Assessment of Monte Carlo N-Particle variance reduction techniques for small solid angle neutron transport

A benchmark experiment for a small solid angle neutron transport system was designed and measured at The University of Texas's Nuclear Engineering Teaching Laboratory. The benchmark problem consists of an MP320 14.1MeV DT Neutron Generator streaming fast neutrons through an opening in a graphite rod. The flux at the far end of the graphite was measured with a Holmium activation foil and used to calculate the absolute source strength of the DT Generator. The experimental measurements were then compared to an analog MCNP simulation of the experimental setup to verify the benchmark MCNP problem was accurate with real world phenomena. Once the analog benchmark was verified, variance reduction methods were included in the MCNP simulation both independently and together in order to evaluate their effectiveness for small solid angle neutron transport problems. Multiple variance reduction techniques were identified to be effective when implemented individually. However, significant performance gains were achieved by implementing multiple complimentary variance reduction techniques in combination.