Krylov Subspace Based Direct Projection Techniques For Low Frequency, Fully Coupled, Structural Acoustic Analysis and Optimization

Noise, Vibration and Harshness (NVH) is a critical consideration in the design of automotive and aerospace vehicles for comfort, and fatigue of components arising from interior structural and acoustic pressure fluctuations due to external structural or acoustic loading. In the low to mid frequency range, current NVH models, obtained by direct, unsymmetric, coupled Eulerian Finite Element discretization (often known as the Cragg’s u/p unsymmetric formulation) cannot provide a highly accurate and computationally efficient lower order model suitable for iterative structural and acoustic design modifications, or for example control applications via state-space techniques. The task of generating an accurate and a computationally efficient lower order model is further complicated due to the addition of non-trivial trim details such as frequency dependent damping resulting in an explicit participation of the damping matrix in the coupled higher dimensional system. In this thesis, lower order, fully coupled, structural-acoustic models are developed using a systematic dimension reduction procedure from the higher dimensional coupled structural-acoustic system to enable efficient, fully-coupled, undamped and damped structural-acoustic analysis and optimization. The proposed dimension reduction techniques do not require the solution of the traditional coupled or uncoupled eigenvalue problems; but instead are based on im-