Nuclear Engineering - Master's degree
http://hdl.handle.net/1721.1/7689
Tue, 21 Feb 2017 00:47:42 GMT2017-02-21T00:47:42ZOptimization and analysis of the laser isotope separation technique SILEX and ensuing proliferation ramifications
http://hdl.handle.net/1721.1/106767
Optimization and analysis of the laser isotope separation technique SILEX and ensuing proliferation ramifications
Baldwin, Aaron Taylor
SILEX is a molecular isotope separation technique that takes advantage of the differing energies of molecular excitations between different isotopes of uranium. This process occurs within a jet of supersonic gas, the gas includes both uranium hexafluoride and a carrier gas, and reportedly allows for high separation factors relative to other isotope-separation processes. Industry interests have argued that it could be readily commercialized. This topic is of particular interest because laser isotope separation technology has seen an increase in interest and funding over the last decade. This suggests some study of the risks that such a technology poses to society may now be in order. To inform policymakers about the risks inherent to a particular enrichment technology, it is necessary to understand the theoretical underpinnings of the technology before one can analyze the impact of the technology. Positions expressed in the current literature are ill-informed and range from deep opposition, citing concerns that SILEX poses greater proliferation risk than centrifuge or gaseous diffusion technology, to claims by scientists that it is not possible to use the technology to produce greater than 50% enriched U-235. A rigorous and holistic view of the technology will better inform policy by improving the accuracy of claims and identifying realistic solutions to problems the technology may pose. This thesis will seek to provide this deeply technical and holistic analysis of the technology, and will use the results to interpret the economic and proliferation impact such a technology will have on the global nuclear enterprise. The holistic analysis in this thesis will present several important conclusions: 1) the enrichment factor of SILEX is not quite as large as proponents suggest; 2) asymmetric cascade designs will be required; 3) SILEX may not be cheaper than centrifuge facilities; 4) SILEX will not be viable without improvements in laser technology; and 5) international policies may be the most effective means of curtailing enrichment schemes like SILEX.
Thesis: S.M., Massachusetts Institute of Technology, Department of Nuclear Science and Engineering, 2016.; Cataloged from PDF version of thesis.; Includes bibliographical references (pages 125-132).
Fri, 01 Jan 2016 00:00:00 GMThttp://hdl.handle.net/1721.1/1067672016-01-01T00:00:00ZSelection and implementation of deterministic and probabilistic models for the prediction of stress corrosion cracking in used nuclear fuel containment material
http://hdl.handle.net/1721.1/106766
Selection and implementation of deterministic and probabilistic models for the prediction of stress corrosion cracking in used nuclear fuel containment material
Ferry, Sara Elizabeth
Stainless steel canisters encased in vented concrete overpacks are used to store used nuclear fuel at interim spent fuel storage installations in the United States. These storage systems are exposed to the elements. There is concern that, over time, a deliquesced salt film could develop on the stainless steel canister. Such a film can create a corrosive environment in stainless steels. If a stress is present on or in then material, stress corrosion cracking (SCC) could be a possibility. Because the canister welds are not stress-relieved in order to avoid sensitization of the steel, residual stresses are expected to be present in the canisters. Thus, there is interest in determining the likelihood that (a) a sufficiently corrosive film develops on a stainless steel used fuel canister (b) there are sufficiently high stresses in the material at the location of the corrosive film (c) SCC initiates and (d) the crack propagates through the canister wall, resulting in canister failure. This thesis begins with the assumption that a corrosive film has developed on the canister surface, and pits have begun to initiate. It investigates various methods of modeling SCC in the canister wall after the point of corrosion pit initiation. An extensive literature review was carried out in order to understand the different SCC models that currently exist in the literature. A figure-of-merit was developed to decide which models were the most likely to be helpful to the modeling of SCC in used fuel canisters. The figure-of-merit was then used to select the most promising models. These models were then used to write MATLAB@ simulations that could be used to predict time-to-failure in canisters due to SCC once corrosion pits have begun to grow. The results of these simulations are then considered and compared, and used to inform recommendations for future development of a useful predictive model of SCC in used fuel canisters.
Thesis: S.M., Massachusetts Institute of Technology, Department of Nuclear Science and Engineering, 2016.; "June 2016." Cataloged from PDF version of thesis.; Includes bibliographical references (pages 391-408).
Fri, 01 Jan 2016 00:00:00 GMThttp://hdl.handle.net/1721.1/1067662016-01-01T00:00:00ZUncertainty quantification and calibration in nuclear safety codes using Gaussian process active learning
http://hdl.handle.net/1721.1/106691
Uncertainty quantification and calibration in nuclear safety codes using Gaussian process active learning
Fugleberg, Eric N. (Eric Nels)
Inverse problems and inverse uncertainty quantification (UQ) are challenging issues when dealing with complex and highly non-linear functions. Methods have been developed to decrease the computational burden by using the Gaussian Process (GP) emulator model framework to approximate the input-output relation of a deterministic computer code. The GP emulator can then be used in place of the computer code to perform Bayesian calibration techniques to determine uncertain parameter distribution. The performance of a GP emulator is largely dependent on the quality of the points in its training set; the best emulator exactly replicates the output of the computer code. The uncertain parameter posterior sample space is not known a priori, resulting in GP training sets covering as much of the prior sample space as possible in hopes of covering the posterior space well enough. This work improves the performance of the simple GP emulator using an active learning methodology to select additional training points which cover the posterior sample space of the unknown parameters. Furthermore, the effect of the covariance function on the performance of the GP is investigated with recommendations made for future GP emulator applications.
Thesis: S.M., Massachusetts Institute of Technology, Department of Nuclear Science and Engineering, 2016.; This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.; Cataloged from student-submitted PDF version of thesis.; Includes bibliographical references (pages 85-87).
Fri, 01 Jan 2016 00:00:00 GMThttp://hdl.handle.net/1721.1/1066912016-01-01T00:00:00ZCalculating reaction rate derivatives in Monte Carlo neutron transport/
http://hdl.handle.net/1721.1/106690
Calculating reaction rate derivatives in Monte Carlo neutron transport/
Harper, Sterling (Sterling M.)
An operating nuclear power reactor is a complex system that is sensitive to many material parameters including densities, temperatures, and compositions. There is great interest in solving the neutron transport with Monte Carlo methods due to their extremely high fidelity, but Monte Carlo methods are too slow to run in an iterative brute-force search of the reactor parameter space. This thesis discusses the derivation, implementation, and applications of differential tallying -- a method which can be used to mitigate the computational cost of mapping out a reactor parameter space with Monte Carlo. With differential tallies, each calculation provides derivatives of tallied quantities like reactivity and fission reaction rates with respect to material density, temperature, etc. These derivatives directly provide reactivity coefficients and they can also be used to extrapolate and predict small changes in reactor parameters. Notably, a novel method is presented which uses the windowed multipole cross section representation to compute temperature derivatives due to the resolved resonance Doppler broadening effect. To demonstrate the utility of differential tallies, this thesis presents example computations of moderator density and fuel Doppler feedback coefficients in pressurized water reactor pincells. With differential tallies, the moderator and fuel Doppler coefficients can be computed 40% and 50x faster, respectively, than by brute-force methods. A calculation of pin-by-pin Doppler coefficients in an assembly is also presented in order to demonstrate that differential tallies are even more efficient for assembly calculations.
Thesis: S.M. and S.B., Massachusetts Institute of Technology, Department of Nuclear Science and Engineering, 2016.; This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.; Cataloged from student-submitted PDF version of thesis.; Includes bibliographical references (pages 63-64).
Fri, 01 Jan 2016 00:00:00 GMThttp://hdl.handle.net/1721.1/1066902016-01-01T00:00:00Z