Finger seals operate over extended periods under complex conditions involving high-pressure differentials, elevated rotational speeds, and rotor radial runout. Intense convective heat transfer arises within the seal, significantly impacting its structural deformation. To elucidate the influence of temperature on finger-seal deformation during convective heat transfer, the present study derives heat transfer energy equations for finger seals based on the Local Thermal Non-Equilibrium (LTNE) model. A three-dimensional porous-media flow-field model incorporating the LTNE framework, along with a solid thermal-deformation model, is developed. The effects of pressure differential and interference-fit magnitude on the structural deformation and average contact pressure of finger seals are analyzed under both the Local Thermal Equilibrium (LTE) and LTNE models. The results indicate that the LTNE model predicts a higher maximum seal temperature and a lower leakage rate compared to the LTE model. In both models, the deformation of individual seal-blade layers increases with rising pressure differentials and interference-fit magnitudes. Furthermore, the overall blade deformation is more pronounced under the LTNE model, suggesting a substantial thermal influence on sealing performance. The effects of pressure difference and interference fit on the thermal deformation of the seal plate are similar: both have the greatest impact on radial deformation, followed by circumferential deformation and axial deformation. Within the pressure difference range, the radial deformation of the third-layer seal plate in the LTNE model increases by 14.55%. When the interference fit increases from 0.05 mm to 0.2 mm, the radial deformation of each layer of the seal plate in the LTNE model increases by 0.18 mm. The average contact pressure increases with both pressure differential and interference-fit magnitude across both models. At a given pressure differential, the LTNE model yields a higher average contact pressure than the LTE model, with a maximum observed difference of 0.01 MPa. When the interference-fit magnitude is small, the pressure difference between the models remains minimal; however, at the maximum interference-fit, the difference reaches 0.08 MPa.
Wang et al. (Thu,) studied this question.