Mass excess estimations using artificial neural networks

Ozdogan H., Uncu Y. A., Sekerci M., Kaplan A.

Applied Radiation and Isotopes, cilt.184, 2022 (SCI-Expanded) identifier identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 184
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1016/j.apradiso.2022.110162
  • Dergi Adı: Applied Radiation and Isotopes
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aquatic Science & Fisheries Abstracts (ASFA), Chemical Abstracts Core, Chimica, Compendex, EMBASE, Food Science & Technology Abstracts, INSPEC, MEDLINE, Pollution Abstracts
  • Anahtar Kelimeler: Mass excess, Artificial neural network, Levenberg-marquardt, RIPL3, TALYS 1.95, GROUND-STATE MASSES, NUCLEAR
  • Akdeniz Üniversitesi Adresli: Evet

Özet

Mass excess knowledge is important to investigate the fundamental properties of atomic nuclei. It is a meaningful and important parameter for the determinations of nucleon binding energy, nuclear reaction Q value, energy threshold and plays an undeniable role in the theoretical calculations of a reaction cross-section value in terms of the quantities it affects. In this research, a new artificial neural network (ANN) based algorithm is proposed to determine the mass excess of nuclei. The experimental data, which were taken from the RIPL3 database have been used for training the ANN. Proton, neutron, and mass numbers have been chosen as the input parameters. The Levenberg-Marquardt (LM) algorithm has been employed for the training section. The correlation coefficients have been found as 0.99984, 0.99977, 0.99984, and 0.99983 for training, validation, and testing, respectively. To validate our ANN results, ANN findings have been given as input parameters on TALYS 1.95 code and 56Fe(p,x) nuclear reactions have been simulated. The obtained results were compared with the literature. In conclusion, the findings of this study point to the ANN as a recommended tool that can be used to calculate estimates of mass information.