About us
Espace utilisateur
INSTN offers more than 40 diplomas from operator level to post-graduate degree level. 30% of our students are international students.
Professionnal development
Professionnal development
Find a training course
INSTN delivers off-the-self or tailor-made training courses to support the operational excellence of your talents.
Human capital solutions
At INSTN, we are committed to providing our partners with the best human capital solutions to develop and deliver safe & sustainable projects.
Home   /   Thesis   /   Detection of anomalies in PWR nuclear reactors cores by artificial intelligence methods using neutron noise measurements

Detection of anomalies in PWR nuclear reactors cores by artificial intelligence methods using neutron noise measurements

Engineering sciences Mathematics - Numerical analysis - Simulation


In the sequel of the studies carried out in the framework of the European CORTEX project, the CEA proposes a research work in artificial intelligence relating to the detection of anomalies in a PWR (Pressurized Water Reactor) reactor.

The objective of the European project EC H2020 CORTEX (CORe monitoring Techniques and EXperimental validation and demonstration) [1] was to assess the feasibility of methods for exploiting "neutron noise" in a nuclear reactor for the early detection of operating anomalies, such as mechanical vibrations of fuel assemblies, disturbances in heat transfer fluid flow rates or other phenomena, and if possible to locate them. Although promising, the CORTEX results showed that neural network models based solely on simulated data were not enough and that it was necessary to adapt them to better take into account real measurement conditions, work that could not be fully carried out before the end of the project.

The objective of this thesis is to look for new methods and tools that would make it possible to better take into account the cross-evolutions of the signals acquired by detectors placed inside the reactor, in order to obtain convincing diagnoses. The thesis will also address the delicate question of pre-processing that will have to be applied to the real measurements acquired in the power reactor before their introduction into the tools developed on the simulated data. Measurements are already available (disseminated as part of the CORTEX project). During the thesis, other measurements acquired in power reactor could be available to test the analytical methods that will have been developed.

The PhD student will work within a team of experimental physicists, which will allow him to acquire skills on the physical phenomena that we want to measure and therefore to better understand the signals that will be analyzed. He will be involved in the European BRAIN project which should follow the CORTEX project. As his work will focus on the adaptation of artificial intelligence models to real data acquired in the reactor, he will have to compare his own approach with those of specialists in artificial intelligence.

[1] http://cortex-h2020.eu/


Département Etude des Réacteurs
Service de Physique expérimentale, d’essais en Sûreté et d’Instrumentation
Laboratoire des programmes expérimentaux et d’essais en sûreté
Top pencilenvelopegraduation-hatlicensebookuserusersmap-markercalendar-fullbubblecrossmenuarrow-down