Numerical Investigation of Discontinuous Ice Effects on Swept Wings

This study utilizes enhanced delayed detached-eddy simulations to demonstrate that discontinuous ice on swept wings causes more severe lift reduction than continuous ice by disrupting leading-edge vortex formation through gap jets, while simultaneously preventing sudden stall and introducing distinct flow oscillation patterns characterized by specific Strouhal numbers.

The Gist

Imagine an airplane wing as a sleek, smooth surfer riding a wave of air. Its job is to stay attached to that wave to keep the plane flying high and efficient. Now, imagine that surfer suddenly gets covered in sticky, jagged ice. This is the problem of ice accretion.

This paper investigates a specific, tricky kind of ice: discontinuous ice. Instead of a solid, smooth sheet of ice covering the front of the wing (like a continuous blanket), this ice forms in chunks with gaps in between, looking more like a row of jagged rocks or a broken zipper.

Here is the story of what happens when these two types of ice hit the wing, explained simply:

1. The Two Types of Ice: The Blanket vs. The Broken Fence

The researchers compared three scenarios:

• The Clean Wing: A smooth surfer.
• Continuous Ice: A solid, smooth block of ice glued to the front. It's ugly, but it acts like a large, smooth blanket. When air hits it, it peels off in one big, predictable sheet, creating a large "bubble" of swirling air behind it. Surprisingly, this big bubble actually helps the wing hold onto some lift (the force that keeps the plane up) for a little while.
• Discontinuous Ice: This is the real villain. It's like a broken fence with gaps between the posts. When air hits this, it doesn't just peel off; it gets shot through the gaps like tiny, high-speed water jets from a hose.

2. The Big Surprise: Why Broken Ice is Worse for Lift

You might think a solid block of ice would be worse than a broken one. But the study found the opposite for lift (keeping the plane up).

• The Continuous Ice (The Blanket): Because the air peels off smoothly, it creates a large, calm swirling bubble. This bubble acts like a cushion, helping the wing stay airborne a bit longer.
• The Discontinuous Ice (The Broken Fence): The "gap jets" (the air shooting through the cracks) act like tiny disruptors. They smash the air's ability to form that helpful cushion. Instead of a smooth flow, the air gets chaotic and messy immediately.
  • Result: The wing loses lift much faster with broken ice than with solid ice. It's like trying to surf on a wave that's been chopped up by a motorboat; you fall off much sooner.

3. The Drag Trade-off: A Silver Lining?

While broken ice is terrible for lift, it's actually slightly better for drag (the air resistance that slows the plane down).

• The solid ice creates a massive, turbulent wake that drags the plane back heavily.
• The broken ice, while chaotic, creates a smaller overall drag penalty. It's a "less bad" scenario for speed, but a "much worse" scenario for staying in the air.

4. The Dance of the Air: Vortices and Frequencies

The researchers looked at how the air moves using super-computers, visualizing it like a dance.

• The Continuous Ice: The air dances in a predictable, rhythmic way, like a slow, heavy waltz. It forms a big separation bubble that stays relatively stable.
• The Discontinuous Ice: The air dances to a chaotic, fast-paced jazz tune. The "gap jets" break up the smooth flow, creating tiny, swirling tornadoes (vortices) that spin off the ice chunks.
  • The Analogy: Imagine throwing a smooth stone into a pond (continuous ice) vs. throwing a handful of jagged rocks (discontinuous ice). The smooth stone makes one big, clean ripple. The jagged rocks make a million tiny, chaotic splashes that crash into each other.

5. The "Heartbeat" of the Ice

The study found that the broken ice has a specific "heartbeat" or rhythm.

• The air swirling off the ice chunks creates a pattern that repeats at specific speeds.
• They found three main "beats" (frequencies). The most important one is a fast beat that happens twice as fast as the main swirling motion.
• Why it matters: This fast beat is caused entirely by the air shooting through the gaps. It's a unique signature of broken ice that doesn't exist with solid ice. This helps engineers know exactly what kind of ice they are dealing with just by listening to the "noise" of the airflow.

The Bottom Line

This research tells us that not all ice is created equal.

• Solid ice is bad, but it creates a predictable, large bubble that the wing can somewhat tolerate.
• Broken (discontinuous) ice is a silent killer. It destroys the wing's ability to generate lift much faster because the gaps in the ice shoot jets of air that scramble the flow.

Why does this matter?
Engineers usually design planes assuming ice will form in smooth, predictable layers. This study warns them: "Watch out for the jagged, broken ice!" It suggests that future safety systems need to account for these chaotic "gap jets" to prevent planes from stalling unexpectedly in icy conditions.

In short: A smooth sheet of ice is a heavy blanket; a broken sheet of ice is a chaotic storm. And for an airplane wing, the storm is much more dangerous.

Technical Summary

1. Problem Statement

Ice accretion on aircraft wings significantly degrades aerodynamic performance by reducing lift, increasing drag, and diminishing control effectiveness. While extensive research exists on simplified or span-averaged ice geometries (such as continuous horn or ridge ice), realistic ice accretion on swept wings often forms complex, discontinuous geometries (e.g., scallop ice) with gaps between ice segments.

• The Gap: Existing numerical studies often rely on simplified continuous ice models or experimental approaches due to the difficulty of meshing complex, high-fidelity discontinuous geometries.
• The Challenge: The flow physics over discontinuous ice are poorly understood. Specifically, it is unclear how the "gap jets" (flow passing through the spaces between ice segments) interact with leading-edge vortices, separation bubbles, and unsteady flow structures compared to continuous ice.
• Objective: This study aims to investigate the aerodynamic performance, flow structures, and characteristic frequencies of infinite-span swept wings with artificially simulated discontinuous ice, comparing them against clean and continuous-ice configurations.

2. Methodology

The study employs a high-fidelity numerical approach to resolve complex unsteady flow phenomena.

• Numerical Solver: The simulations utilize CFL3D, a structured finite-volume solver for the Navier–Stokes equations.
• Turbulence Modeling: The primary method is Enhanced Delayed Detached-Eddy Simulation (IDDES) based on the Anisotropic Minimum-Dissipation (AMD) model.
  • Why AMD-IDDES? Standard IDDES can suffer from the "gray area" problem (premature transition from RANS to LES) and sensitivity to highly anisotropic grids. The AMD formulation dynamically adjusts the subgrid length scale based on local flow features, making it more robust for the complex, anisotropic grids required around ice geometries.
• Geometry:
  • Airfoil: NACA 23012 with a 30° sweep angle.
  • Configurations: Three cases were simulated: (1) Clean wing, (2) Continuous ice, and (3) Discontinuous ice (simulated using an artificial ice shape with a duty cycle of 52.8%, consisting of ice segments and gaps).
  • Domain: Infinite-span configuration using periodic boundary conditions in the spanwise direction to eliminate tip/root effects.
• Validation: The AMD-IDDES method was first validated against experimental data for a NACA 0012 wing with glaze ice, showing good agreement in lift coefficients and pressure distributions.
• Analysis Tools:
  • Power Spectral Density (PSD) for frequency analysis.
  • Triple Decomposition of the velocity gradient tensor to distinguish between normal straining, pure shearing, and rigid-body rotation.
  • Proper Orthogonal Decomposition (POD) to extract dominant flow modes and energy distribution.

3. Key Contributions

1. High-Fidelity Simulation of Discontinuous Ice: This study provides one of the first detailed numerical investigations of the flow field around realistic, discontinuous ice geometries on swept wings using AMD-IDDES.
2. Mechanism of Lift Loss: It identifies that while continuous ice creates a large separation bubble that partially sustains lift, discontinuous ice disrupts this mechanism via gap jets, leading to a more severe lift reduction.
3. Flow Physics Characterization: The study characterizes the flow over discontinuous ice as a superposition of two canonical patterns: a separating shear layer and Kármán vortex shedding, but notes that the shear layer is highly irregular due to gap jet interference.
4. Frequency Signatures: It identifies specific Strouhal numbers associated with discontinuous ice that are absent in continuous ice cases, linking them directly to gap jet dynamics.

4. Key Results

A. Aerodynamic Performance

• Stall Behavior: Discontinuous ice causes the earliest stall (at $\alpha = 0^\circ$), followed by continuous ice ($\alpha = 2^\circ$), and the clean wing ($\alpha = 6^\circ$).
• Lift vs. Drag:
  • Lift: Discontinuous ice causes a more severe reduction in lift than continuous ice. Continuous ice forms a large separation bubble with a low-pressure region that helps maintain some lift; discontinuous ice destroys this bubble via gap jets.
  • Drag: Interestingly, while lift loss is worse for discontinuous ice, the drag penalty is less severe than that of continuous ice.
• Stall Drop: Unlike the continuous-ice wing, which exhibits a sudden lift drop at stall, the discontinuous-ice wing shows a gradual decline because the flow is already turbulent and separated at low angles of attack.

B. Flow Structures

• Continuous Ice: Exhibits a classic large separation bubble downstream of the ice tip, characterized by Kelvin-Helmholtz (K-H) instability and the formation of organized hairpin vortices.
• Discontinuous Ice:
  • Gap Jets: Flow passing through the gaps creates jets that interfere with the leading-edge flow, preventing the formation of a large, coherent separation bubble.
  • Vortex Dynamics: The flow is dominated by counter-rotating vortex pairs shed from the ice segments. These vortices convect downstream, break up rapidly, and induce strong turbulence.
  • Shear Layer: The separating shear layer is highly irregular and distorted compared to the smooth, coherent layer seen in continuous ice.

C. Frequency Analysis

• Characteristic Frequencies: Three chord-based Strouhal numbers ($St$) were identified for discontinuous ice: 11.3, 22.6, and 33.9.
  • $St = 11.3$: Corresponds to the shedding of counter-rotating vortex pairs. When normalized by ice width, $St \approx 0.58$, which is higher than the canonical cylinder wake ($St \approx 0.176$).
  • $St = 22.6$: This is the dominant frequency for lift and drag fluctuations (twice the shedding frequency). This phenomenon is unique to discontinuous ice and is induced by the gap jets.
  • $St = 33.9$: A higher harmonic associated with smaller-scale vortex breakup.
• Continuous Ice: Shows no dominant characteristic frequency in lift/drag fluctuations, indicating a more chaotic, broadband unsteadiness without a single periodic driver.

D. POD Analysis

• The flow energy in the discontinuous-ice case is more dispersed across many modes compared to continuous ice (where energy is concentrated in the first few modes). This confirms the flow field is more complex, with weaker large-scale coherent structures and stronger interactions between multiple flow patterns.

5. Significance

• Safety and Design: The findings highlight that discontinuous ice (common in real-world scallop formation) is more dangerous for lift generation than previously assumed based on continuous ice models. This necessitates more conservative safety margins or advanced de-icing strategies.
• Modeling Advancement: The successful application of AMD-IDDES demonstrates its capability to handle complex, anisotropic grids and anisotropic flows, providing a reliable tool for future high-fidelity icing simulations.
• Control Strategies: The identification of specific frequencies ($St=22.6$) driven by gap jets offers potential targets for active flow control or sensor-based detection systems to mitigate ice-induced aerodynamic degradation.
• Future Work: The study establishes a baseline for infinite-span wings; future research should extend these findings to finite wings to account for 3D tip effects and root interactions.