A Space-Time Galerkin Boundary Element Method for Aeroacoustic Scattering

This paper presents a robust and unconditionally stable space-time Galerkin time-domain boundary element method for predicting acoustic scattering and shielding by complex geometries, which is validated through analytical cases and successfully applied to a practical trailing edge-mounted propeller scenario with good agreement to experimental data.

The Gist

Here is an explanation of the paper, translated into everyday language with some creative analogies.

The Big Picture: The "Echo Chamber" Problem

Imagine you are shouting in a room full of furniture. The sound doesn't just travel in a straight line from your mouth to the listener's ear. It bounces off the table, wraps around the chair, and creates a complex mix of echoes and shadows.

In the world of aviation, this is a huge problem. When a plane's propeller spins, it creates noise. But that noise doesn't just fly straight out; it hits the wings, the tail, and the fuselage. These surfaces scatter (bounce) the sound and shield (block) it. This changes how loud the plane sounds to people on the ground.

For a long time, predicting these "echoes" has been like trying to solve a puzzle while wearing blindfolds. Engineers had to choose between methods that were fast but inaccurate, or methods that were accurate but took so long to compute they were useless for designing new planes.

The Solution: A New "Time-Traveling" Calculator

The authors of this paper (Maks Groom and Beckett Zhou from Georgia Tech) have built a new mathematical tool called a Space-Time Galerkin Boundary Element Method (TDBEM).

Here is the breakdown of what makes it special, using simple metaphors:

1. The "Snapshot" vs. The "Movie"

• Old Methods (Frequency Domain): Imagine trying to understand a movie by looking at a single, frozen photograph. You can see the actors, but you can't tell how they are moving or if the scene is chaotic. Old methods usually look at sound as a steady hum (a single frequency). This is great for a steady tone, but terrible for the complex, changing roar of a propeller.
• This New Method (Time Domain): This method watches the whole movie. It simulates the sound wave as it moves through time, second by second. This allows it to handle "broadband" noise (a mix of many frequencies at once) and sudden changes (transients) perfectly. It's like watching the sound wave dance in real-time rather than taking a snapshot.

2. The "Unbreakable" Foundation

• The Stability Issue: Many old simulation methods are like a house of cards. If you tweak the numbers just a little bit (like changing the frequency of the sound), the whole calculation collapses and gives you garbage results. Engineers often had to "tune" the math with special knobs and dials to keep the house standing, but they didn't always know what the right settings were.
• The Galerkin Advantage: The authors used a mathematical approach called "Galerkin." Think of this as building a house out of solid concrete blocks instead of cards. It is unconditionally stable. It doesn't matter how complex the sound is or what the shape of the plane is; the math holds up without needing any "tuning knobs." It just works.

3. The "Magic Trick" with Math

• The Hard Part: The biggest problem with this "concrete block" method is that the math requires calculating a massive, double-layered integral (imagine trying to measure the volume of every single drop of water in a swimming pool and the time it takes for each drop to move, all at once). This is usually so computationally expensive that it would take a supercomputer years to finish.
• The Breakthrough: The authors developed a clever "decomposition" strategy. Instead of trying to measure the whole pool at once, they broke the problem down into tiny, manageable geometric shapes (triangles) and solved the math for those shapes exactly. It's like realizing you don't need to count every grain of sand on a beach; you just need to know the shape of the beach and the density of the sand to get the answer instantly. This made the calculation fast enough to be useful.

How They Tested It

To prove their new calculator works, they ran three "test drives":

1. The Smooth Ball (Sphere): They simulated sound bouncing off a perfect sphere. The new method matched the perfect mathematical answer almost exactly.
2. The Sharp Coin (Disk): They simulated sound hitting a flat, thin disk with sharp edges. This is harder because sound behaves strangely at sharp corners. The new method handled this perfectly, even though the disk was so thin it was essentially a 2D sheet in a 3D world.
3. The Windy Wall (Plane with Flow): They simulated sound hitting a flat wall while wind was blowing past it. They used a mathematical "lens" (variable transformation) to bend the sound waves correctly around the wind, and again, the results were spot on.

The Real-World Test: The Propeller on a Wing

Finally, they took the method to the real world. They modeled a propeller mounted on the back edge of a flat plate (simulating a wing).

• The Setup: They used data from a wind tunnel experiment where a real propeller was spinning near a plate.
• The Comparison: They compared their computer simulation against the actual microphone measurements from the experiment.
• The Result: The simulation predicted exactly where the sound would get louder (amplification) and where it would get quieter (shielding) as the propeller moved. The "echoes" and "shadows" matched the real-world data very well.

Why This Matters

This paper presents a tool that is fast, stable, and accurate.

• For Engineers: It means they can design quieter drones, electric planes, and helicopters earlier in the process. They can simulate how a new propeller shape will interact with the wing without building a physical model and testing it in a wind tunnel.
• For the Public: Quieter aircraft mean less noise pollution for communities near airports and a more pleasant flying experience.

In short, the authors built a "sound crystal ball" that lets engineers see exactly how noise will bounce around a plane, helping them design quieter, more efficient aircraft for the future.

Technical Summary

Here is a detailed technical summary of the paper "A space-time Galerkin boundary element method for aeroacoustic scattering" by Groom and Zhou.

1. Problem Statement

Vehicle noise, particularly from aeroacoustic sources like propellers and rotors, is a critical design constraint for modern aircraft. A significant challenge in predicting overall noise levels is the scattering and shielding effects caused by vehicle surfaces (e.g., fuselage, wings, or mounting plates).

• Limitations of Current Methods:
  • Grid-based methods (CFD/LES): Accurate but computationally prohibitive due to the need for volume discretization and specialized schemes to avoid numerical dispersion.
  • Acoustic Analogies: Often fail to capture scattering effects unless coupled with high-fidelity compressible flow solutions, which are expensive.
  • Frequency-Domain BEM: Struggles with broadband, transient, and rotating sources (like propellers) because they require multiple simulations for different frequencies.
  • Collocation-based Time-Domain BEM (TDBEM): While capable of handling broadband signals, they often suffer from numerical instabilities near critical frequencies (interior resonances) and require tuned coupling parameters (e.g., Burton-Miller formulation) to stabilize.
  • Equivalent Source Methods (ESM): Require tuning of source positions and struggle with thin geometries.

The authors aim to develop a robust, unconditionally stable, and computationally efficient time-domain boundary element method capable of handling complex, rotating, broadband aeroacoustic sources and thin scattering geometries without tuned numerical parameters.

2. Methodology

The paper proposes a Space-Time Galerkin Time-Domain Boundary Element Method (TDBEM) based on the mathematical formulation of Ha-Duong for the wave equation.

Governing Equations & Transformations

• The method solves the linearized potential wave equation in a steady, irrotational background flow.
• To handle background flows (e.g., uniform flow), the authors employ variable transformations (Prandtl-Glauert-Lorentz for uniform flow, Taylor's transformation for low Mach non-uniform flow). These transform the convected wave equation into the standard wave equation in transformed coordinates, allowing a single algorithmic approach for various flow conditions.
• The problem is formulated as a Neumann problem (sound-hard boundary condition) where the scattered field is computed to cancel the normal velocity of the incident field on the scattering surface.

Discretization Strategy

• Space-Time Galerkin Formulation: Unlike point collocation methods, this approach uses a weak formulation with test and trial functions.
  • Basis Functions: Piecewise linear (P1) basis functions are used for both spatial and temporal discretization.
  • Stability: The Galerkin formulation is unconditionally stable and does not require the introduction of artificial coupling parameters (unlike Burton-Miller collocation).
  • Thin Geometries: The method naturally handles thin surfaces (like plates) that do not enclose a volume by setting the single-layer potential to zero, avoiding the need for double-layer meshes.

Numerical Implementation Challenges & Solutions

The primary computational bottleneck of Galerkin TDBEM is the evaluation of double space-time integrals, which contain hypersingularities and "light cone" singularities (due to discontinuities in the temporal basis functions).

• Decomposition-Based Quadrature: The authors developed an efficient integration strategy to resolve these singularities:
  1. Geometric Decomposition: The integration domain (pairs of triangles) is decomposed into simpler subdomains (sectors and triangles) relative to the light cone boundaries ($|x-y| = n\Delta t$).
  2. Coordinate Transformation: Local cylindrical coordinates are used to analytically cancel the $O(|x-y|^{-1})$ singularity of the Green's function.
  3. Analytical Integration: The radial integration is performed exactly using derived analytical expressions for specific polynomial terms.
  4. Dimensional Reduction: This reduces the inner spatial integral from a 2D numerical quadrature to a 1D numerical integral (Gauss-Legendre), significantly lowering computational cost.
• Marching-on-in-Time (MOT): The system is solved using a time-marching algorithm where the influence matrix is precomputed, and the solution advances step-by-step, requiring only the solution of a relatively small linear system at each step.

3. Key Contributions

1. Unconditionally Stable Formulation: The space-time Galerkin approach eliminates the need for tuned coupling parameters, making it a robust predictive tool for unknown geometries and source characteristics.
2. Efficient Quadrature Scheme: The novel decomposition-based integration strategy overcomes the high computational cost and singularity issues traditionally associated with Galerkin TDBEM, enabling practical application.
3. Thin Surface Capability: The method efficiently models thin scattering surfaces (e.g., plates, wings) using a single layer of elements, drastically reducing problem size compared to volume-based or double-layer methods.
4. CFD Coupling: The framework successfully couples with CFD-derived incident fields (via Ffowcs Williams-Hawkings analogy) to simulate realistic installed aeroacoustic configurations.

4. Results and Validation

The method was validated against analytical solutions for three distinct test cases and applied to a practical engineering problem:

• Case 1: Sphere (Smooth, Enclosed Body):

  • Validated against harmonic and broadband scattering.
  • Results showed excellent agreement with analytical shielding factors across frequencies ($ka = 2$ to $16$).
  • The method remained stable at critical frequencies ($ka = n\pi$) where collocation methods typically fail.
  • Convergence: Refinement studies demonstrated second-order convergence in both space and time. A Courant-Friedrichs-Lewy (CFL) number of $\approx 0.5$ was identified as optimal for minimizing total error.

• Case 2: Disk (Flat, Sharp Edges):

  • Validated scattering from a thin surface with sharp edges.
  • Demonstrated the ability to model thin geometries with a single element layer.
  • Showed strong agreement with analytical solutions for harmonic forcing.

• Case 3: Plane in Uniform Flow (Transient Source):

  • Validated scattering of a transient Gaussian source in a uniform flow using PGL transformation.
  • Results matched analytical solutions (method of images) perfectly, capturing flow-induced amplification effects.

• Application: Trailing Edge-Mounted Propeller:

  • Applied to a propeller mounted near the trailing edge of a flat plate (replicating Hanson et al. experiments).
  • Incident Field: Derived from RANS CFD simulations using the FWH analogy.
  • Findings: The TDBEM successfully predicted the Sound Pressure Level (SPL) changes due to scattering and shielding.
  • Agreement: The numerical predictions matched experimental measurements well, capturing key trends such as amplification on the reflection side, shielding on the shadow side, and the movement of crossover angles as the propeller position ($H/D$, $L/D$) changed.
  • Discrepancies were generally within 5 dB, attributed to experimental setup details (e.g., wind tunnel nozzles) not included in the numerical model.

5. Significance

This work presents a significant advancement in computational aeroacoustics by providing a predictive, stable, and efficient tool for analyzing noise scattering in complex aircraft configurations.

• Design Impact: It enables early-stage design assessment of vehicle noise by accurately predicting how airframe components scatter and shield noise from rotating sources (propellers, rotors), which is crucial for meeting noise regulations and improving public acceptability.
• Methodological Leap: By resolving the computational difficulties of space-time Galerkin methods, the authors make a theoretically superior (stable, parameter-free) approach practically viable for industrial applications.
• Broad Applicability: The method is versatile, handling harmonic, broadband, and transient sources, as well as uniform and non-uniform background flows, making it suitable for a wide range of emerging propulsion technologies (e.g., distributed electric propulsion, open rotors).