How can we harness wind and improve airflow to benefit society?
For millions of people around the world the topic of airflow and ventilation has taken on an increased urgency over the past several years. You might even be wondering right now how you can cool the warm room you’re sitting in without contributing to climate change. Or how the airflow in that room is affecting your health and that of those around you.
Studying these issues, and much more, is Catherine Gorlé, an assistant professor of civil and environmental engineering at Stanford, and the founder of Stanford’s Wind Engineering Lab. Her research focuses on how wind affects sustainability and resiliency of buildings and cities. For example, she studies how to sustainably improve ventilation in homes and buildings, and how we can harness the wind to provide other benefits for society.
To find out, she uses a type of modeling tool called computational fluid dynamics, which uses high-performance computers to study fluids in motion. And as her results make their way into the hands of designers, policymakers, and fellow scientists and engineers, she also hopes to work with researchers who study human behavior to nudge people into adopting habits that benefit the planet and people’s health. “We need to understand how to change behavior; for example, getting people to open their windows at the right time.”
Gorlé first became involved with this area of research in 2005 when she worked on bridge aerodynamics. In 2002 and 2005, respectively, she received her BSc and MSc degrees in aerospace engineering from the Delft University of Technology in the Netherlands, and in 2010 her PhD from the von Karman Institute for Fluid Dynamics in cooperation with the University of Antwerp. After postdoctoral fellowships at Stanford and then positions at the von Karman Institute and Columbia University, she returned to Stanford in 2016 to focus on sustainable design in buildings and in urban areas.
Because wind and ventilation are universal, Gorlé’s work addresses a diverse set of problems. Along with her students and postdocs, she is developing models to understand everything from how wind flows impact the dispersion of pollution and air quality in urban areas to wind loading on buildings during extreme wind events, and predicting how natural ventilation can be used to cool buildings without using air conditioning and to reduce the concentration of airborne pathogens, including the COVID-19 virus.
A different methodology
Her methodology differs from many scientists and engineers who study wind and airflow. Traditionally, researchers have used wind tunnels, placing models of a structure, such as a building, in a tunnel and subjecting it to wind generated by fans. As the air flows past, through, and against the model, they can then see how different design elements impact the building’s resilience and how it will fare when faced with extreme winds.
When computer simulations first came along, research focused on validating the simulations against these wind tunnel experiments. This kind of validation has resulted in many model improvements, but it also has limitations. For starters, wind tunnels are only a model of reality, and it is difficult to fully capture the variability and uncertainty in the wind conditions a structure will be subjected to.
By contrast, Gorlé uses computational fluid dynamics (CFD) in combination with uncertainty quantification (UQ). CFD combines math, physics, and computers to identify solutions to issues related to the flow of fluids like wind. (When we think of the word “fluid,” we may automatically picture liquids like water, but when in movement, air also behaves like a fluid.) UQ aims to quantify the uncertainty in the solutions related to the variability and uncertainty in real urban wind flow. Gorlé says that the combination of UQ and CFD enables engineers to play with all the parameters of a scenario and look at their effects.
For instance, when testing how a pollutant disperses in the air, CFD models allow engineers to put the source of pollutants in different locations of a structure and then see how that impacts the concentration of pollutants inside. “The amount of information you can get from CFD is a huge benefit,” she says.
CFD does have drawbacks. Current modeling methods are computationally expensive, and even when using high-performance computing clusters, a simulation can take days to weeks, depending on how much detail is included. An ongoing issue is building confidence in CFD models. Wind tunnels have been around for more than a century; the first was used in 1871. By comparison, CFD is a fresh-faced newcomer full of promise. But Gorlé and her lab are working to show that this promise is warranted by perfecting CFD models and validating them with field experiments.
“I strongly believe that at some point these models are going to replace a significant fraction of wind tunnel experiments,” says Gorlé, “but they need to become more affordable, and we need to build confidence in all of the physics models that have to go in them.”
Exploring natural ventilation
While CFD models haven’t replaced wind tunnels just yet, Gorlé is currently using CFD to evaluate natural ventilation as a way to lower the incidence of respiratory diseases in informal settlements and to recommend design solutions to boost natural ventilation. In 2017, she teamed up with Stephen Luby, a Stanford University professor of medicine, to develop a CFD model to test ventilation strategies for homes in informal settlements in Dhaka, the capital of Bangladesh; in the South Asian country about 22% of deaths among children under the age of 5 are due to respiratory diseases like pneumonia.
Luby’s previous work suggested that increasing ventilation in homes could reduce the prevalence of pneumonia. While many of the solutions to increase ventilation, such as opening a window, may seem intuitive, Gorlé’s CFD models added nuance to these solutions, such as figuring out exactly how much airflow is created by different window configurations, or how effective a design solution such as a skylight is for ventilation.
An added complexity of this modeling work is that, in informal settlements, the structure of homes and how people use their homes are highly variable. But using measurements of temperature and tracer concentration decay collected from homes in Dhaka, which involved injecting a tracer gas into the homes and recording the dilution of this tracer over time, Gorlé and her lab were able validate the CFD model they had built. “Now we can take that model and apply it to different settings and have confidence in the fact that it should be accurate because the physics are the same.”
As the research in Dhaka wraps up in 2022, Gorlé is already working on other projects related to natural ventilation and health, including work in refugee camp emergency health care facilities. What will she focus on after that? Wherever the wind takes her.