Dr. Gregory Delipei
Research Scholar
Department of Nuclear Engineering
North Carolina State University

Abstract

Artificial intelligence and machine learning (AI/ML) methods have been increasingly used in nuclear engineering applications. The need to improve the understanding of these methods, to develop guidelines and to provide a framework for assessing their predictive performance lead to the development of the OECD/NEA Task Force on AI/ML in 2022. One of the main goals of this Task Force is to develop benchmark exercises for evaluating the AI/ML models. In this lecture, we will provide an overview of this Task Force, present the currently developed benchmark exercises, and highlight some future steps. The preliminary results and analysis from the execution of the first benchmark exercise on the Critical Heat Flux prediction will be discussed.

Biography

Dr. Gregory Delipei (NCSU) is a Research Scholar in the Department of Nuclear Engineering at NCSU. Prior to his current position, he was a Postdoctoral Research Scholar at NCSU and he received his PhD in Nuclear Engineering from University of Paris Saclay in France. His research expertise is in the domain of uncertainty quantification for multi-physics, multi-scale and multi-fidelity calculations and the development of machine learning approaches for reactor core applications. The application domain of his research spans both LWR and Advanced Reactors. He is currently actively involved in many ongoing OECD/NEA international activities including the management and coordination of the Task Force on Task Force on Artificial Intelligence and Machine Learning.

January 16. 2024
4:10 pm seminar

zoom link upon request