Artificial Intelligence (AI) and energy generation are inexorably linked through a reciprocal, cyclical, and iterative relationship. Maintaining our advances in AI technology requires greater and greater amounts of energy, but these AI tools are being used to make advances in power generation.
One of Texas A&M Institute of Data Science’s (TAMIDS) Ph.D. Student Ambassadors, Zavier Ndum Ndum, presented his research paper on this topic at the IEEE 2024 Winter Simulation Conference in Orlando, Florida.
Zavier Ndum attending the 2024 IEEE Winter Simulation Conference
Ndum’s research paper, titled “A Digital Twin-Based Simulator for Small Modular and Microreactors,” details the development of a digital twin simulator for a conceptual 4.5 MWth lead-cooled fast reactor (LFR). The digital twin simulator, built using MATLABand Simulink, models the reactor’s complex physics and interactions between structures. Specifically, it simulates the neutronics and thermohydraulics through a coupled modular approach. The researchers validated the model against existing LFR data and incorporated a user interface for real-time visualization and interaction.
“We think such a development is important in the nuclear industry because our reactor digital twin could aid in design visualization, testing, and optimization at low cost. It could be used as a tool for training reactor operators and nuclear engineering students.”
-Zavier Ndum
The study demonstrates the digital twin model’s capability to analyze various reactor designs and operational factors, creating a risk-free method for training and experimentation. The simulation highlights the potential for accelerating LFR development and operator training to meet the growing need for energy sources that are safe, reliable, and do not emit greenhouse gases. Ndum’s future work includes expanding the model’s capabilities and integrating machine learning algorithms for predictive maintenance.
A digital twin is a virtual representation of a physical entity, process, or system, synchronized with a real-world counterpart. They are much more than a simulation, they offer real-time data integration by continuously receiving data from sensors and other sources to create physics-based modeling and feedback of the physical system. TAMIDS’s Digital Twin Lab, led by co-author Dr. Jian Tao, is dedicated to developing these innovative modeling, visualization, computing, and networking technologies to aid in the deployment of digital twins for a variety of real-world applications.
Dr. Tao is TAMIDS’s Assistant Director for Project Development and an Assistant Professor in Texas A&M University’s School of Performance, Visualization & Fine Arts. He is working with fellow co-author Dr. Yang Liu, Assistant Professor in the Department of Nuclear Engineering, on a TAMIDS-funded project, “AutoFLUKA: An AI-Assisted Framework for Automating Monte Carlo Simulations.”
Students presenting at the Energy Research Society Conference
Part of TAMIDS’s Seed Program for AI, Computing, and Data Science (SPAICD) grant initiative, Drs. Tao and Liu aim to simplify complex Monte Carlo simulations, making them more accessible and efficient by using an AI-assisted framework that integrates Large Language Models (LLMs) with nuclear energy domain-specific knowledge. Their collaboration exemplifies the AI-Energy nexus, where innovative AI approaches like the AutoFLUKA project not only accelerate research and development but also contribute to the creation of more sustainable and reliable energy solutions.
TAMIDS is dedicated to leveraging data science to solve real-world problems and encourages collaboration between data scientists and domain experts, like Ndum and Tao’s work in the Digital Twins Lab. As a TAMIDS Student Ambassador, Ndum hosts engaging events throughout the year to promote data literacy and connect new individuals to Texas A&M’s growing data science community. His first event of 2025 is “Bring Your Own Data (BYOD): Exploring AI, ML, and Data Science Applications in Nuclear Engineering,” a workshop for nuclear engineering students who are interested in learning and incorporating data science into their research or capstone projects. In addition to our Student Ambassadors’ events, TAMIDS also engages with student organizations like the Texas A&M Energy Research Society (ERS). TAMIDS was a proud sponsor of the ERS annual conference in 2024, which showcased groundbreaking advancements across the energy industry that focused on the conference’s theme of “Sustainable Energy for All: Accelerating the Transition to a Low-Carbon Future.” Through these initiatives and partnerships, TAMIDS continues to foster interdisciplinary collaboration and innovation, empowering students and researchers to drive impactful advancements in research.
Research Paper: Z. N. Ndum et al., “A Digital Twin-Based Simulator for Small Modular and Microreactors,” 2024 Winter Simulation Conference (WSC), Orlando, FL, USA, 2024, pp. 2963-2974, doi: 10.1109/WSC63780.2024.10838736.
