An inverse method for radiation transport

Conference · Thu Jan 01 04:00:00 EST 2004

DOI:https://doi.org/10.1093/rpd/nci204· OSTI ID:977488

Favorite, J A ^[1]; Sanchez, R ^[2]

---

1. Jeffrey A.
2. Richard

Adjoint functions have been used with forward functions to compute gradients in implicit (iterative) solution methods for inverse problems in optical tomography, geoscience, thermal science, and other fields, but only once has this approach been used for inverse solutions to the Boltzmann transport equation. In this paper, this approach is used to develop an inverse method that requires only angle-independent flux measurements, rather than angle-dependent measurements as was done previously. The method is applied to a simplified form of the transport equation that does not include scattering. The resulting procedure uses measured values of gamma-ray fluxes of discrete, characteristic energies to determine interface locations in a multilayer shield. The method was implemented with a Newton-Raphson optimization algorithm, and it worked very well in numerical one-dimensional spherical test cases. A more sophisticated optimization method would better exploit the potential of the inverse method.

• Research Organization: Los Alamos National Laboratory
• Sponsoring Organization: DOE
• OSTI ID: 977488
• Report Number(s): LA-UR-04-1330
• Country of Publication: United States
• Language: English

Related Subjects

61 RADIATION PROTECTION AND DOSIMETRY
BOLTZMANN EQUATION
DATA
OPTIMIZATION
RADIATION PROTECTION
RADIATION TRANSPORT
RADIATIONS
SCATTERING
SHIELDING
TOMOGRAPHY
TRANSPORT