Understanding the radiological inventory of a nuclear installation is paramount to ensure the control of safety and criticality risks throughout the facility's lifespan, as well as to manage the radiological impact on both human health and the environment. Managing the radiological condition of the processes and equipment within an installation enables the definition of safe intervention scenarios to optimize maintenance operations or decontamination/dismantling activities in hostile environments and enhance waste management. During the dismantling of a nuclear facility, decontamination operations are necessary to reclassify areas from nuclear waste zones to conventional waste zones. This step necessitates the prior creation of an accurate map of artificial radioactivity within the concrete civil engineering structures.
To address the limitations of existing methodologies and in accordance with the directives of the Nuclear Safety Authority (ASN), the proposed doctoral research aims to develop an innovative and efficient non-destructive radiological characterization methodology using machine learning algorithms (artificial intelligence) for automated analysis of in-situ gamma spectrometry measurements with a highly resolved Germanium detector. The ultimate goal is to establish real-time three-dimensional maps of contaminant distribution within contaminated civil engineering structures of nuclear facilities.
The characterization methodology developed as a result of this project holds significant potential for industrial applications, particularly in the field of nuclear facility decontamination and dismantling.
The doctoral candidate will work within a team with extensive experience in the development and in situ implementation of non-destructive radiological characterization techniques (alpha and gamma imaging techniques, alpha, beta, and gamma spectrometry techniques). The candidate will have the opportunity to evaluate the proposed solutions on some of the world's largest dismantling sites.
The desired profile is a candidate with an engineering school or Master's degree (M2) background and strong knowledge in nuclear instrumentation and measurement, particularly in the physical phenomena related to the interactions of ionizing radiation with matter. An inclination towards, and preliminary proficiency in, statistical data processing methods and machine learning (computer programming in Python) are likewise highly valued.