Key points are not available for this paper at this time.
The rise in greenhouse gas emissions in cities and the excessive consumption of fossil energy resources has made the development of green spaces, such as green roofs, an increasingly important focus in urban areas. This study proposes a novel smart energy-comfort system for green roofs in housing estates that utilises integrated machine learning (ML), DesignBuilder (DB) software and Taguchi design computations for optimising green roof design and operation in buildings. The optimisation process maximises energy conservation and thermal comfort of the green roof buildings for effective parameters of green roofs including Leaf Area Index (P1), leaf reflectivity (P2), leaf emissivity (P3), and stomatal resistance (P4). The optimal solutions can result in a 12.8% increase in comfort hours and a 14% reduction in energy consumption compared to the base case. The ML analysis revealed that the adaptive network-based fuzzy inference system is the most appropriate method for predicting Energy-Comfort functions based on effective parameters, with a correlation coefficient greater than 97%. This novel smart framework for the optimal design of green roofs in buildings offers an innovative approach to achieving energy conservation and thermal comfort in urban areas.
Mousavi et al. (Thu,) studied this question.