Attila4MC was developed to help MCNP® users become more productive. While the accuracy of MCNP is widely established, it relies on command line input for defining model geometry, calculation input, and variance reduction, which can often create significant bottlenecks limiting calculation throughput. Attila4MC provides an easy-to-use graphical interface, allowing novice and advanced MCNP users to easily set up, run, and visualize MCNP solutions from CAD data. Attila4MC leverages the state-of-the-art Attila deterministic solver to easily generate highly efficient weight windows variance reduction, shortening MCNP calculation times and substantially reducing user investment over manual variance reduction techniques. Once a calculation is performed, Attila4MC provides intuitive and advanced solution visualization options, aiding solution verification and allowing solution field insights to be easily conveyed to external parties.
Attila4MC benefits include:
Robust CAD Integration
A major bottleneck when using MCNP is the creation of Constructive Surface Geometry (CSG), which can be extremely time consuming to both build and verify. Attila4MC provides CAD integration with MCNP, allowing 3D solid models to be imported into Attila4MC, meshed, then exported in a format readable by the unstructured mesh capabilities in MCNP. This approach is consistent with state-of-the-art technology in well-established computer aided engineering (CAE) fields such as finite element analysis (FEA) and computational fluid dynamics (CFD), and enables MCNP to be integrated more effectively into the design cycle.
SpaceClaim® is offered as an option with Attila4MC and Attila, providing users with easy-to-use 3D solid modeling and best-of-class geometry cleanup and simplification features. With SpaceClaim, users can import a broad range of CAD model formats and uniquely modify the imported CAD data as if it was originally constructed in SpaceClaim. Hundreds of bolt holes can be filled with a single command, as can removal of small parts and features such as fillets and chamfers.
Where pre-existing CAD data is not available, SpaceClaim offers simple to use features for rapidly building complex models that are ideal for part-time users.
MCNP Calculation Setup
Attila4MC provides an intuitive, process based graphical user interface (GUI) for setting up MCNP calculations. Most fixed source calculations can be setup entirely through the GUI, including volume source definitions, material creation, material-to-region assignments, solver controls, variance reduction, and tally specifications. To aid in verification, regions can be visualized by material assignment, allowing regions to be visualized by material type. Time saving features, such as automated material-to-region assignments based on CAD part names, help minimize user errors and streamline the setup of complex models.
The Attila4MC generated MCNP input decks are heavily commented and easily readable, facilitating verification and modification for advanced cards not supported in the GUI.
MCNP can be directly run from within the Attila4MC GUI, or Attila4MC can automatically pack the MCNP input files to be run on another machine.
CADIS & FW-CADIS Variance Reduction
Attila4MC supports CADIS1 and FW-CADIS2 variance reduction powered by the deterministic Attila solver. The CADIS method in Attila automatically generates optimized weight windows from an Attila adjoint calculation, including energy source biasing. CADIS is ideal for deep penetration cases where the solution is desired at a single detector, or multiple detectors in near vicinity. FW-CADIS also employs a forward Attila calculation, to generate weight windows that are optimized for multiple detectors in disparate locations, large regions, or even the entire computational domain. FW-CADIS is ideal for cases where it is desired to visualize the MCNP solution everywhere, which traditionally has been very challenging for Monte Carlo methods.
Both CADIS and FW-CADIS can be fully set up from within the Attila4MC GUI, and require only a handful of additional steps relative to an analog Attila4MC calculation. For deep penetration cases, CADIS and FW-CADIS weight windows can achieve Figures of Merit (FOMs) as much as 4 or 5 orders of magnitude higher than analog MCNP.
Images show a scenario with a Co-60 source inside a 30 cm thick steel cask, inside a 100 cm thick concrete walled room. FW-CADIS was used to calculate dose in the outer room, where the dose drops off ~13 orders of magnitude from inside the cask. MCNP using Attila4MC was run with 5.0E+07 histories. The entire calculation, including deterministic calculation time for weight window generation, was accomplished in 1 hour on a laptop with 8 cores.
1 J Wagner, A Haghighat, “Automated Variance Reduction of Monte Carlo Shielding Calculations Using the Discrete Ordinates Adjoint Function”, Nuclear Science and Technology, Vol 128(2), Feb 1998
2 J Wagner, D Pepolow, S Mosher, “FW-CADIS Method for Global and Regional Variance Reduction of Monte Carlo Radiation Transport Calculations”, Nuclear Science and Technology, Vol 176(1), 2014
Attila4MC automatically converts the output from MCNP to the Tecplot 360® format, which is included as an option with Attila4MC. Through Attila4MC, global MCNP solutions can be visualized directly on the unstructured tetrahedral mesh, allowing locally adaptive, body-fitted visualization not possible with traditional Cartesian mesh tallies. Additionally, the Attila4MC GUI supports the setup of tetrahedral mesh based tally definitions, allowing quantities such as flux, energy deposition, or user defined response functions to be output for each tetrahedral element. Statistical uncertainties can additionally be output, allowing users to visually assess local and global solution convergence.
Solutions can be visualized through a broad range of methods, including section planes, contour plots, iso-surfaces, and animations. Collectively, these features provide users with valuable solution field insight, which can be used to make more informed design decisions, verify that the calculation was correctly setup, and effectively convey solution data to third parties.
Attila® is a trademark of Silver Fir Software
MCNP® and Monte Carlo N-Particle® are registered trademarks owned by Triad National Security, LLC, manager and operator of Los Alamos National Laboratory.
SpaceClaim® is a trademark of ANSYS, Inc. Tecplot 360® is a trademark of Tecplot, Inc.