Fluid Force Identification

Abstract

In this thesis a method to model and identify the force inducing a vibration on the control rods of a nuclear power plant is verified, but could not be validated. The aim of the method is to be able to predict wear in future nuclear power plant designs. This study is based on a study by Bodel [1], and reviews, adjusts and extents its contents. The liquid coolant in a nuclear power plant, flowing passed the control rods induces them to vibrated, wearing them out at the positions of the guide plates. The method aims to model and identify the fluid forces of the liquid coolant, by looking at the dynamic behaviour of a model of the control rod. With this modelled fluid force, predictions can be made on the wear in future power plant designs. To describe the dynamic behaviour of the model of the rod properly, the mode-shapes of the dynamic system need to be mass normalised taking into account the added mass of the water surrounding the model. After the fluid force is modelled and identified it should be validated by applying it to a model which is closer to the real control rod. By comparing the results to the experimental results obtained on this model, the fluid forces can be validated. The results are however, that the experimental mode-shapes were not well mass normalised by the measurement system and the dynamic system in water could therefore not be obtained properly. By using ‘hand’-normalised mode-shapes the study is continued and the method could be verified. Only the modelling of the fluid force needs to be reconsidered, because it can be shown that the way the fluid force is modelled will not give the desired results. The validation of the fluid forces could not by accomplished, because the more realistic model of the control rod does not function properly. Furthermore, a fundamental problem with the method at hand emerges, when using the Polymax algorithm to identify the fluid force.