Influence of orthotropic stiffness and damping in hydropower rotors

In this thesis the dynamic behaviour of the rotor and the bearings (tilting pad bearings) of a hydro power station is studied. More specifically, the influence of anisotropic bearing stiffness is studied, when the bearing stiffness is different in X and in Y direction. The influence of damping on rotor vibration and bearing force is also studied. In the thesis three different models are studied. The first and the second are Jeffcott’s rotor model (point mass attached at the middle med of mass less shaft). In the first rigid supports are assumed and in the second elastic supports are used. They are both quite simple model. The last model is a more realistic rotor including damping of the bearings and air drag. From the results one can observe the influence of anisotropy which results in different behaviour in x and y direction for the force and for the displacement. The anisotropic case shows that the best working point for low displacement in the bearings is not the best for the force in the bearings. The effect of damping in the bearing is also studied by its influence on rotor vibration and bearing force. The simulations show that there exist an optimal damping value between the lowest and the highest value. The high bearing damping value is in many cases the worst choice resulting in high rotor amplitude and high force in the bearing. Another interesting observation is the effect of the relation of bearing and shaft stiffness to the wpeak value (Frequency with highest amplitude value) for different values of damping. When the damping value is increased there will be a significant shift in the frequency of wpeak. The reason is that the eigenfrequency will suddenly change when the bearing damping is changed. The results can assist designers of real hydropower rotors to design suitable bearings for each application and give some theoretical models for determining the optimum choice of damping. An interesting continuation of the project would therefore be to study the effects on a more realistic model which include the whole system. Especially on a rotor with dynamic problems, it would be interesting to compare the results with measurements and hopefully indicate how the system could be improved.Validerat; 20101217 (root