Genetic Algorithms: A New Method for Neutron Beam Spectral Characterization
A revolutionary new concept for solving the neutron spectrum unfolding problem using genetic algorithms (GAs) has recently been introduced. GAs are part of a new field of evolutionary solution techniques that mimic living systems with computer-simulated chromosome solutions that mate, mutate, and evolve to create improved solutions. The original motivation for the research was to improve spectral characterization of neutron beams associated with boron neutron capture therapy (BNCT). The GA unfolding technique has been successfully applied to problems with moderate energy resolution (up to 47 energy groups). Initial research indicates that the GA unfolding technique may well be superior to popular unfolding methods in common use. Research now under way at Kansas State University is focused on optimizing the unfolding algorithm and expanding its energy resolution to unfold detailed beam spectra based on multiple foil measurements. Indications are that the final code will significantly outperform current, state-of-the-art codes in use by the scientific community.
- Research Organization:
- Kansas State University, Manhattan, KS (US)
- Sponsoring Organization:
- none (US)
- OSTI ID:
- 786188
- Report Number(s):
- ISSN 0003-018X; CODEN TANSAO; ISSN 0003-018X; CODEN TANSAO; TRN: US0109305
- Resource Relation:
- Conference: 2000 Annual Meeting, San Diego, CA (US), 06/04/2000--06/08/2000; Other Information: Transactions of the American Nuclear Society, Volume 82; PBD: 4 Jun 2000
- Country of Publication:
- United States
- Language:
- English
Similar Records
Avoiding excess computation in asynchronous evolutionary algorithms
Genetic algorithms and their use in Geophysical Problems