In the evolving landscape of nuclear energy, ensuring the safety and efficiency of nuclear reactors remains paramount, particularly with the increasing demands for energy and a concurrent rise in global temperatures. A significant aspect of nuclear safety involves maintaining the integrity of the fuel rods, which are susceptible to Turbulence Induced Vibrations (TIV) resulting from axial flows of the coolant liquid. TIV can instigate severe repercussions including structural damage such as fatigue, wear, and stress corrosion cracking, posing substantial threats to reactor safety. Despite the historical attention this phenomenon has garnered since the 1950s, conventional semi-empirical methods offer limited predictive accuracy and do not facilitate extrapolations for multi-rod scenarios effectively.  Recent developments have turned to Fluid-Structure Interaction (FSI) simulations as a powerful tool to study fuel rods’ behavior under TIV effects, capitalizing on the increase of computational power available today. While Direct Numerical Simulation (DNS) and Large Eddy Simulation (LES) offer more accurate predictions, their computational demands make them unsuitable for complex FSI simulations. This has led to a preference for Unsteady Reynolds-Averaged Navier-Stokes (URANS) simulations, despite them underpredicting the displacement amplitudes of the vibrations.  Evolving from this shortfall, the current study directs its focus on a recently developed Anisotropic Pressure Fluctuation Model (AniPFM). This model generates a synthetic velocity fluctuations field, which is used to solve for the pressure fluctuations. The use of this model together with URANS poses as a possible way to inexpensively simulate the excitation mechanisms of TIV of fuel rods. While previous research has highlighted the potential of this model, it is important to note the considerable level of uncertainty still associated with it. Additionally, there are parameters, definitions and constants whose impacts on the model are not yet fully understood or even explored. This calls for a comprehensive research to fine-tune the model, optimize its performance and further validate it. This is precisely the goal of this study, carried out through the analysis of two pure flow and two FSI cases. Hypotheses were formulated and tested in pure flow scenarios before being further validated in FSI cases. Key advancements were made by optimizing the time correlation method used on the generated velocity fluctuations, which significantly reduced the model’s uncertainty. This method was then calibrated using DNS data of turbulent channel flow. Further calibration was undertaken, this time in the parameters part of the modelling of the turbulent kinetic energy spectrum, to address the overprediction of pressure fluctuations near the wall observed in the baseline model. Moreover, the turbulent annular flow was used as the second flow only case, providing more complexity compared to channel flow, by adding curvature, as well as another opportunity to test the hypothesis made.  Furthermore, the hypotheses underwent additional validation via FSI simulations, through a brass beam in turbulent axial flow, showing a substantial decrease in the average difference from the experimental data to 19% from a previous 68%, over the range of inflow velocities considered. Notably, the calibrated AniPFM surpassed LES in accuracy while requiring fewer computational resources. The results obtained are promising, but further validation is needed. This thesis also outlines and lays the foundation for further validation work, through the setup and initial simulations of a flexible cantilever rod in turbulent axial water flow.