Use of 3D-VAR and Kalman Filter Approaches for Neutronic State and Parameter Estimation in Nuclear Reactors

In this paper, data assimilation (DA) techniques that have proven to be efficient in the fields of meteorological forecast and oceanography are applied to neutronics problems. Two applications, the techniques of which can be used to enhance optimization, are presented: three-dimensional (3-D) neutronic field interpolation in online core monitoring, and parameter estimation in code qualification procedures. First, the main bases of DA theory are shortly presented. Calibration and estimation procedures that are in use today at Electricité de France (EDF) are then briefly introduced. We also analyze the main limitations of these procedures and the potential improvements that the use of various DA techniques can provide. We present the MANARA mock-up, which computes a 3-D field interpolation and has been implemented in PALM, a DA-dedicated coupling platform developed at the Centre Européen de Recerche et de Formation Avancée en Calcul Scientifique (CERFACS). Result validation and comparison with former interpolation procedure CAMARET are also presented. Next, the principles of the KAFEINE mock-up, based on an extended Kalman filter approach, are displayed. This application covers the field of optimal parameter calibration using the in-core measures. In conclusion, these first two applications to neutronics, carried out in partnership between CERFACS and EDF Research and Development, seem very promising, especially considering the new generation of neutronic solvers that are being developed in the frame of the DESCARTES project.