Typescript. / Digitized by Kansas Correctional Industries

Pressurizer modeling for predicting the dynamic pressure of the PIUS system is presented. The transient behavior of this model for the PIUS system was investigated. The validity of this model for the PIUS system is limited to transients that are neither too large nor too long in duration. For example, the model is not capable of describing events following a complete loss of liquid for the pressurizer. However, the model can be used for qualitative prediction of the PIUS system behavior for a wide variety of severe transients. A review of pressurizer modeling indicates that the neglecting of the change in the internal energy of the subcooled water during transients is an acceptable assumption. The inherently safe feature of the PIUS system was confirmed through the self-shutdown of the reactor or, in some cases, through reactor power reduction as a result of the ingress of the pool boric acid solution into the primary system. This dynamic model was constructed of three major components: (1) The primary loop, (2) The secondary loop, and (3) The natural convection loop through the pool. A lumped parameter model, uniform heat transfer, and point kinetics have been the main approximations in this model. Other approximations are mentioned during the modeling of each component of the model. The dynamic model was simulated using the DARE-P continuous system simulation language which was developed in the Electrical Engineering Department at the University of Arizona.

Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1981. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING. / Bibliography: leaves 182-188. / by Juan J. Manzano-Ruiz. / Ph.D.

Thesis. 1978. Nucl.E.--Massachusetts Institute of Technology. Dept. of Nuclear Engineering. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Includes bibliographical references. / by Gilberto Gomes De Andrade. / Nucl.E.

Benchmark calculations are performed for neutron transport in a two material (binary) stochastic multiplying medium. Spatial, angular, and energy dependence are included. The problem considered is based on a fuel assembly of a common pressurized water nuclear reactor. The mean chord length through the assembly is determined and used as the planar geometry system length. According to assumed or calculated material distributions, this system length is populated with alternating fuel and moderator segments of random size. Neutron flux distributions are numerically computed using a discretized form of the Boltzmann transport equation employing diffusion synthetic acceleration. Average quantities (group fluxes and k-eigenvalue) and variances are calculated from an ensemble of realizations of the mixing statistics. The effects of varying two parameters in the fuel, two different boundary conditions, and three different sets of mixing statistics are assessed. A probability distribution function (PDF) of the k-eigenvalue is generated and compared with previous research. Atomic mix solutions are compared with these benchmark ensemble average flux and k-eigenvalue solutions. Mixing statistics with large standard deviations give the most widely varying ensemble solutions of the flux and k-eigenvalue. The shape of the k-eigenvalue PDF qualitatively agrees with previous work. Its overall shape is independent of variations in fuel cross-sections for the problems considered, but its width is impacted by these variations. Statistical distributions with smaller standard deviations alter the shape of this PDF toward a normal distribution. The atomic mix approximation yields large over-predictions of the ensemble average k-eigenvalue and under-predictions of the flux. Qualitatively correct flux shapes are obtained, however. These benchmark calculations indicate that a model which includes higher statistical moments of the mixing statistics is needed for accurate predictions of binary stochastic media k-eigenvalue problems. This is consistent with previous findings. / Graduation date: 2005

Oregon State University has hosted an International Atomic Energy Agency (IAEA) International Collaborative Standard Problem (ICSP) through testing conducted on the Multi-Application Small Light Water (MASLWR) facility. The MASLWR facility features a full-time natural circulation loop in the primary vessel and a unique pressure suppression device for accident scenarios. Automatic depressurization system (ADS) lines connect the primary vessel to a high pressure containment (HPC) which dissipates steam heat through a heat transfer plate thermally connected to another vessel with a large cool water inventory. This feature drew the interest of the IAEA and an ICSP was developed where a loss of feedwater to the steam generators prompted a depressurization of the primary vessel via a blowdown through the ADS lines. The purpose of the ICSP is to evaluate the applicability of thermal-hydraulic computer codes to unique experiments usually outside of the validation matrix of the code itself. RELAP5-3D 2:4:2 was chosen to model the ICSP. RELAP5-3D is a best-estimate code designed to simulate transient fluid and thermal behavior in light water reactors. Modeling was conducted in RELAP5-3D to identify the strengths and weaknesses of the code in predicting the experimental trends of the IAEA ICSP. This extended to nodalization sensitivity studies, an investigation of built-in models and heat transfer boundary conditions. Besides a qualitative analysis, a quantitative analysis method was also performed. / Graduation date: 2013

Several computer codes based on one and two-group diffusion theory models were developed for SHUFFLE. The programs were developed to calculate power distributions in a two-dimensional quarter core geometry of a pressurized power reactor. The various coarse-mesh numerical computations for the power calculations yield the following: the Borresen's scheme applied to the modified one-group power calculation came up with an improved power distribution, the modified Borresen's method yielded a more accurate power calculations than the Borresen's scheme, the face dependent discontinuity factor method have a better prediction of the power distribution than the node averaged discontinuity factor method, Both the face dependent discontinuity factor method and the modified Borresen's methods for the two-group model have quite attractive features. / Graduation date: 1992

The main objective of this research is: (1) to develop a model and perform numerical simulations to evaluate the radiation field and the resulting dose to personnel and activation of materials and structures throughout the IRIS nuclear power plant, and (2) to confirm that the doses are below the regulatory limit, and assess the possibility to reduce the activation of the concrete walls around the reactor vessel to below the free release limit. IRIS is a new integral pressurized water reactor (PWR) developed by an international team led by Westinghouse with an electrical generation capacity of 335 MWe and passive safety systems. Its design differs from larger loop PWRs in that a single building houses the containment as well as all the associated equipment including the control room that must be staffed continuously. The resulting small footprint has positive safety and economic implications, and the integral layout provides additional shielding and thus the opportunity to significantly reduce the activation, but it also leads to significantly more challenging simulations. The difficulty in modeling the entire building is the fact that the source is attenuated over 10 orders of magnitude before ever reaching the accessible areas. For an analog Monte Carlo simulation with no acceleration (variance reduction), it would take many processor-years of computation to generate results that are statistically meaningful. Instead, to generate results for this thesis, the Standardized Computer Analyses for Licensing Evaluation (SCALE) with the package Monaco with Automated Variance Reduction using Importance Calculations (MAVRIC) will be used. This package is a hybrid methodology code where the forward and adjoint deterministic calculations provide variance reduction parameters for the Monte Carlo portion to significantly reduce the computational time. Thus, the first task will be to develop an efficient SCALE/MAVRIC model of the IRIS building. The second task will be to evaluate the dose rate and activation of materials, specifically focusing on activation of concrete walls around the reactor vessel. Finally, results and recommendations will be presented.

Thesis (Ph. D.)--University of North Texas, May, 2007. / Title from title page display. Includes bibliographical references.

Thesis (M.S.)--University of North Texas, May, 2008. / Title from title page display. Includes bibliographical references.