- Restrictions on Access:
- Open Access.
- The tiltrotor configuration offers the ability to combine vertical flight with high-speed, efficient cruise performance. With proprotors close to the wing tip, the reconfigurable vehicle can transition from a hover mode to an airplane mode. At high speeds in the airplane configuration, tiltrotors are known to experience an aeroelastic instability known as whirl-flutter which is a dynamic coupling between the wing and rotor. Unfortunately, even the most recent literature shows an inability to consistently predict the onset of whirl flutter when compared to test data. The research presented in this work seeks to improve whirl-flutter prediction capabilities in two ways. First, modeling of induced velocity in current state-of-the-art methods is largely neglected. Here, both a dynamic inflow model and a mid-fidelity vortex particle method (VPM) are used to make tiltrotor aeroelastic stability predictions. Second, the most common method of performing aeroelastic stability analysis is through linearizing the dynamic equations of motion and using eigenanalysis to extract modal frequency and damping values. A transient approach is developed and used in this research which enables the ability to perform stability analysis using the vortex particle method. Together, these two additions to tiltrotor whirl-flutter modeling are implemented on two semi-span configurations and comparisons are made with current standard practice methods to identify potential benefits. First, stability analysis of a 4-bladed soft-in-plane tiltrotor model is performed using transient simulations with three wake models: uniform inflow, dynamic inflow, and VPM. In these results, a clear trend of increasing wake modeling fidelity leading to an increased prediction in damping is established. A maximum variation in predicted damping of the wing beam-bending mode of 0.75\% critical from uniform inflow to VPM is shown. Additionally in one configuration, a variation in predicted whirl-flutter speed ranges from 121 knots using uniform inflow to 149 knots using VPM. Next, these results are built upon by adding transient frequency and damping predictions of the wing chord-bending and torsion modes to the wing beam-bending predictions for the XV-15. Additionally, linearized predictions are made using uniform inflow and dynamic inflow. Particle resolution is shown to be a critical variable for consideration when using VPM analysis, particularly at high-speed flight conditions. A detailed resolution study at 370 knots shows that a coarse particle resolution predicts a damping value of 0.3\% critical for the wing beam bending mode, while a fine particle resolution predicts 1.5\% critical. At low speeds, excellent agreement between the wake models and analysis methods is shown. For speeds greater than 250 knots, uniform inflow predicts a lower damping than either dynamic inflow or VPM, with VPM predicting the highest damping for the wing beam-bending and chord-bending modes. For the baseline XV-15, a variation in damping of 1.2\% critical is seen between uniform inflow and VPM transient predictions at 350 knots. A reduced stiffness variant of the XV-15 is used to destabilize the model. Here, the wing chord-bending mode is predicted to become unstable at 300 knots when using transient uniform-inflow analysis and 325 knots when using transient VPM analysis. In addition to variation between wake models, comparisons are made between linearized and transient predictions. Again, low-speed predictions are consistent between the two methods. A curve-veering interaction between the rotor lag mode and the wing beam-bending mode for the XV-15 is identified as contributing to large differences in damping between the two methods. The linearized model predicts the curve-veering interaction between the modes as the two frequencies approach each other, while the transient prediction of the wing beam-bending mode frequency smoothly passes through this range of airspeeds. In turn, the predicted damping trends and amplitudes are significantly different between the two methods.
- Dissertation Note:
- Ph.D. Pennsylvania State University 2020.
- Technical Details:
- The full text of the dissertation is available as an Adobe Acrobat .pdf file ; Adobe Acrobat Reader required to view the file.
View MARC record | catkey: 30586022