image: Lehigh University PhD student Saskia Putri (civil and environmental engineering), Farrah Moazeni, assistant professor of civil and environmental engineering, and Javad Khazaei, assistant professor of electrical and computer engineering, P.C. Rossin College of Engineering and Applied Science. The team is collaborating with Siemens to advance microgrid technology for data centers.
Credit: Lehigh University
Hyperscale data centers are behemoths of computing power. Spanning over a million square feet in a single facility alone, they house thousands of servers that process, store, and analyze massive datasets that drive artificial intelligence.
Not surprisingly, such infrastructure also requires an enormous amount of power.
“Data centers have distinctive load profiles,” says Saskia Putri, a fourth-year Lehigh University PhD student advised by Farrah Moazeni, an assistant professor of civil and environmental engineering in the P.C. Rossin College of Engineering and Applied Science. “When they’re training AI models, their power requirements can jump suddenly and significantly. That kind of rapid change can create instability in the electrical grid.”
To tackle this problem, Siemens sponsored a one-year research project led by Farrah Moazeni and co-led by Javad Khazaei, an assistant professor of electrical and computer engineering. As part of this agreement, Putri was awarded a year-long fellowship to work with researchers at Siemens on advancing microgrid technology for more resilient and reliable energy systems. The agreement stems from a new research collaboration between Siemens—a global leader in automation and energy systems—and Lehigh’s Center for Advancing Community Electrification Solutions (ACES).
Under the Siemens-Lehigh research agreement, one of Putri’s tasks is to develop a data center microgrid—a localized power network capable of operating independently of the main grid. Her work will focus on modeling, simulating, and embedding real-time control systems to enhance the microgrid’s stability and reliability in powering energy-intensive operations such as AI training.
“In addition to real-time control, we also aim to build a real-time monitoring tool that lets users know when a disturbance or maintenance issue is likely to occur,” she says. “The idea is to design a system that is reliable and efficient and can predict and respond to changes instantly, ensuring rapid voltage stabilization under dynamic conditions.”
Throughout the fellowship, Putri will make several visits to Siemens’ research and development department to refine her designs. She will also use resources from Moazeni’s Interconnected Critical Infrastructure Systems Engineering (CONCISE) Laboratory and the Integrated, Resilient, and Intelligent Energy Systems (INTEGRITY) Laboratory, led by Javad Khazaei. Lehigh’s own data center will serve as a testing ground for Putri’s models, allowing her to validate simulations using real operational data.
The collaboration underscores the value of mutually beneficial university-industry partnerships, a theme highlighted during a recent national workshop on industry-academia collaboration hosted by Lehigh in Washington, D.C.
For Siemens, which aims to incorporate new control and optimization algorithms into its software and hardware platforms, partnering with Lehigh bridges academic innovation and industrial application, and promises access to promising technologies.
“They focused on continuously improving their monitoring and automation tools across diverse industrial applications,” says Putri. “By collaborating with universities like Lehigh, they can bring in new ideas and research that haven’t yet been implemented in the real world.”
Such advances could translate to significant energy savings. According to the International Energy Agency, data centers already account for about 1.5 percent of global energy consumption—a figure expected to climb as AI demand grows.
“That's why we want to use a microgrid instead of relying solely on the main grid,” says Putri. “We plan to incorporate solar panels, wind turbines, and energy storage systems. By combining renewable generation with storage, we can save energy and reduce emissions.”
For Putri, the partnership offers an opportunity to address the unique challenges of modeling hyperscale data center microgrids at a time when the world is increasingly reliant on artificial intelligence.
“The design of these microgrids is not typical,” she says. “The load profile is different, the scale is massive. How can I automate that? How can I combine AI with microgrid design? These things have never been done before in the industry. I’m fascinated by how algorithms can control and optimize infrastructure to make it more resilient and sustainable—and that’s what makes this partnership so exciting.”
The experience will also give her a glimpse into her potential future. Still weighing a career in academia versus industry, Putri sees the fellowship as a chance to explore what it’s like to pursue her passion within a global company.
“When I think about going into industry, I think about companies like Siemens that prioritize research and development,” she says. “My passion is research—and if we can use what we’re learning here to make data centers greener and grids more stable, that would be amazing.”
Related Links:
- Lehigh University: Center for Advancing Community Electrification Solutions (ACES)
- Rossin College Faculty Profile: Farrah Moazeni
- Interconnected Critical Infrastructure Systems Engineering (CONCISE) Laboratory
- Rossin College Faculty Profile: Javad Khazaei
- Integrated, Resilient, and Intelligent Energy Systems (INTEGRITY) Laboratory
- Lehigh News: "Lehigh Hosts National Workshop on Rethinking PhD Training in Washington, D.C."
- Lehigh University: Institute for Cyber Physical Infrastructure and Energy
- I-CPIE News: "Doctoral Student Saskia Putri Optimizes Interconnected Water and Power Systems"