Woodruff School of Mechanical Engineering
Nuclear & Radiological Engineering and Medical Physics Programs
Exact-to-Precision Generalized Perturbation Theory Development for Reactor Analysis Calculations
Dr. Hany Abdel-Khalik
North Carolina State University
Thursday, October 25, 2012 at 11:00:00 AM
Boggs Building, Room 3-49
Dr. Said Abdel-Khalik
Generalized perturbation theory has been one of the most prominent mathematical tools used in nuclear engineering. It was first introduced by Wagner in 1945 and since then received numerous developments by reactor physicists such as Gandini, Williams, Stacey, Cacuci, and Rahnema, and others. Its primary function is to calculate variations in responses of interest resulting from parameter perturbations to an often expensive model, e.g. transport equation. This function is needed because in reactor analysis, one needs calculate the impact of various core conditions on responses of interest, such as criticality, reactivity coefficients, peaking factors, etc.
Although the mathematical aspects of GPT have matured over the years, its application has always been limited to investigative studies only. This is mainly because of the computational cost associated with its execution for realistic reactor models, expected to be nonlinear, multi-physics, associated with many responses and parameters. In this talk, I overview recent developments at NCSU based on the work of Mr. Congjian Wang (a PhD candidate) focused on introducing a new framework for GPT to overcome existing challenges. In particular, EpGPT is designed to handle models with many parameters and responses. Moreover, unlike existing GPT, EpGPT allows one calculate all orders of variations combined to a given user-defined tolerance. Given the demonstrated potential of EpGPT, we believe one could improve the efficiency and accuracy of existing reactor design calculations.
Hany Abdel-Khalik received his undergraduate degree (graduating college valedictorian) in nuclear engineering from Alexandria University (Egypt) in 2000. He received his Ph.D. from North Carolina State University (NCSU) in 2004 in the area of computational reactor physics focusing on the development of innovative sensitivity, uncertainty, and inverse methods to enhance the accuracy of nuclear reactor calculations. After graduation, he joined the reactor analysis methods group at AREVA-NP’s PWR Engineering Unit. He is currently an assistant professor at NCSU, department of nuclear engineering.