Led by Assistant Professor Syed Bahauddin Alam of the Department of Nuclear, Plasma and Radiological Engineering (NPRE), the project integrates advanced machine learning with high-performance computing to create virtual sensors capable of rapidly and accurately predicting key physical conditions inside nuclear reactors. These AI-driven models produce results up to 1,400 times faster than conventional Computational Fluid Dynamics (CFD) simulations.
The study, published in npj Materials Degradation, showcases how Deep Operator Neural Networks (DeepONet) can infer thermal-hydraulic behavior across the entire reactor domain using limited physical input data. Trained on data generated using supercomputing infrastructure provided by Illinois Computes and the National Center for Supercomputing Applications (NCSA), the models function as real-time monitoring tools that sidestep the constraints of physical sensors.
Physical sensors often struggle to operate in the high-temperature, high-radiation environments found within reactor cores. Traditional modeling methods, while accurate, are too computationally intensive to deliver timely predictions. The DeepONet framework fills this critical gap, offering a virtual map of reactor behavior by predicting thermal and flow parameters in key areas like the hot leg of pressurized water reactors.
"Our research introduces a new way to keep nuclear systems safe by using advanced machine-learning techniques to monitor critical conditions in real-time," Alam said. "Traditionally, it's been incredibly challenging to measure certain parameters inside nuclear reactors because they're often in hard-to-reach or extremely harsh environments. Our approach leverages virtual sensors powered by algorithms to predict crucial thermal and flow conditions without needing physical sensors everywhere.
"Think of it like having a virtual map of how the reactor is operating, giving us constant feedback without having to place physical instruments in risky spots. This not only speeds up the monitoring process but also makes it significantly more accurate and reliable. By doing this, we can detect potential issues before they become serious, enhancing both safety and efficiency."
The project was made possible through the Illinois Computes program, which provided access to Delta's NVIDIA A100 GPU nodes and CPU infrastructure for model training and data generation. NCSA graduate students Kazuma Kobayashi and Farid Ahmed contributed to the technical development alongside Alam, with AI and HPC support from NCSA experts Dr. Diab Abueidda and Dr. Seid Koric.
"In this Illinois Computes project, we have fully utilized the unique high-performance computing resources and multidisciplinary expertise at NCSA and the Grainger College of Engineering to advance translational and transformative engineering research in Illinois," said Koric.
Abueidda added, "This collaboration exemplifies the synergy that emerges when advanced AI methods, high-performance computing resources and domain expertise converge. Working alongside Dr. Alam's team and NCSA's AI and HPC experts, we leveraged the U.S. National Science Foundation-funded Delta's cutting-edge capabilities to push the boundaries of real-time monitoring and predictive analysis in nuclear systems. By uniting our specialized skill sets, we have accelerated research while enhancing the accuracy and reliability of critical safety measures.
"We look forward to continuing this interdisciplinary approach to drive transformative solutions for complex energy systems. Ultimately, these breakthroughs highlight the promise of computational science in addressing the pressing challenges of nuclear energy."
Related Links
Department of Nuclear, Plasma and Radiological Engineering
Nuclear Power News - Nuclear Science, Nuclear Technology
Powering The World in the 21st Century at Energy-Daily.com
| Subscribe Free To Our Daily Newsletters |
| Subscribe Free To Our Daily Newsletters |
