ALGORITHM BASED ON ARTIFICIAL BEE COLONY FOR UNFOLDING OF NEUTRON SPECTRA OBTAINED WITH BONNER SPHERES

Area of Science:

• Nuclear Physics
• Radiation Detection and Measurement
• Computational Physics

Background:

• Characterizing occupational neutron fields, ranging from thermal to MeV energies, is vital for accurate radioprotection quantity estimation.
• Neutron spectra unfolding, based on limited measurements, is essential for precise radiation safety assessments.
• Traditional unfolding methods may require prior spectral information, limiting their application in unknown fields.

Purpose of the Study:

• To implement and evaluate the Artificial Bee Colony (ABC) algorithm for neutron spectra unfolding.
• To assess the performance of the ABC algorithm in characterizing neutron fields using experimental data.
• To compare the efficacy of the ABC algorithm against a conventional method (SPUNIT) for neutron spectra unfolding.

Main Methods:

• The Artificial Bee Colony (ABC) algorithm, inspired by bee foraging behavior, was adapted for neutron spectra unfolding.
• Experimental measurements were conducted using Bonner spheres and a 6LiI(Eu) detector.
• Irradiations were performed in a thermal neutron flux and three reference fields: 241Am-Be, 252Cf, and 252Cf (D2O).

Main Results:

• The ABC algorithm successfully unfolded neutron spectra with good accuracy, even without prior spectral information.
• Results from the ABC algorithm demonstrated closer agreement with expected spectra compared to the SPUNIT algorithm.
• The study validated the capability of the ABC algorithm in characterizing diverse neutron fields.

Conclusions:

• The Artificial Bee Colony (ABC) algorithm is a viable and effective tool for neutron spectra unfolding in occupational settings.
• The ABC algorithm offers an advantage over traditional methods by not requiring prior spectral knowledge.
• This approach enhances the accuracy of radioprotection quantity estimations in complex neutron fields.