2017 Annual Meeting

Presentation information

Oral presentation

III. Fission Energy Engineering » 307-1 Computational Science and Engineering

[1G01-07] Numerical simulation and Visualization of Thermal Fluid Dynamics and Nuclear Fuel

Mon. Mar 27, 2017 10:00 AM - 11:55 AM Room G (16-207 Building No.16)

Chair: Nakajima Norihiro (JAEA)

11:15 AM - 11:30 AM

[1G06] Study on Fuel Loading Pattern Optimization using Deep Learning and Reinforcement Learning

(1)Prediction of Nuclear Characteristics of Reactor Core with Loading Pattern Infromation

*Masahiro Tatsumi1 (1. NEL)

Keywords:Loading Pattern Optimization, Deep Learning, Reinforcement Learning, Neural Networks, Nuclear Characteristics

We constructed a deep layer neural network with infinite multiplication factor vector as input and realized high speed core characteristic evaluation. We showed the possibility of evaluating the combustion characteristics of the core, with the infinite multiplication factor of the next step predicted as a new input.