Numerical prediction of cavitation for a horizontal axis tidal turbine

Adriano Evangelisti*, Giuliano Agati, Domenico Borello, Luca Mazzotta, Paolo Capobianchi, Paolo Venturini

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

13 Downloads (Pure)

Abstract

This paper aims at assessing cavitation in a scaled tidal turbine geometry through numeri-cal simulations. Cavitation occurrence is predicted by using the Singhal cavitation model, based on the Rayleigh-Plesset equation, for treating bubble dynamics. Turbulence is mod-elled adopting a Reynolds Averaged Navier Stokes (RANS) approach, specifically employ-ing the Shear Stress Transport (SST) k-ω model to simulate the fluid flow. The Reboud den-sity function is applied to adjust the eddy viscosity computation in the cavitation region. Initially, cavitation and turbulence models are validated using a NACA 66 (mod) hydrofoil profile as a test case. Numerical and experimental pressure coefficients are compared on the hydrofoil suction side for a selected cavitation condition. A Mesh Sensitivity Analy-sis (MSA) is performed to ensure simulation accuracy, comparing numerical results with experimental data on the Horizontal Axis Tidal Turbine (HATT) scaled domain. Based on this analysis, the optimal computational grid is selected. Experimental and numerical power and thrust coefficients are then compared across different tip speed ratios. Finally, cavitation occurrence is evaluated for four different regimes, namely the cut-in, the peak-power, the curve highest velocity and the off-set tip speed ratios. Computational Fluid Dynamics (CFD) solutions reveal vapor formation around turbine components, highlight-ing regions most exposed to cavitation onset.
Original languageEnglish
JournalFlow, Turbulence and Combustion
Early online date2 Dec 2024
DOIs
Publication statusE-pub ahead of print - 2 Dec 2024

Keywords

  • CFD
  • cavitation
  • bubble dynamics
  • MSA
  • HATT

Fingerprint

Dive into the research topics of 'Numerical prediction of cavitation for a horizontal axis tidal turbine'. Together they form a unique fingerprint.

Cite this