-family house owners access to specific information on energy saving for energy renovation strategies. With machine learning algorithms that extract case-specific knowledge from Swedish Energy Performance Certificates, GIS, and Swedish National Land Survey, a database will be established with alternative energy renovation solutions. The method will be validated with data from single-family houses.
The project is financed by Formas.
Since the mid-1970s, sustainable building construction has mainly focused on energy efficiency in the user phase. Sweden, which early gained a world-leading position in sustainable construction, can, for example, refer to that the idea for the so-called passive houses can be traced back to Professor Bo Adamsson's research activities at Lund University in the 1980s. Thanks to the strategic use of sustainable construction solutions for several decades, it can now be stated that the energy efficiency in today's newly produced buildings is significantly better than before.
Despite this, the settlement process still uses a significant proportion of society's energy. It takes so long to reduce the buildings' use of energy because new ones replace few existing buildings. Therefore, fast and efficient methods are required to renovate existing buildings. For a long time, research and development in the public construction sector have focused on new production. Renovation, refurbishment, and extensions have been left behind. In Sweden, Formas has recently funded a national research collaboration on the renovation. It has made essential contributions to renovation research. While mainly commercial and multifamily buildings of various kinds have been examined, renovation of single-family houses is still a relatively unexplored area. Also, the single-family house sector is challenging to convert. One reason is that homeowners are a very inhomogeneous group that has completely different conditions and interests.
The purpose of this project is to provide decision support for single-family house owners in energy renovation. A data-driven model with Big Data will be developed. It shall be based on information from the National Board of Housing, Building and Planning's database with Swedish Energy Performance Certificates, Geographic Information System, and Swedish National Land Survey. Information management is based on a data-driven model. The model consists of machine learning algorithms, which extract case-specific knowledge based on technical solutions and actual energy use. In this way, the developed methodology, which becomes open-access, gives single-family house owners access to specific information on energy savings for various energy renovation methods. The method will be validated by comparing collected data from real single-family homes with the methodology's predictions.
By reaching a large part of the single-family house owners with user-friendliness and open access, the project actively supports a transition to a more sustainable energy system. Workshops will be organized that can enable the dissemination of results to relevant stakeholder groups. The project is based on Swedish conditions, but methods and results will be possible to transfer and apply at an international level.
