Grant funds research that could pave the way for new skyscraper designs

San Francisco’s skyline, showing a skyscraper under construction

Funding will support student researchers using machine learning and wind tunnels

What do you see when you look at a city skyline? If you looked through the eyes of Associate Professor of Civil Engineering Zhaoshuo Jiang, you’d see sameness — and potential. Jiang was recently awarded a $285,491 grant from the National Science Foundation for research that could open up brand new possibilities for our skylines by flipping the way skyscrapers are designed on its head.

“We’re trying to find the most optimal shapes that we can use to get around the wind load,” Jiang said. “With that, we’ll be able to build taller and more exciting buildings.”

Wind, Jiang explains, is a skyscraper’s greatest enemy. The taller the building, the more wind constrains the building’s design, so the development of a skyscraper often begins with a very small set of tried-and-true shapes that are able to hold up to wind gusts. Only much later in the development process will architects send the building’s design to engineers so it can be tested using building models and wind tunnels. By collaborating with architects and engineers on that testing phase earlier in the process, Jiang and his collaborators hope to open up a vast swath of new potential architectural designs and shapes that might never have been considered otherwise.

In the system the researchers developed to accomplish that, a robot arm builds a model skyscraper in a wind tunnel, which collects data on the building’s performance. That data gets fed into a computer, which incorporates the new information and instructs the robot arm to build a differently shaped model that might perform better, starting the cycle over again. The system combines the speed and analytical power of machine learning with the realism of wind-tunnel testing, automating what would typically be a time- and labor-intensive process.

“As far as we know, this is the first effort to marry physical testing with intelligent computing and robot arms in the wind tunnel,” Jiang said. “What we’re trying to do is to take advantage of the pros of all these different approaches and marry them together to give us a tool to drive our future designs.”

But there’s no point in testing designs that architects won’t use. To avoid wasting time on dead-end designs, Jiang used his past experience in industry to forge a partnership with Skidmore, Owings & Merrill LLP, an influential engineering and architectural firm that’s responsible for notable skyscrapers like Dubai’s Burj Khalifa, the world’s tallest building. The researchers will also work with a government organization, the National Institute for Standards and Technology, so their findings can be translated into codes and standards that govern skyscraper construction.

Those partnerships also have a second purpose: to give students the opportunity to collaborate and make connections with professional engineers in the workforce. Both graduate and undergraduate students will have chances to get involved with the project, starting this summer with the Research Experience for Undergraduates program Jiang runs. Student researchers will also have the opportunity to take part in wind-tunnel tests at the University of Florida, home to the grant’s second lead researcher, Associate Professor of Civil & Coastal Engineering Brian Phillips, as well as one of the largest wind tunnel testing facilities of its kind.

Jiang says the project will allow the researchers to deepen ties between SF State and industry and offer more opportunities for students. As for what new and strange architectural designs they might stumble upon? That’s still an open question.

“I’m always joking with my students: ‘Later on, if you see flower-shaped buildings, you’ll know this project was a success,’” Jiang said. “Currently, we explore a tiny, tiny bit of the universe. But now, this is an opportunity for us to open our eyes and look at the bigger picture.”