Berkeley IEOR Professor Javad Lavaei, in partnership with his former postdoctoral scholar, now Assistant Professor at Virginia Tech, Ming Jin, has been awarded an $800,000 grant from the National Science Foundation for their pioneering project, “Safety under Distributional Shift in Learning-Enabled Power Systems.” This project is poised to transform the design of learning-enabled, safety-critical systems, with particular emphasis on power systems.
In an era marked by rapid environmental and technological shifts, power systems face unprecedented challenges. Safeguarding the operation of these systems requires meticulous adherence to mathematical models that capture the interplay of parameters such as voltage, frequency, and equipment health. However, in an increasingly unpredictable environment, ensuring end-to-end safety has become a formidable task. Lavaei and Jin’s project acknowledges this challenge, striving to bolster power systems with improved safety and resilience.
This project introduces the concept of ‘antifragility,’ which views change not as a disruption but as an opportunity for system enhancement. Lavaei and Jin’s project embraces this new perspective as crucial for navigating the distributional shifts in our rapidly evolving environment. This transformation will be achieved by developing proactive, memory-based antifragile systems, exploring the potential of multi-agent systems, exploring multi-agent systems for cooperative decision-making, and applying advanced techniques for validation and rigorous stress testing.
These innovative methodologies promise to not only withstand change but to thrive amidst it, bringing us closer to safer, more resilient power systems that can effectively adapt to the evolving demands of today’s world.
Learn More about Professor Javad Lavaei