Analysis of the AMMON experimental program in the EOLE facility supporting the qualification of the JHR neutron and photon tools

Abstract

The first divergence of the international Jules Horowitz Material Testing Reactor under construction at CEA Cadarache is foreseen in 2016. In order to perform all the design and safety studies, a specific neutron and photon calculation tool, HORUS3D/N&P, dedicated to the JHR simulation, was developed. This development follows the V&V-UQ process; in this frame, the validation and qualification steps aim at quantifying all the biases and associated uncertainties of HORUS3D/N&P calculations. These biases originate from both the nuclear data and the calculation scheme, for calculations at step zero (JHR BOL core) or during depletion (JHR core at equilibrium). This paper focusses on the quantification of the biases due to nuclear data and the associated uncertainties. A new experimental program, named AMMON, was performed between 2010 and 2013 to provide experimental data for this quantification. A large part of the measurements has been today analyzed with the Monte Carlo TRIPOLI-4 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> code and the JEFF3.1.1 and EPDL97 data libraries relative to neutron/photon transport and photon emission. The C/E results were transposed from AMMON to JHR. The synthesis of the present analysis is very satisfactory. The calculation biases remains limited. In addition, thanks to a very good control to the experimental uncertainties during measurements, it is shown that the a priori uncertainties due to nuclear data for parameters at step zero was reduced up to a factor of 2.