A3MCNP
Automatic Adjoint Accelerated MCNP


Developed by:

John C. Wagner
Alireza (Ali) Haghighat
Oak Ridge National Laboratory
University of Florida
P.O. Box 2008, Bldg 6011
Department of Nuclear & Radiological Engineering
Oak Ridge, TN 37831-6370
Gainesville, FL 32611
 Phone: 865 241-3570   FAX: 865 576-3513
 Phone: 352 392-1401   FAX 352 392-3380
Email: wagnerjc@ornl.gov
Email: haghighat@ufl.edu
H&S Advanced Computing Technologies

    Greetings and Welcome to the A3MCNP web page !

What is A3MCNP?

     A3MCNP is an enhanced version of the general-purpose Monte Carlo radiation transport code MCNP that features automatic variance reduction for accelerating Monte Carlo shielding calculations.  Numerous applications have already been efficiently solved with A3MCNP.

1. Introduction

     A3MCNP automatically generates and utilizes consistent source biasing parameters and weight window lower bounds based on a three-dimensional
     discrete ordinates adjoint functions for accelerating Monte Carlo shielding calculations.  From the MCNP input file (with a few additional parameters),
     A3MCNP does the following:
  1. Generates an input file for the 3-D discrete ordinates TORT code. This includes spatial mesh generation with a "back-thinning" feature for effective mesh economy (see example below).
  2. Generates an input file for the cross-section mixing code GIP and executes GIP to generate mixed cross sections for the adjoint TORT calculation.
  3. Reads the adjoint function from the TORT binary flux file and couples the original (MCNP) source distributions with the adjoint function to calculate consistent source biasing parameters and weight window lower bounds based on the CADIS methodology.
  4. Performs the transport calculation using the calculated source biasing parameters and weight window lower bounds. The discrete ordinates spatial mesh and energy group structure are used within A3MCNP to facilitate the space- and energy-dependent importance distribution (weight windows), and thus, cell subdivision for the purposes of specifying spatial importances is not necessary.

2. Mesh Generation Example

     This simplified shipping cask problem demonstrates the mesh generation capabilities of A3MCNP. From the MCNP input geometry, a TORT mesh
     distribution is automatically generated. The following figures (which were generated by A3MCNP) illustrate TORT mesh distributions generated by
     A3MCNP for this problem.

3. Related Publications

    Applications of A3MCNP

  1. A. HAGHIGHAT, G.E. SJODEN and J.C. WAGNER, "Advanced 3-D Deterministic and Monte Carlo Codes for Simulation of Real-Life Complex Nuclear Systems," Trans. Am Nuc. Soc., 82, June 2000.

  2. A. HAGHIGHAT, H. HIRUTA and B. PETROVIC, "Performance of A3MCNPTM for Calculation of 3-D Neutron Flux Distribution in a BWR Core Shroud," Reactor Dosimetry, ASTM STP 1398, John G. Williams, David W. Vehar, Frank H. Ruddy and David M. Gilliam, Eds., American Society for Testing and Materials, West Conshohoken, PA, 2000.

  3. A. HAGHIGHAT,  J.C. WAGNER, E.L. REDMOND II, and S. ANTON, "Performance of A3MCNP for Determination of Neutron and Gamma Dose on a Shipping Cask Surface," submitted for presentation at the Radiation and Protection Conf., 2000, Spokane, WA.

  4. A. PATCHIMPATTAPONG and A. HAGHIGHAT, "Effectiveness of A3MCNP for a Purely Absorbing Medium with Void Region," Trans. Am Nuc. Soc., 81, pp. 257-260, Nov. 14-18, 1999.

  5. A. HAGHIGHAT, H. HIRUTA, B. PETROVIC, and J.C. WAGNER, "Performance of the Automated Adjoint Accelerated MCNP (A3MCNP) for Simulation of a BWR Core Shroud Problem," Proceedings of the International Conference on Mathematics and Computation, Reactor Physics, and Enviromental Analysis in Nuclear Applications, Madrid, Spain, September 27-30, 1999.

  6. J.C. WAGNER and A. HAGHIGHAT, "Monte Carlo PWR Cavity Dosimetry Calculations using an Automatic Variance Reduction Technique," invited, Joint Int. Conf. on Mathematical Methods and Supercomputing for Nuclear Applications, Saratoga Springs, October 6-10, 1997.

  7. J.C. WAGNER and A. HAGHIGHAT, "Acceleration of Monte Carlo Reactor Cavity Dosimetry Calculations with the Discrete Ordinates Adjoint Function," Ninth Intl. Symposium on Reactor Dosimetry, Prague, Czech Republic, September 2-6, 1996.

    Development and Methods of A3MCNP

  1. J.C. WAGNER and A. HAGHIGHAT, "Automatic Variance Reduction of Monte Carlo Shielding Calculations Using the Discrete Ordinate Adjoint Function," Nucl. Sci. Eng., 128, 186 (1998).

  2. J.C. WAGNER, " Acceleration of Monte Carlo Shielding Calculations with an Automated Variance Reduction Technique and Parallel Processing," PhD Thesis, The Pennsylvania State University, Nuclear Engineering (1997).

  3. J.C. WAGNER and A. HAGHIGHAT, "Automatic Variance Reduction for Monte Carlo Shielding Calculations with the Discrete Ordinates Adjoint Function," Joint Int. Conf. on Mathematical Methods and Supercomputing for Nuclear Applications, Saratoga Springs, October 6-10, 1997.

  4. J.C. WAGNER and A. HAGHIGHAT, "Application of the Discrete Ordinates Adjoint Function to Accelerating Monte Carlo Reactor Cavity Dosimetry Calculations," 1996 Radiation Protection & Shielding Topical Meeting, No. Falmouth, MA, 1 345-398 (Apr. 1996).

  5. J.C. WAGNER and A. HAGHIGHAT, "Deterministic Adjoint Functions for Biasing Monte Carlo Reactor Cavity Dosimetry Calculations," Trans. Am. Nucl. Soc., 73, 432-434 (1995).

4. Additional Information

     Additional discussion, including some examples of computational performance, can be found here.

5. Availability

     Consult the following link for availability: H&S Advanced Computing Technologies

John C. Wagner wagnerjc@ornl.gov