DOI: 10.55176/2414-1038-2020-4-26-32
Authors & Affiliations
Gomin E.A., Davidenko V.D., Shirokov R.V.
National Research Centre “Kurchatov Institute”, Moscow, Russia
Gomin E.A. – Senior Researcher.
Davidenko V.D. – Department Head, Dr. Sci. (Techn.) . Contacts: 1, pl. Kurchatov, Moscow, Russia, 123182. Tel.: +7 (499) 196-96-73; e-mail:
Shirokov R.V. – Senior Researcher, Dr. Sci. (Phys.-Math.).
Abstract
The article presents the results of testing the KIR code designed to solve the neutron transfer equation by the Monte Carlo method, based on available experimental data from the international Bank ICSBEP on solution benchmarks. The KIR program is part of the DAREUS program complex designed for modeling dynamic processes in solution reactors. Computer models of 13 critical assemblies were created. The total number of calculated configurations was 137. The results were analyzed and the calculation errors were determined both in comparison with the experiment and with other programs. In most cases, the calculated keff multiplication factor falls within the experimental error. The obtained calculation results show a mean square deviation of the multiplication coefficient of 0.7 % for all calculated configurations of experimental assemblies. However, for a number of assemblies, there are significant differences in the assessment. Similar deviations in the calculated values of keff were observed in a variety of previously performed calculations for other codes. The article presents the main factors affecting the results of calculations. Mainly they are related to the insufficiently complete description of experiments submitted to the ICSBEP International Bank.
Keywords
Monte Carlo method, solution reactor, benchmark, multiplication factor, statistical error, experimental
error
Article Text (PDF, in Russian)
UDC 621.039
Problems of Atomic Science and Technology. Series: Nuclear and Reactor Constants, 2020, issue 4, 4:3