COUPLED 2025

Vibration Prediction of a Rotating Structure Submerged in a Dense Fluid by Fluid-Structure Interaction

  • Ceusters, Kamiel (Belgian Nuclear Research Centre)
  • Iarmonov, Mikhail (Belgian Nuclear Research Centre)
  • Makhov, Kirill (Belgian Nuclear Research Centre)
  • Pacio, Julio (Belgian Nuclear Research Centre)
  • Verstraete, Tom (Belgian Nuclear Research Centre)
  • Degroote, Joris (Belgian Nuclear Research Centre)

Please login to view abstract download link

In this paper a methodology is developed to characterize the vibration behaviour of rotating geometries submerged in a dense fluid. While the proposed methodology has various applications, the focus in this case is on the main coolant pumps of nuclear reactors. In this field, research is being conducted at the Belgian Nuclear Research Centre into pumps for liquid metal cooled reactors. A simulation technique is required to predict sources of resonance in rotating machinery. Therefore, multiple simulations are performed as part of the methodology to obtain the vibration modes of interest. Firstly, a structural solver is used to determine the Eigenmodes in vacuum. Additional factors can be introduced such as centrifugal stiffening or Rayleigh damping. Secondly, the flow field is initialized with a specified rotational velocity in a fluid solver, taking into account all the relevant phenomena, e.g. vortex shedding, flow transition and boundary layer effects. Finally, both solvers are combined in a fully coupled fluid-structure interaction simulation (FSI). The single-physics solvers are coupled through a shared interface and convergence of the coupling iterations is accelerated by using a quasi-Newton algorithm, developed at Ghent University. The structure is subjected to a forced excitation, after which the vibration parameters are extracted from its free response. The methodology is applied to an existing vibration experiment, performed by Chen et al. In this experiment, the frequency and damping of a propeller rotating and vibrating in water are measured for different rotational speeds. The aforementioned methodology was applied to this case to validate the computational approach. The FSI simulations show strong agreement with the experimental results, accurately predicting the frequency and damping.