Abstract
Accurate prediction of building energy need plays a fundamental role in building design, despite the high computational cost to search for optimal energy saving solutions. An important advancement in the reduction of computational time could come from the application of machine learning models to circumvent energy simulations. With the goal of drastically limiting the number of simulations, in this paper we investigate the regression performance of different machine learning models, i.e., Support Vector Machine, Random Forest, and Extreme Gradient Boosting, trained on a small data-set of energy simulations performed on a case study building. Among the XX algorithms, the tree-based Extreme Gradient Boosting showed the best performance. Overall, we find that machine learning methods offer efficient and interpretable solutions, that could help academics and professionals in shaping better design strategies, informed by feature importance.
Highlights
Indoor heating and cooling of buildings are among the most energy consuming activities in Europe [1] and a few important laws, acts and regulations aim at reducing their environmental impacts [2]
Among the most relevant ones, we find building energy simulations, characterizing different scales [4] and in different locations [5,6]
We show how Machine Learning (ML) feature importance can rank the impact of the different building variables on the energy need
Summary
Indoor heating and cooling of buildings are among the most energy consuming activities in Europe [1] and a few important laws, acts and regulations aim at reducing their environmental impacts [2]. A variety of approaches try to address this issue, such as the creation of new energy saving materials, the imposition of progressively stricter requirements, and the realisation of more efficient systems and equipment [3]. Simulate different building configurations, returning precise and accurate results that take into account all variables affecting building energy consumption for the indoor climate control (e.g., envelope materials, orientation, systems, weather data). Even though a single simulation is relatively fast (taking from a few seconds up to a couple of hours depending on the modelled building and the adopted computer), each simulation requires an operator who inserts, analyses, assesses, makes decisions, leading to a time consuming process that needs constant human supervision to avoid unintended effects [8,9,10]
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call