Energy-saving potential prediction models for large-scale building: A state-of-the-art review

Energy-saving potential prediction models play a major role in developing retrofit scheme. Reliable estimation and quantification of energy saving of retrofit measures for these models is essential, since it is often used for guiding political decision-makers. The aim of this paper is to provide up-to-date approaches of predicting energy-saving effect for building retrofit in large-scale, including data-driven, physics-based, and hybrid approaches, while throwing light on workflow and key factors in developing models. The review focuses on pointing out pivotal aspects that are not considered in current models of predicting energy-saving effect for building retrofit in large-scale. It is concluded that the validation of proposed models mainly focuses on an aggregated level, which makes it ignore performance gap differences between buildings. The models exist the problem of prebound-and rebound effects due to uncertainty factor. Occupant's willingness to retrofit is ignored in all three categories of models, which can lead to the prediction result deviate from the actual situation in a certain extent. This paper promotes the development of models for predicting energy-saving potential for large-scale buildings, and help to formulate appropriate strategies for the retrofit of existing buildings.

Automated summary

This paper provides up-to-date approaches for predicting energy-saving effects in large-scale building retrofit, including data-driven, physics-based, and hybrid approaches. It highlights several key issues in current prediction models, such as ignoring performance differences, prebound and rebound effects, and occupant willingness to retrofit. The study is of great importance in promoting the development of energy-saving potential prediction models and formulating appropriate retrofit strategies for large-scale buildings.