Variance Reduction with MCNP
Class Length: 4 days (See schedule)
Time: 8:30 a.m. to 4:30 p.m.
Place: Los Alamos National Laboratory
Number accepted in class: 25
Minimum enrollment required: 10
email address: karinh@lanl.gov
This class focuses on applying variance reduction techniques in MCNP5 for fixed source neutron,
photon, and coupled neutron/photon calculations. It is an intermediate to advanced level class,
intended for those users with at least some familiarity with MCNP and understanding of variance
reduction techniques. Class examples provide hands-on experience at running the code,
applying/tuning variance reduction techniques, working with event logs, and understanding
the code output.
Variance reduction techniques/topics to be covered include:
- Implicit capture and weight cut-off
- Geometry splitting and Russian roulette
- Forced collisions
- Exponential transform
- Weight windows and weight windows generator
- Weight windows and adjoint functions
- Source biasing
- DXTRAN
- Point and Ring detectors
The class includes lectures and hands-on computer use for a variety of variance
reduction calculations. Time will be available to discuss individual questions
and problems with the MCNP5 developers, and to pursue additional variance reduction
topics.
The course material is roughly organized as
Day 1: Introduction and review of basic variance reduction methods
Day 2: Advanced techniques and problems
Day 3: Advanced techniques and problems
Day 4: Advanced hands-on practicum, individual problems, and wrap-up
Course attendees are encouraged to have basic experience working with MCNP.
Attendees should be familiar with working on a Microsoft Windows XP environment
using either Windows or Cygwin tools.
Address all correspondence regarding this class to Karin Hendrickson,
karinh@lanl.gov, phone: 505-606-1740.
Register
