High Order Wavelet-Based Multiresolution Technology for Airframe Noise Prediction

An integrated framework is proposed for efficient prediction of rotorcraft and airframe noise. A novel wavelet-based multiresolution technique and high-order accurate WENO scheme is proposed for efficient capturing of noise sources and unsteady flowfield. A wavelet compression is used to store the flowfield as a multi-level representation in functional space. The primary solution progresses using a coarse grid. The regularity of the flow field data is used to identify regions of steep variation. These regions are selectively solved recursively in the finer grid-levels and accurate information is injected into the coarse grids to correctly represent all flow features. In Phase I, a three-dimensional wavelet-based multiresolution algorithm, and an acoustic analogy module based on the Kirchhoff-Ffowcs Williams and Hawking methodology will be developed. The feasibility of the proposed technology will be demonstrated by prediction of three-dimensional noise source and acoustic waves of vortex-blade interaction problems. The proposed technology will provide 2-3 orders-of-magnitude reductions in CPU requirements over existing techniques. In Phase II, the wavelet compression methodology will be integrated into a high-fidelity CFD module. An efficient data structure will be developed to store and update the multiresolution data. The modules will be coupled with a nonlinear finite-element structure dynamic module for noise prediction of flexible structures.