Enhancing Building Sustainability through Aerodynamic Shading Devices: An Integrated Design Methodology Using Finite Element Analysis and Optimized Neural Networks

Abstract

In the quest for sustainable building solutions, attention has increasingly turned to innovative structural designs that minimize environmental impact. Our study focuses on integrating advanced sun-breakers with aerodynamic features to enhance building sustainability through optimized shading device design. We analyze recent advancements in shading devices, specifically optimizing airflow for improved aerodynamic performance. Given the computational intensity of Finite Element Analysis (FEA), we employ an Artificial Neural Network (ANN) based on Stochastic Gradient Descent (ANN-SGD) to address these challenges. Our predictive model estimates key fluid flow variables, including wind velocity, static pressure, air density, and turbulent kinetic energy. Initially, short-run workbench simulations are conducted to validate the ANN-SGD model, which demonstrated excellent regression accuracy with a Root Mean Square Error (RMSE) of 4.66%, a Coefficient of Determination (R²) of 0.995, and a Coefficient of Variation of RMSE (CVRMSE) of 8.33%. These findings contribute valuable perspectives to the future of sustainable building design, offering enhanced methodologies for aerodynamic shading devices.