Richard Vega

  • Program Year: 4
  • Academic Institution: Texas A&M University
  • Field of Study: Computational Neutron Transport
  • Academic Advisor: Marvin Adams
  • Practicum(s):
    Lawrence Livermore National Laboratory (2016)
    Sandia National Laboratories, New Mexico (2018)
  • Degree(s):
    B.S. Physics, and B.S. Nuclear Engineering, Texas A&M University, 2014; A.A. General Studies, Lake-Sumter State College, 2010

Summary of Research

At the core of nuclear engineering is the neutral particle Boltzmann transport equation which is an integro-differntial equation for the angular flux as a function of seven independent variables. With such a large phase space, anayltical solutions are hard to come by except for the simplest of cases. It is thus necessary to resort to numerical solutions, and in particular fast and efficient numerical solutions. With the advancements in high performance computing, this means massively parallel computation on super computers with as many central processing units (CPUs) as possible. The CPU has been the work horse for the computational physicist for many years, and when it was realized that there would be a physical limit on the speed of a single CPU, parallel computing allowed for the solution of larger more complicated problems by utilizing many CPUs concurrently; however, there is another hardware component found in many laptops and desktops that has recently been gaining interest.

The graphics processing unit (GPU) can be thought of as a collection of many "lightweight" processing units designed to perform the same operations on different sets of data. This single instruction multiple data (SIMD) parallelism makes them ideal for video game production where each core is used to calculate the value of a single pixel on a screen. It was eventually realized that if a problem could be parallelized to resemble this pipelined computation of pixel evaluation, significant performance gains could be achieved.

If I had to describe my research in as few words as possible, it would be the development of algorithms to solve the neutral particle Boltzmann transport equation faster and more efficiently by utilizing a GPU wherever possible. This is motivated by the fact that the next generation of super computers at the national labs are going to have many GPUs in addition to CPUs, and to put it succinctly, if you aren't using them, you are wasting them!

Publications

Development of a genetic algorithm for neutron energy spectrum adjustment.
Richard M. Vega, Edward J. Parma
Presented at the Joint International Conference on Mathematics and Computation (M&C), Supercomputing in Nuclear Applications (SNA) and the Monte Carlo (MC) Method 2015.

Reactivity effects at the Mayak Production Association, 2 January 1958
criticality accident using Serpent 2 and OpenFOAM.
Richard M. Vega, Taylor K. Lane, John A. Miller, and Norman F. Schwers
To be presented at the International Conference on Nuclear Criticality Safety, ICNC 2015.

Radiation Characterization Summary: ACRR Central Cavity Free-Field Environment with the 32-Inch Pedestal at the Core Centerline (ACRR-FF-CC-32-cl), SAND2015-6483
Edward J. Parma, Gerald E. Naranjo, Richard M. Vega, Lance L. Lippert,
David W. Vehar, and Patrick J. Griffin

Neutron reference benchmark field specification: ACRR 44 inch lead-boron (LB44) bucket environment (ACRR-LB44-CC-32-CL), SAND2015-5359 R
Richard M. Vega, Edward J. Parma, Patrick J. Griffin, and David W. Vehar

Neutron reference benchmark field specification: ACRR free-field environment (ACRR-FF-CC-32-CL), SAND2015-5360 R
Richard M. Vega, Edward J. Parma, Patrick J. Griffin, and David W. Vehar

Neutron reference benchmark field specification: ACRR polyethylene-lead-graphite (PLG) bucket environment (ACRR-PLG-CC-32-CL), SAND2015-5358 R
Richard M. Vega, Edward J. Parma, Patrick J. Griffin, and David W. Vehar

GenSpec: A Genetic Algorithm for Neutron Energy Spectrum Adjustment, SNAD2015-11036
Richard M. Vega, Edward J. Parma

Thin polycrystalline diamond films protecting Zirconium alloys surfaces: from technology to layer analysis and application in nuclear facilities,
P. Ashcheulov, R. Skoda, J. Skarohlíd, A. Taylor, L. Fekete, F. Fendrych, R. Vega,L. Shao, L. Kalvoda, S. Vratislav, V. Cháb, K. Horáková, K. Kusová, L. Klimsa, J. Kopecek, P. Sajdl, J. Macák, S. Johnson , I. Kratochvílová,
Applied Surface Science, vol 359, pp. 621-628, 2015

Transport Sweeps Using an Improved Slice Balance Approach with LDFE and GPU Acceleration.
Richard M. Vega, Marvin L. Adams
Presented at the International Conference on Mathematics & Computational Methods Applied to Nuclear Science & Engineering,
Jeju, Korea, April 16-20, 2017

Awards

DOE NNSA Stewardship Science Fellow

American Nuclear Society
Alan F. Henry/ Paul A. Greebler Memorial Scholarship

Texas A&M University
Recipient of the Crawford and Hattie Jackson Scholarship
Recipient of the Marianne and Robert Hamm Scholarship
Recipient of the Adams Family Scholarship
Dean's Honor Award (Every semester of attendance)
Most Outstanding Undergraduate Thesis Award

Lake Sumter Community College
Outstanding Achievement Award in Mathematics
Outstanding Achievement Award in Chemistry
Presidents List (Every Semester of Attendance)
Academic Excellence Award

Phi Theta Kappa
All-State Academic Team
Excellence in Leadership Award