Issue |
2014
SNA + MC 2013 - Joint International Conference on Supercomputing in Nuclear Applications + Monte Carlo
|
|
---|---|---|
Article Number | 05110 | |
Number of page(s) | 2 | |
Section | 5. Poster Session: a. Computational Nuclear Applications | |
DOI | https://doi.org/10.1051/snamc/201405110 | |
Published online | 06 June 2014 |
Feasibility study for a realistic training dedicated to radiological protection improvement
1 EDF-R&D, 6 quai watier, 78400 Chatou, France
2 EDF-DPN-UNIE, 1 place Pleyel, 93200 Saint-Denis, France
3 EDF DPN-CIV, BP 64, CNPE CIVAUX, 86320 Civaux, France
* Estelle Courageot, E-mail: Estelle.courageot@edf.fr
Any personnel involved in activities within the controlled area of a nuclear facility must be provided with appropriate radiological protection training. An evident purpose of this training is to know the regulation dedicated to workplaces where ionizing radiation may be present, in order to properly carry out the radiation monitoring, to use suitable protective equipments and to behave correctly if unexpected working conditions happen. A major difficulty of this training consist in having the most realistic reading from the monitoring devices for a given exposure situation, but without using real radioactive sources. A new approach is developed at EDF R&D for radiological protection training. This approach combines different technologies, in an environment representative of the workplace but geographically separated from the nuclear power plant: a training area representative of a workplace, a Man Machine Interface used by the trainer to define the source configuration and the training scenario, a geolocalization system, fictive radiation monitoring devices and a particle transport code able to calculate in real time the dose map due to the virtual sources. In a first approach, our real-time particles transport code, called Moderato, used only an attenuation low in straight line. To improve the realism further, we would like to switch a code based on the Monte Carlo transport of particles method like Geant 4 or MCNPX instead of Moderato. The aim of our study is the evaluation of the code in our application, in particular, the possibility to keep a real time response of our architecture.
Key words: Realistic training / Monte Carlo code
© Owned by the authors, published by EDP Sciences, 2014