The analysis of helicopter dynamics is challenging due to complex mechanical systems, strong coupling with aerodynamics, and parametric uncertainties. High-fidelity finite element models capture full structural and vibration behavior but are computationally too expensive for transient simulations, extensive parameter studies, or iterative design tasks. This thesis presents approaches for model order reduction, coupled simulation frameworks for vibration analysis, and uncertainty quantification including parameter identification and propagation.
The goal of model order reduction is to reduce computational complexity while accurately preserving key properties such as eigenfrequencies and transfer functions, especially near the rotor’s main excitation frequencies. This work combines modal truncation with moment matching based on dominant aerodynamic forces, improving the approximation of critical frequencies compared to conventional modal methods. Two coupling schemes are presented to simulate helicopter vibrations: loose coupling, efficient for harmonic-dominated problems and compatible with standard trimming analysis, and tight coupling, a co-simulation of CFD and structural dynamics capturing unsteady loads at higher computational costs.
Dynamic scatter between different helicopters of the same type is systematically addressed with uncertainty quantification based on possibility theory. Parameter distributions are derived from measurements and their influence on eigenfrequencies quantified. Variations between individual structures of the same type are considered, and possibilistic uncertainty descriptions are used to predict behavior of new configurations.