New Course: Computational Radiation Transport

Term: Fall 2012 Semester
Course Number: ENGR/ME 2118
Time: Thursday 6:00 – 8:30 PM
Instructors: Dr. David Griesheimer, Dr. Dan Gill & Dr. Justin Pounders.

Description: This course provides an in-depth review of modern computational techniques used for solving the linear Boltzmann equation, with specific applications to neutron and photon radiation transport problems. Topics to be covered include: an introduction to the physical processes that govern radiation transport through materials, Monte Carlo methods for the simulation of radiation transport, a first-principles derivation of the Boltzmann radiation transport equation for multiplying and non-multiplying systems, the multi-group, diffusion, and discrete ordinates approximations to the transport equation, expansion of the scattering kernel in Legendre polynomials, and numerical methods for approximating solutions to the transport equation. In addition, the course will review many commonly used numerical methods for solving integral and differential equations, including: finite differencing, numerical quadrature, harmonic analysis, and the power method for solving eigenvalue problems. Topics covered in the class will be reinforced with weekly programming exercises designed to illustrate the different methods for solving the Boltzmann radiation transport equation and demonstrate how these methods can be used to solve realistic problems related to nuclear reactor and radiation shielding analysis. The course will also places a strong emphasis on formal quality assurance methods (and best-practices) for the development, verification, and validation of scientific computer codes intended for use in engineering design calculations of record. Students will immediately take what they have learned in lecture and write a computer program that is capable of solving a real problem (up to 2D).

The three instructors are adjunct faculty that work at Bettis Atomic Power Laboratory. They have experience working as a part of integrated software development teams to develop, verify, and validate large-scale scientific computer programs for use in engineering design calculations. Dr. Griesheimer has experience leading a team of 13 programmers in the development of a large-scale (100,000+ lines of code) Monte Carlo transport solver for reactor analysis calculations. The instructors all have a PhD in Nuclear Engineering with emphasis on scientific computing and computational methods for solving the neutron transport equation.