Interview - Gérard Cattiaux

06th February, 2014


IRSN (the French National Institute of Radiation Protection and Nuclear Safety) is the national public expert on nuclear and radiological risks. It contributes to the safety of nuclear facilities through technical support for nuclear safety authorities. It assists the authorities in charge of the protection of nuclear materials, facilities and transportation against malicious acts. IRSN monitors environmental radiation and worker exposure to ionizing radiation. It conducts on an independent basis any research required to assess risks and contributes to informing the public on these risks.

IRSN is a public authority with industrial and commercial activities, placed under the joint authority of the Ministry for Ecology, Sustainable Development, Transportation and Housing and of the Ministry for the Economy, Finance and Industry, the Ministry for Higher Education and Research, the Ministry of Defense and the Ministry for Employment and Health.


Gérard Cattiaux

First of all, could you tell us in which service you are based in the IRSN? What is your role in this structure? And in which branch do you work?

I work in the Expert evaluation of Equipment and Structures Service of the “Safety of nuclear installations and systems” Pole. One of the missions of this service is to perform expert safety evaluations related to the conception, making, mechanical behavior, in-service control and aging of mechanical equipment and civil engineering crafts and structures, for all the nuclear installations or equipment concerned by safety. This service, composed of many specialists and experts, performs specific expert evaluations in all these areas on behalf of the Nuclear Safety Authorities. It also conducts studies and researches supporting the expertise in the fields of activity of the service.

My role as an expert is to perform expert evaluations in the field of Non Destructive Testing (NDT), and to initiate researches in this area to support expert evaluations, which may include digital simulation and the development of new means of control, for demonstration purposes before industrialization.


When did you start using CIVA and what convinced you to use simulation?

I started to use CIVA at its early beginning, as soon as the first PC-oriented versions were proposed. In fact, IRSN wished to have digital simulation tools to evaluate by its own means the performances of NDT methods when the methodology for qualification of testing methods has been set up in Europe. This methodology was defined in the early 90’s, in the frame of the work of the European Network for Inspection Qualification (ENIQ) and the Nuclear Regulators Working Group (NRWG).

In this context, IRSN supported financially, orientated its work, and contributed scientifically to the development and the implementation of simulation tools in CIVA for uses specific to the nuclear domain, in cooperation with the development teams of the CEA. The conclusive and very encouraging results obtained at the time, especially for ultrasound simulation, led IRSN to pursue and intensify the collaborative works, and to extend the works of digital simulation to Ultrasound, Eddy current and Radiography techniques.
All IRSN works achieved in the frame of the use and the validation of the models, relative to all the techniques, are accompanied by experimental measurements on mock ups, to have elements of comparison between simulation and experimentation.


Which advantages did it bring to your work? (Saving time? Lowering costs? Better precision?) Do you associate it with safety criteria?

CIVA allows in many cases to assess the performances of Non Destructive Testing methods, and also to study their limits, by changing some influent parameters. Regarding IRSN, the tool mostly allows us to support recommendations generally formulated during technical expertise evaluations, after the study of the technical documents of industrials. This contributes to improving safety.
A good appreciation of the validity of the models and their limits of application is essential. This goes necessarily through experimental validations and a good knowledge of the comparisons between experimentation and simulation.


What would you like to see in the future versions of CIVA?

IRSN contributes actively to the development of models dedicated to uses specific to the nuclear industry, in the fields of Ultrasounds, Eddy Currents and Radiography, through its multi-year research programs realized in collaboration with the CEA development teams. Because of that, IRSN has a good vision of the new functionalities or future improvements that will result from the works it has initiated. As an example, the modeling of internationally used Eddy Current probes like the + Point probe will soon be possible. In another field, more specifically radiography, simulations with digital films for which experimental validations are currently running should also be proposed in a future version.
In parallel to these developments, I think it is necessary to orient more efforts towards the improvement of the Man to Machine interface, in order to make the definition of specimens or defects easier for the users. For example, it would be a good thing to be able to create and manipulate groups of defects, including defects with a complex shape, through the use of a CAD software, before incorporating them in CIVA.
Moreover, simple defects with ramifications (Y shaped or others) are not correctly taken into account when computing the beam-defect interactions when they are created with a CAD software. These same defects are correctly interpreted when they are directly defined through the CIVA interface. Whichever way is used to create a defect, this defect should be correctly taken into account in CIVA.
Furthermore, in cases of computations at the limit of the use of the models, more warnings to the attention of the users could pop up, in order to better point out the level of confidence about the results obtained with this computation, relatively to the experimental validations realized or to be realized.


Which tool do you particularly like in CIVA currently?

I mainly use the Ultrasound simulation module, because the technique is widely used in the nuclear industry, and leads to a large number of diverse application cases. I also use the radiography simulation module, with a growing interest as new features are integrated.