Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Programmes to learn for Nuclear Engineering Masters

  1. Jul 26, 2017 #1
    I am about to begin an MSc in Nuclear Engineering. I am competent in MatLab language but wish to practice/learn another in the time before my course starts. Would I be well served in continuing to practice MatLab or to learn another from scratch (C/Java/Labview)? Are there any particular areas within these programmes particularly relevant to nuclear physics that I should focus on?
  2. jcsd
  3. Jul 26, 2017 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    It used to be that most scientist/engineers would learn Fortran, since much legacy code was written in Fortran. Now C/C++ has become a standard scientific language, and there is an emphasis on object-oriented language. It's probably helpful to have familiarity with both.

    The broad area of application is now 'computational physics', or 'computational multiphysics', in which one attempts to simulate a broad range of physical phenomena over multiple time and length scales, from atoms to planet size, and larger, and picoseconds to billions of years, depending on the physical system being modeled.

    The challenge is to understand the physical phenomena involved and the mathematical equations that describe the system in sufficient detail to allow a reliable or realistic prediction of the behavior.

    See for example - http://farside.ph.utexas.edu/teaching/329/329.pdf

    There is a broad array of computational tools from which to choose:

    In engineering, one would be concerned with movement, heat transfer, fluid flow, electric/magnetic fields (and currents), forces/stress, . . . .

    To this end, there are numerous computational systems, e.g., Comsol, ANSYS, ABAQUS, . . . . The US DOE is supporting a system called MOOSE (based on C++).

    Perhaps one should explore Python as well as exposure to C++ and Fortran.


    So one should explore 'computational physics' and/or 'numerical methods' with language as a qualifier, and see what various universities are teaching.
    e.g., https://courses.physics.ucsd.edu/2017/Spring/physics142/Labs/FinalProject/NumMethods.pdf

    There are still plenty of legacy codes written in Fortran.
  4. Jul 26, 2017 #3
    Thank you for your help and will look into those links shortly. Having a familiarisation with Matlab already I am keen to grow my knowledge of that. Do you think this would be a waste of my time?
  5. Jul 26, 2017 #4


    User Avatar
    Staff Emeritus
    Science Advisor

    Apparently, programs at various universities teach MatLab, or use MatLab in course work. It's probably useful to learn, but one should branch out into other languages, especially if one is interested in modeling and simulation.

    It's best to get into a language like Python and C++ or Fortran and solve systems of equations in order to understand the physics and the mathematics involved in describing the physics. One might find oneself having to work on a problem with something other than MatLab.

    The key areas in nuclear power systems are reactor physics or neutronics, thermal-hydraulics and structural mechanics (including fuel performance). If one does fusion engineering, then plasma physics would be in the reactor physics. In the past, detailed simulations were done more or less separately, but the trend has been to couple these areas. There are other related areas like radiation effects on materials.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted