# Nuclear Science and Techniques

《核技术》(英文版) ISSN 1001-8042 CN 31-1559/TL     2019 Impact factor 1.556

Nuclear Science and Techniques ›› 2014, Vol. 25 ›› Issue (1): 010602

• NUCLEAR ENERGY SCIENCE AND ENGINEERING •

### A neural network to predict reactor core behaviors

Juan Jos´e Ortiz-Servin, David A. Pelta, Jos´e Alejandro Castillo

1. 1Instituto Nacional de Investigaciones Nucleares, Carretera Mexico Toluca S/N, La Marquesa Ocoyoacac, Estado de Mexico, CP 52750, Mexico.
2ETS Ingenier´?a Inform´atica y Telecomunicaciones, Universidad de Granada, C/Daniel Saucedo Aranda, s/n 18071, Granada, Spain.
3Instituto Nacional de Investigaciones Nucleares, Carretera Mexico Toluca S/N, La Marquesa Ocoyoacac, Estado de Mexico, CP 52750, Mexico.
• Contact: Juan Jos′e Ortiz-Servin E-mail:juanjose.ortiz@inin.gob.mx;
• Supported by:

Supported in part by Campus CEI-BioTic GENIL, from University of Granada. D. Pelta acknowledges support from Projects TIN2011- 27696- C02-01 from the Spanish Ministry of Economy and Competitiveness and P11- TIC-8001 from Andalusian Government. The authors gratefully acknowledge the Departamento de Gesti´on de Combustible of the Comisi´on Federal de Electricidad of M´exico, the support given by CONACyT from Mexico, through the research project CB-2011-01-168722 and the ININ through the research project CA-215.

Juan Jos′e Ortiz-Servin, David A. Pelta, Jos′e Alejandro Castillo. A neural network to predict reactor core behaviors.Nuclear Science and Techniques, 2014, 25(1): 010602

Abstract:

The global fuel management problem in BWRs can be understood as a very complex optimization problem, where the variables represent design decisions and the quality assessment of each solution is done through a complex and computational expensive simulation. This last aspect is the major impediment to perform an extensive exploration of the design space, mainly due to the time lost evaluating non promising solutions. In this work, we show how we can train a Multi-Layer Perceptron (MLP) able to predict the reactor behavior for a given configuration. The trained MLP is able to evaluate the configurations immediately, thus allowing performing an exhaustive evaluation of the possible configurations derived from a stock of fuel lattices, fuel reload patterns and control rods patterns. For our particular problem, the number of configurations is approximately 7.7×1010; the evaluation with the core simulator would need above 200 years, while only 100 hours were required with our approach to discern between bad and good configurations. The later were then evaluated by the simulator and we confirm the MLP usefulness. The good core configurations reached the energy requirements, satisfied the safety parameter constrains and they could reduce uranium enrichment costs.

Key words: BWR, Neural Networks, Optimization