A Multiobjective Deep Learning Solution for Optimizing Cooling Rates of Urban Courtyard Blocks in Hot Arid Zones

Abstract

In response to the rapid urbanization and housing demands, there has been a shift from traditional courtyards to multi-story city structures. Unfortunately, this transition can significantly affect the local climate and overall comfort due to increased heat. To overcome these challenges, our proposed approach suggests implementing multi-objective optimization techniques to strike a balance between various competing goals. These goals may encompass outdoor thermal comfort, energy efficiency, and urban sustainability when designing urban courtyard blocks. This study has many potential benefits for sustainable living and aligns with several Sustainable Development Goals (SDGs) like Energy Efficiency (SDG 7 -Affordable and Clean Energy), Sustainable Cities and Communities (SDG 11 -Sustainable Cities and Communities) and Good Health -Well-being (SDG 3 -Good Health and Well-being). The outcomes from this paper will help reduce the effects of climate change by making a positive contribution to sustainable development. This research aims to anticipate the cooling load per unit area (cooling/m 2 ) of buildings in hot arid zones based on building features such as overall height, orientation, and other considerations of buildings. The deep learning algorithms used are MLP Regressor, RNN LSTM, and RBFN. This research aims to create a model to properly forecast cooling load per unit area and provide insights into the best building design for lowering cooling loads in hot arid zones. RBFN outperformed MLP Regressors and RNN LSTM in forecasting cooling rates in urban courtyard blocks, according to the findings.