Criticality accident alarm system analysis using MCNP6.2 constructive solid geometry/unstructured mesh hybrid

Criticality Accident Alarm System (CAAS) design, analysis, and simulation require a combination of methodologies and techniques used for both criticality and shielding problems. MCNP6.2 is a general-purpose Monte Carlo radiation transport package with continuous-energy neutron and photon physics making it suitable for CAAS analysis and simulation. Use of MCNP6 for analysis of criticality accident alarm systems has typically been done by building constructive solid geometry (CSG) models that may include extensive effort spent in specification. In this work a constructive solid geometry/unstructured mesh hybrid model is created for use with MCNP6.2. The model is chosen for appropriateness to demonstrate the MCNP6 CSG/UM capability for solving problems relevant to CAAS analysis and for comparison to traditional CSG models used for analyzing criticality accident alarm systems. CAAS problems typically require the use of variance reduction techniques. Methods designed to increase the efficiency of the calculations are discussed. Deterministic variance reduction methods available in Atilla10.0 are applied to the hybrid CSG/UM model. Monte Carlo/deterministic hybrid variance reduction, in which FW-CADIS (forward-weighted consistently adjoint driven importance sampling) is used on the mesh in a deterministic calculation to generate weight windows and source biasing parameters which are applied in the MCNP6.2 simulations. Results for CAAS analysis are presented and compared for MCNP6.2 unstructured mesh, constructive solid geometry, and unstructured mesh/constructive solid geometry hybrid models. Visualization of MCNP6.2 results, including use of the MCNP6 UM elemental edit output files and FMESH tallies over the geometry under consideration is presented. (authors)