Optimal Control of Distributed Nuclear Reactors
Optimal Control of Distributed Nuclear Reactors

Author: G.S. Christensen

Publisher: Springer Science & Business Media

Published: 2013-11-11

Total Pages: 246

ISBN-13: 1489936025

DOWNLOAD EBOOK

This book is devoted to the mathematical optimization theory and modeling techniques that recently have been applied to the problem of controlling the shape and intensity of the power density distribution in the core of large nuclear reactors. The book has been prepared with the following purposes in mind: 1. To provide, in a condensed manner, the background preparation on reactor kinetics required for a comprehensive description of the main problems encountered in designing spatial control systems for nuclear reactor cores. 2. To present the work that has already been done on this subject and provide the basic mathematical tools required for a full understand ing of the different methods proposed in the literature. 3. To stimulate further work in this challenging area by weighting the advantages and disadvantages of the existing techniques and evaluating their effectiveness and applicability. In addition to coverage of the standard topics on the subject of optimal control for distributed parametersystems, the book includes, at amathemati cal level suitable for graduate students in engineering, discussions of con ceptsoffunctional analysis, the representation theory ofgroups, and integral equations. Although these topics constitute a requisite for a full understanding of the new developments in the area of reactor modeling and control, they are seidom treated together in a single book and, when they are, their presenta tion isoften directed to the mathematician.They are thus relatively unknown to the engineering community.

Energy Research Abstracts
Energy Research Abstracts

Author:

Publisher:

Published: 1990

Total Pages: 614

ISBN-13:

DOWNLOAD EBOOK

Semiannual, with semiannual and annual indexes. References to all scientific and technical literature coming from DOE, its laboratories, energy centers, and contractors. Includes all works deriving from DOE, other related government-sponsored information, and foreign nonnuclear information. Arranged under 39 categories, e.g., Biomedical sciences, basic studies; Biomedical sciences, applied studies; Health and safety; and Fusion energy. Entry gives bibliographical information and abstract. Corporate, author, subject, report number indexes.

Optimization of In-core Nuclear Fuel Management in a Pressurized Water Reactor
Optimization of In-core Nuclear Fuel Management in a Pressurized Water Reactor

Author: Richard Bartholomew Stout

Publisher:

Published: 1972

Total Pages: 316

ISBN-13:

DOWNLOAD EBOOK

Fuel loading patterns which have a minimum power peak are economically desirable to allow power reactors to operate at the highest possible power density and to minimize the possibility of fuel failure. A computer code called SHUFLE was developed for pressurized water reactors which shuffles the fuel in search of the lowest possible power peaking factor. An iterative approach is used in this search routine. A radial power distribution is calculated from which the program logic Selects a movement of fuel elements in an attempt to lower the radial power peak. Another power calculation is made and the process repeated until a predetermined convergence is reached. The logic by which the code decides the fuel movement is presented, along with the criteria for accepting or rejecting the move after a power calculation of the new loading pattern is made. A 1.5 group course mesh diffusion theory method was used to obtain the power distribution for each SHUFLE iteration. Convergence to a final loading pattern varies from about 10 to 40 shuffling iterations depending on the initial loading presented to the code. Since the typical computer running time for a one-quarter core power distribution with this 1.5 group method is only one to a few seconds, depending on the loading, convergence to a good loading pattern takes on the order of one minute on a Univac 1108. The low computer cost plus ease of operation should make this code of considerable use in determining loading patterns with minimum power peaking for any given set of fuel elements. The program also has burnup capability which can be used to check power peaking throughout core life. A parametric analysis study of fuel cycle costs for a PWR is also presented. Cost parameters analyzed were variation in the cost of yellow cake, enrichment, money, fabrication, and reprocessing plus changes in burnup, load factors, power densities, and the effect of forced early discharge. Figures are presented to indicate total fuel costs as a function of burnup for these cost parameters. Linear relationships for minimum cost and optimum burnup are presented for each parameter.

A Hybrid Method for In-core Optimization of Pressurized Water Reactor Reload Core Design
A Hybrid Method for In-core Optimization of Pressurized Water Reactor Reload Core Design

Author:

Publisher:

Published: 1995

Total Pages: 209

ISBN-13:

DOWNLOAD EBOOK

The objective of this research is the development of an accurate, practical, and robust method for optimization of the design of loading patterns for pressurized water reactors, a nonlinear, non-convex, integer optimization problem. The many logical constraints which may be applied during the design process are modeled herein by a network construction upon which performance objectives and safety constraints from reactor physics calculations are optimized. This thesis presents the synthesis of the strengths of previous algorithms developed for reload design optimization and extension of robustness through development of a hybrid liberated search algorithm. Development of three independent methods for reload design optimization is presented: random direct search for local improvement, liberated search by simulated annealing, and deterministic search for local improvement via successive linear assignment by branch and bound. Comparative application of the methods to a variety of problems is discussed, including an exhaustive enumeration benchmark created to allow comparison of search results to a known global optimum for a large scale problem. While direct search and determinism are shown to be capable of finding improvement, only the liberation of simulated annealing is found to perform robustly in the non-convex design spaces. The hybrid method SHAMAN is presented. The algorithm applies: determinism to shuffle an initial solution for satisfaction of heuristics and symmetry; liberated search through simulated annealing with a bounds cooling constraint treatment; and search bias through relational heuristics for the application of engineering judgment. The accuracy, practicality, and robustness of the SHAMAN algorithm is demonstrated through application to a variety of reload loading pattern optimization problems.