News & Analysis
/
Article

Restoring wind-induced responses of supertall buildings

AUG 02, 2024
Machine learning model can reconstruct missing displacement measurements of supertall buildings during typhoons, enabling real-time structural health monitoring.
Restoring wind-induced responses of supertall buildings internal name

Restoring wind-induced responses of supertall buildings lead image

Monitoring the displacement of supertall buildings during typhoons or other extreme wind conditions is essential for ensuring these structures remain safe and serviceable. However, severe wind may cause visual obstructions and data transmission interruptions that render traditional displacement monitoring techniques, such as GPS, interferometric radar, and sensor-based methods, unusable.

Zhou and Li developed an approach to reconstruct unmeasured displacements of supertall buildings during windstorms. The method uses a physics-informed machine learning model known as Long Short-Term Memory to estimate the missing strain responses of structures, then combines these with limited field measurements to reconstruct displacement measurements.

The authors verified their method by reconstructing the displacement of an actual supertall building during a super typhoon. It can be easily integrated into existing monitoring systems and minimizes reliance on extensive sensor networks, allowing real-time monitoring of structural deformations during windy conditions.

“The significance of this study lies in its ability to accurately reconstruct displacement in extreme wind conditions while reducing sensor dependency, costs, and complexity, providing a more reliable and economical solution for structural health monitoring,” said author Qi Zhou.

This approach can also be used in engineering practice to provide valuable references and support for future wind-resistant design of supertall buildings. The authors plan to integrate this approach with physics models of actual supertall buildings to create digital twins of these structures.

“The integration of machine learning with structural monitoring not only has significant implications for civil engineering but also opens up new avenues for interdisciplinary applications,” Zhou said. “We are excited to explore these possibilities and expand the application scope of this method.”

Source: “Reconstruction of displacement responses of a supertall building during typhoons based on limited field measurements and a physics-informed machine learning model,” by Qi Zhou and Qiu-Sheng Li, Physics of Fluids (2024). The article can be accessed at https://doi.org/10.1063/5.0222336 .

This paper is part of the Flow and Civil Structures Collection, learn more here .

More Science
APS
/
Article
/
Article
Improving terrestrial gravitational wave detection with “multi-loop” geometry
/
Article
Advancements and potential research directions in multimodal skin-like sensors
/
Article
First known mountain ridge impact found at Zhangguang Cailing
/
Article
Using magnetic resonance velocimetry facilitates precise structural modifications based on flow characteristics and turbulent kinetic energy