This research is devoted to the identification of stochastic loads applied to fuel assemblies using an uncertain computational model and experimental measurements of responses. The stochastic loads applied to the structure are induced by a turbulent flow. The structure is made up of a nonlinear complex dynamical system. The experimental responses of the structure are obtained from strain sensors located on the structure. There are several sources of uncertainties in this experimental identification problem of the stochastic loads: uncertainties on the nonlinear dynamical computational model of the structure (fuel assemblies), uncertainties on parameters of the mathematical model of the stochastic loads themselves and finally, measurements errors. All these sources of uncertainties are identified and taken into account in the identification process of the stochastic loads. Then, the stochastic nonlinear dynamical computational model of fuel assemblies on which the identified stochastic loads are applied yield interesting results concerning the robustness of the estimation of the frettingwear of fuel rods.
