In addition to Wang, the fellowship was awarded to Gabe Mantegna, a graduate student in mechanical and aerospace engineering.

The Maeder Graduate Fellowship is awarded to one or two graduate students each year who demonstrate strong potential to develop solutions for a sustainable energy and environmental future. The fellowship is supported by the Paul A. Maeder ’75 Fund for Innovation in Energy and the Environment and will cover the students’ tuition and stipend for the 2025–2026 academic year.

As the rise of artificial intelligence drives a surge in energy-intensive data centers, delivering electricity efficiently from the grid to the computing chips at the heart of those data centers has never been more important.

Wang, a graduate student in electrical and computer engineering, is working to make critical components of those power delivery systems more energy efficient. In collaboration with his adviser Minjie Chen, an associate professor of electrical and computer engineering and the Andlinger Center for Energy and the Environment, Wang has focused much of his graduate career on developing better tools for modeling power magnetics.

“Magnetics are a part of every single power delivery system,” said Wang. “At the same time, they take up a lot of physical space within the overall power conversion unit, and they are a major contributor to the energy losses that happen during the conversion process.”

While improvements to other components of power delivery systems have yielded higher efficiencies and smaller sizes, making improvements to magnetics has often been challenging. For example, the Steinmetz equation, which is used to estimate power losses in magnetic materials, was developed over 100 years ago for devices operating in vastly different conditions from today.

For his project, Wang has assembled a database of over a million data points about the characteristics of today’s power magnetic materials. Using machine learning tools, he has developed models from that database to predict the characteristics and effectiveness of power magnetics under different operating conditions.

A major priority of Wang’s work is ensuring his research is open-source and available as a resource for other researchers from academia and industry.

“Our code is open source. Our database is open source,” Wang said. “We really want the entire international community to join us, so that we can work together to expedite the push for better power magnetics models.”

In a similarly collaborative spirit, Wang is a member of the organizing committee for the IEEE PELS-Google-Enphase-Princeton MagNet Challenge, an international power magnetics modeling competition that brings together researchers from a wide range of disciplines to develop better software for predicting magnetic characteristics.

Now in its second iteration, the competition is challenging 40 research teams to model the characteristics of power magnetics materials under transient conditions, such as those that occur when power is quickly switched on or off to a device. Wang said understanding how magnetics perform in these circumstances is critical for building more efficient power delivery systems, as nearly all circuit design models are built to simulate transient conditions.

“If we can develop a better model for magnetic materials, then it would be a huge step forward for magnetics designers,” said Wang. “We could potentially reduce the everyday computer charger’s size by 30 to 40%.”

Wang said the Maeder Graduate Fellowship will allow him to continue making connections in the research community, with the goal of having his research used as much as possible by power electronics engineers.

“It’s important to me that my research can be applied to real-world problems. That’s really what motivates me most about the work I’m doing at the Andlinger Center and Princeton,” Wang said.