Four NSSPI graduate students will be walking across the stage to receive their degrees this May. All four will be earning their Master’s of Science in Nuclear Engineering.
Ryan Coogan
Adviser: Dr. Craig Marianno
Master’s of Science in Nuclear Engineering
Abstract: The challenge of an adequately detailed smuggling network problem is that the number of variables required to adequately capture the problem also makes the problem computationally exhaustive. A well bounded problem, although simple, can provide meaningful information to a decision-maker. Limiting the problem to a comparison of two technologies, a decision-maker can prioritize how to best allocate resources, by reinforcing the border with stationary Radiation Portal Monitors (RPMs) which can be perceived, or by investing in Mobile Radiation Detection Systems (MRDS) which are harder for an adversary to detect but may have other weaknesses. An abstract, symmetric network is studied to understand the impact of initial conditions on the network, and the most conservative choices are made in an asymmetric network loosely modeled on the state of Texas transportation system. This asymmetric network is then examined for the technology that will maximally suppress the adversary’s success rate at minimal cost.
Katie Cook
Adviser: Dr. Craig Marianno
Master’s of Science in Nuclear Engineering
Abstract: When it comes to saving lives after a destructive and catastrophic crisis, urban search and rescue (USAR) dogs are an essential emergency response component, where each dog can perform the equivalent work of 20 to 30 people. However, based on current practices, if any crisis contained the dispersal of nuclear material, these dogs and their handlers may not be able to take part in their lifesaving missions due to few protective guidelines. In this study a 9.29 m2 area was sprayed with 200 MBq of 18F, and a dog executed minor search activities in this contaminated area. Using a positron emission tomography (PET) scanner both internal and external contamination from the dog was localized and quantified. Total contamination on the dog as quantified by the PET scan was 3.4 kBq with external and internal contamination being 2.1 kBq and 1.3 kBq, respectively. Total external dose received to the dog during the exercise was 0.19 mGy, and total internal dose was 1.1 μGy. Overall, this contamination exercise proved a viable method to simulate a radioactive environment safe enough for a dog to participate in but strong enough to create detectable contamination. This will allow researchers to gain insight into health concerns that may arise if a USAR dog took part in a real-world contamination event.
Jee Hoon Moo
Adviser: Dr. Sunil Chirayath
Master’s of Science in Nuclear Engineering
Abstract: The objective of this study is to predict the damage that could be created on a reinforced concrete wall due to the impact of Tri-Nitro Toluene (TNT) shaped charge using a numerical simulation software. There are many commercial numerical simulation software that can be used to solve engineering problems in real world scenarios, such as ANSYS, ABAQUS, LS-DYNA, etc. ANSYS Simulation Software is one of the strongest and user-friendly commercial finite element analysis (FEA) tools that uses computer-based numerical techniques. In this study, two different types of simulation software codes which can be coupled for the advanced analysis are used within ANSYS Simulation Software system. One is ANSYS Explicit Dynamic STR (Structure) software and the other one is ANSYS AUTODYN software. ANSYS Explicit Dynamic STR software provides suitable solutions of nonlinear dynamic events for a short duration, including a drop and impact testing with low velocity or high velocity, deformation by high pressure, explosion, etc. ANSYS AUTODYN software also provides suitable solutions of nonlinear dynamic events similar to ANSYS Explicit Dynamic STR, but this software is focused on complicated nonlinear dynamic events like high explosions and detailed damage responses of materials such as cracks and fragments. The reinforced concrete wall target is located at a distance of 50 meters from a TNT shaped charge design. Various TNT shaped charge designs are studied by changing the amount of TNT and liner fragment thickness, etc. in order to make the hole-size big enough on the concrete wall target so that at least a person can pass through it at a time. In addition, a physical protection system vulnerability assessment is performed with a hypothetic nuclear research reactor assuming that TNT shaped charge is used in order to reach to a desired target from the offsite to the nuclear research reactor. The probability of interruption (PI) is calculated with the Adversary Sequence Diagram (ASD) using Estimate of Adversary Sequence Interruption (EASI) model to conduct the vulnerability assessment, which provides the most vulnerable path to reach to the target while minimizing time delay (td) and the probability of detection (PD).
Charles “Rob” Schafer
Adviser: Dr. Sunil Chirayath
Master’s of Science in Nuclear Engineering
Abstract: Monte Carlo N-Particle Transport Code (MCNP) is a Monte Carlo computational neutron transport code with multi core parallel simulation functionality developed by Los Alamos National Laboratory and is widely used in nuclear reactor modeling and nuclear fuel burnup/depletion simulations. In burnup simulations, MCNP calculates neutron reaction rates and their corresponding stochastic uncertainties at each differential time step of fuel depletion, however these reaction rate uncertainties are not currently propagated through multiple depletion time steps. Moreover, these reaction rate uncertainties are not propagated into the concentrations of fission products and actinides produced in depleted nuclear fuel.
The objective of this thesis research is to develop a methodology to quantify the stochastic uncertainties in actinide and fission product concentration estimates from performing nuclear fuel burnup simulations using MCNP. Key trace intra-element fission product isotope ratios in separated plutonium are of specific interest in nuclear forensics, as these ratios can give insight into the reactor-type that produced the plutonium, fuel burnup and time since irradiation A mathematical methodology using the reaction rates given by an MCNP burnup simulation was developed to estimate the concentrations of various nuclides over the entirety of the depletion simulation. A Monte Carlo sampling procedure of reaction rates using the predicted stochastic uncertainty was implemented into the model, where each necessary reaction rate was sampled for each depletion time step. This procedure was run multiple times and a statistical analysis was performed to estimate the overall stochastic uncertainty in selected fission product and actinide concentrations predicted at the end of multiple time step depletion simulations by MCNP. The results from the Multi Stage Monte Carlo Methodology were then compared to an MCNP data set, containing data from over 100 identical MCNP burnup simulations and results from new methodology were found effective in predicting overall stochastic uncertainties.