Doctoral Dissertations
Date of Award
3-1987
Degree Type
Dissertation
Degree Name
Doctor of Philosophy
Major
Nuclear Engineering
Major Professor
H. L. Dodds Jr.
Committee Members
P. F. Pasqua
Abstract
A new approach for the physics design and analysis of LWR reload cores is developed and demonstrated through several practical applications. This new design philosophy uses first- and second-order response derivatives to predict the important reactor performance characteristics (power peaking, reactivity coefficients, etc.) for any number of possible material configurations (assembly shuffling and burnable poison loadings). The response derivatives are computed using generalized perturbation theory (GPT) techniques.
This report describes in detail an idealized GPT-based design system. The idealized system would contain individual modules to generate the required first-order and higher-order sensitivity data. It would also contain at least two major application codes; one for core design optimization and the other for evaluation of several safety parameters of interest in off-normal situations. This ideal system would be fully automated, user-friendly, and quite flexible in its ability to provide a variety of design and analysis capabilities.
Three major components of the system are discussed in detail. First, a GPT-based optimization algorithm is developed, implemented, and illustrated in several applications. The extremely good results obtained in these demonstrations clearly show the potential of the overall methodology. Next, a new first-order GPT formulation that includes macroscopic depletion, k-reset via soluble boron control, and poison and thermal-hydraulic feedback effects is developed. The feedback treatment in the adjoint mode is verified through analytical and numerical illustrations. Finally, several aspects of a practical approach to higher-order GPT are addressed. In particular, a formal second-order GPT formulation is developed and three promising techniques for calculating the required flux derivatives are identified.
The information gained from these three studies gives a good foundation for the development of a complete integrated design package. The initial testing and applications documented herein verify the feasibility of the basic concept and illustrate the unique capabilities and overall potential of the GPT-based design methodology. The actual implementation of a production-oriented GPT-based design system is identified as the logical follow-up to this work.
Recommended Citation
White, John R., "Conceptual development of a complete LWR reload design methodology based on generalized perturbation theory. " PhD diss., University of Tennessee, 1987.
https://trace.tennessee.edu/utk_graddiss/12182