Aerodynamic Shape Optimization of Double Façades to Mitigate Wind-Induced Effects on Tall Buildings

Michael Huntley, Dr. Alice Alipour, and Dr. Mohammad Jafari. Department of Civil Engineering, Iowa State University, 813 Bissell Rd, Ames IA 50011

The growing urbanization worldwide and advancement in construction technology has led to an increase in developing taller and more slender high-rise buildings. These high-aspect-ratio structures are especially susceptible to wind loading, as the wind speed increases exponentially with the height. Wind loading can induce vibrations in the building, which can cause severe damage to both structural and non-structural components of the building, as well as occupant discomfort. So far, many strategies have been tested to mitigate wind loading and the effects of the wind-induced vibrations. These approaches include aerodynamic modifications, such as corner chamfering and rounding, auxiliary damping devices, and structural design adjustments. Until very recently, double skin facades have been researched and utilized primarily for energy saving applications. Recent studies prove that double-skin facades can considerably reduce the wind loads exerted on tall buildings and such structures can modify the building’s aerodynamic properties. This project seeks to explore smart morphing double-skin façade systems as a possible solution to the wind-induced vibration problem using computational fluid dynamic (CFD) technique to simulate the response of various double-skin façade designs under numerous wind loading conditions. This research focuses on machine learning applications in predicting flow characteristics around tall buildings based on a CFD database simulated for the elliptical, triangle, and rectangular buildings. For this purpose, the convolutional and deconvolutional neural networks are supposed to be used for predicting the pressure and velocity contours along with aerodynamic coefficients at different Reynolds numbers, angles of attack, and aspect ratios of the building’ cross-section. Such a framework helps to come up with better aerodynamic shapes for double façade or even building in the design stage. Additionally, this open-source prediction model can be employed to optimize the overall shape providing the least wind-induced load and response.

Additional Abstract Information

Presenter: Michael Huntley

Institution: Iowa State University

Type: Poster

Subject: Civil Engineering

Status: Approved

Time and Location

Session: Poster 4
Date/Time: Tue 11:00am-12:00pm
Session Number: 3631