EDN: BVSMPE
Authors & Affiliations
Pasko D.V., Smolnikov N.V., Anikin M.N., Naimushin A.G., Lebedev I.I., Ushakov I.A.
National Research Tomsk Polytechnic University, Tomsk, Russia
Pasko D.V. – Master's Student. Contacts: 30, Lenin Ave., Tomsk, Russia, 634050. Tel.: +7 (923) 412-75-58; e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it..
Smolnikov N.V. – Postgraduate.
Naimushin A.G. – Associate Professor, Cand. Sci. (Phys.-Math.).
Anikin M.N. – Assistant Professor, Cand. Sci. (Tech.).
Lebedev I.I. – Assistant.
Ushakov I.A. – Senior Lecturer.
Abstract
One of the most important neutronic parameter of small-size heterogeneous nuclear reactors is the power density distribution (PDD). Control and minimization of PDD heterogeneity is an approach to achieve high discharge burnup of fuel and reduce its cost. In case of nuclear research reactors that use partial refueling pattern to refuel spent fuel, PDD changes significantly every fuel cycle. However, heterogeneity of PDD can be reduced by optimization of fuel loading pattern and permutation steps.
In this paper artificial intelligence package based on supervised machine learning models and genetic algorithm is proposed to process, evaluate and find desired optimum fuel loading pattern. Genetic algorithm performs stochastic optimization search while machine learning models used to generate PDD for different fuel assemble combinations within reactor core. Fitness function of genetic algorithm consists of independent objective functions among which are reactor reactivity margin, power peaking factor and symmetry criteria for core parts. It is shown that developed genetic algorithm is able to generate and evaluate more than 25000 fuel patterns in 3–6 minutes and find optimum solution among them.
Keywords
power density distribution, fuel loading pattern, power peaking factor, machine learning, metaheuristic optimization, genetic algorithm, computational intelligence, nuclear research reactor IRT-T, modeling, MCU-PTR
Article Text (PDF, in Russian)
References
- Suman S. Artificial intelligence in nuclear industry: Chimera or solution? Journal of Cleaner Production, 2021, vol. 278, p. 124022.
- Glukhov G.G., Didenko A.N. The IRT-T reactor at Tomsk Polytechnical Institute Nuclear Physics Research Institute: Research and applications. Soviet Atomic Energy, 1988, vol. 64, issue 5, pp. 423–426.
- Shchurovskaya M.V. et al. Validation of the MCU-PTR computational model of beryllium poisoning using selected experiments at the IRT-T research reactor. Annals of Nuclear Energy, 2018, vol. 113, pp. 436–445.
- Shchurovskaya M.V. Control rod calibration simulation using Monte Carlo code for the IRT-type research reactor. Annals of Nuclear Energy, 2016, vol. 96, pp. 332–343.
- Chertkov Y.B., Anikin M.N., Lebedev I.I., Naimushin A.G., Smol’nikov N.V. et al. Calculation and Experimental Determination of the Neutronics Characteristics of the IRT-T Research Reactor. At Energy, 2021, vol. 131, issue 1, pp. 42–45. DOI: https://doi.org/10.1007/s10512-022-00834-y.
- Kumar A., Tsvetkov P.V. A new approach to nuclear reactor design optimization using genetic algorithms and regression analysis. Annals of Nuclear Energy, 2015, vol. 85, pp. 27–35.
- Jayalal M.L. Application of Genetic Algorithm methodologies in fuel bundle burnup optimization of Pressurized Heavy Water Reactor. Nuclear Engineering and Design, 2015, vol. 281, pp. 58–71.
- Ahmad A., Ahmad S.-I. Optimization of fuel loading pattern for a material test reactor using swarm intelligence. Progress in Nuclear Energy, 2018, vol. 103, pp. 45–50.
- Shams S., Azgomi H., Asghari A. Fuel assemblies loading pattern optimization of pressurized water reactors using the trees social relations algorithm. Annals of Nuclear Energy, 2023, vol. 192, p. 109963.
- Katoch S., Chauhan S.S., Kumar V. A review on genetic algorithm: past, present, and future. Multimedia Tools and Applications, 2021, vol. 80, no. 5, pp. 8091–8126.
- Slowik A., Kwasnicka H. Evolutionary algorithms and their applications to engineering problems. Neural Computing and Applications, 2020, vol. 32, no. 16, pp. 12363–12379.
- Alhijawi B., Awajan A. Genetic algorithms: theory, genetic operators, solutions, and applications. Evolutionary Intelligence, 2024, vol. 17, no. 3, pp. 1245–1256.
UDC 621.039.52
Problems of Atomic Science and Technology. Series: Nuclear and Reactor Constants, 2025, no. 2, 2:2
