Temperature transformation recovering the compressible law of the wall for turbulent channel flow

This paper proposes new Van Driest-type and semi-local-type temperature transformations for compressible turbulent channel flow, derived from momentum and energy balance analyses, which successfully recover the incompressible law of the wall with high accuracy by accounting for mixing length effects, body force work, and turbulent kinetic energy flux.

The Gist

Imagine you are trying to predict how hot a car engine gets or how air flows over a supersonic jet. In the world of fluid dynamics, there's a famous "rule of the road" called the Law of the Wall. It's like a universal traffic sign that tells engineers exactly how fast the air is moving and how hot it is right next to a surface, whether that's a pipe, a wing, or a wall.

For slow-moving air (incompressible flow), this rule works perfectly. But when things get fast—like a jet breaking the sound barrier—the air gets squished, heats up, and behaves chaotically. The old "traffic signs" stop working. The air isn't just moving; it's changing density and temperature in complex ways that confuse the math.

This paper is like a team of mechanics (Xu, Schmidt, and Adams) who have built a new, universal GPS for these high-speed, hot, squishy airflows. They figured out how to translate the chaotic, high-speed "traffic" back into the simple, predictable patterns we already know.

Here is the breakdown of their discovery using simple analogies:

1. The Problem: The "Distorted Map"

Imagine you have a perfect map of a city (the slow, incompressible flow). Now, imagine you take a photo of that city while driving a rollercoaster at 500 mph. The buildings look stretched, the streets warp, and the distances change. This is what happens to air at high speeds.

Scientists have been trying to "un-distort" the speed of the air (velocity) for a long time. They have a good "un-distortion filter" for speed. But they didn't have a good one for temperature. Trying to predict the heat distribution in high-speed flow was like trying to read a thermometer while spinning in a centrifuge—the numbers were all over the place.

2. The Solution: The "Universal Translator"

The authors created two new "filters" (called transformations) to fix the temperature map. Think of these as a special pair of glasses that, when you look through them, make the chaotic, high-speed heat patterns look exactly like the calm, slow-speed patterns we already understand.

They developed two types of glasses:

• The "Van Driest" Glasses (VD-type): These are the classic style. They work okay, but they get a bit blurry near the wall (the surface).
• The "Semi-Local" Glasses (SL-type): These are the new, high-tech upgrade. They account for local changes in the air's density and stickiness (viscosity). The paper shows these are much sharper and clearer, especially right next to the wall where the action happens.

3. The Secret Ingredients: What Makes the Glasses Work?

To make these glasses work, the authors realized they needed to add three specific "lenses" to their formula. If you miss even one, the picture stays blurry.

• Lens 1: The Mixing Length (The "Crowd Size"):
  Imagine the air molecules as a crowd of people. In a calm room, they move in straight lines. In a turbulent room, they bump into each other. The "mixing length" is a measure of how far a person can walk before bumping into someone else. The authors realized that in high-speed flow, this "walking distance" changes shape near the wall. They fixed their formula to account for this changing shape, ensuring the "crowd" behavior is modeled correctly.

• Lens 2: The Body Force (The "Treadmill"):
  In their computer simulations, they had to push the air to keep it moving, like a treadmill pushing a runner. This "push" adds energy to the system. Previous models often ignored this extra energy, leading to errors. The authors added a lens that accounts for the work done by this push, ensuring the energy balance is perfect.

• Lens 3: The Turbulent Energy Flux (The "Hidden Current"):
  This is the most complex part. Imagine a river with a fast current on the surface and a hidden, swirling undercurrent. The "turbulent kinetic energy flux" is that hidden swirl carrying heat. Previous models often ignored this hidden current. The authors found that in high-speed flows, this hidden current is huge. They added a lens to track it, which prevented the "temperature map" from having weird spikes and errors.

4. The Results: A Perfect Fit

When they tested these new glasses against massive supercomputer simulations (which act like digital wind tunnels), the results were amazing.

• The Data Collapsed: Before, the temperature data for different speeds and pressures looked like a messy pile of spaghetti. After applying their new transformation, all those different spaghetti strands lined up perfectly into a single, neat line.
• High Accuracy: For most cases, their prediction was off by less than 2%. That's like predicting the temperature of a cup of coffee and being within a fraction of a degree.
• The "Wake" Problem: They also noticed that in the middle of the channel (far from the wall), the heat didn't just stop; it had a "wake" effect, like the wake behind a boat. Their new model captures this wake perfectly, extending the "Law of the Wall" further out than ever before.

5. Why Does This Matter?

Why should you care if a jet engine gets 2% hotter or cooler in a simulation?

• Better Design: Engineers can now design supersonic jets, rockets, and hypersonic vehicles with much greater confidence. They can predict exactly where the metal will melt and where the fuel will burn efficiently.
• Simpler Math: Instead of running massive, expensive supercomputer simulations for every new design, engineers can use these "transformations" to take simple, slow-flow math and instantly convert it to high-speed predictions. It's like having a cheat code for engineering.
• Safety: By understanding the heat distribution better, we can build safer aircraft that won't overheat or fail during high-speed flight.

The Bottom Line

This paper is a masterclass in translation. The authors took a chaotic, high-speed, high-temperature problem that was hard to solve and found a mathematical "Rosetta Stone" that translates it back into a simple, universal language. They didn't just fix the speed; they fixed the heat, giving engineers a powerful new tool to conquer the skies.

Technical Summary

Here is a detailed technical summary of the paper "Temperature transformation recovering the compressible law of the wall for turbulent channel flow" by Xu, Schmidt, and Adams.

1. Problem Statement

Wall-bounded turbulent flows are critical in aerospace and industrial applications. While the Law of the Wall (LoW) for velocity in compressible flows has been extensively studied and validated through various transformations (e.g., Van Driest, Semi-local), a robust, well-established temperature transformation remains an open issue.

Existing approaches face several challenges:

• Temperature-Velocity (TV) Relations: Many rely on boundary layer edge conditions (centerline values), which are difficult to define in asymmetric flows (e.g., mixed isothermal/adiabatic walls) or internal flows where the centerline is not a priori known.
• Existing Transformations: Previous temperature transformations (e.g., by Huang et al., Chen et al.) often fail to collapse data perfectly in the viscous sublayer and buffer layer, exhibit incorrect logarithmic slopes, or suffer from singularities (spikes) due to energy imbalances when applied to adiabatic walls or high-Mach-number flows.
• Missing Physics: Many prior models neglect the work of body forces, specific mixing length variations in the outer layer, and the turbulent kinetic energy (TKE) flux, leading to inaccuracies in the energy balance.

2. Methodology

The authors propose new Van Driest (VD-type) and Semi-local (SL-type) temperature transformations derived from a rigorous analysis of the momentum and energy balance equations in the overlap layer of compressible turbulent channel flow.

Key Derivation Steps:

1. Governing Equations: The derivation starts from the Reynolds-averaged mass, momentum, and energy equations. The energy equation is decomposed to explicitly account for:
   • Molecular and turbulent heat conduction.
   • Molecular and turbulent diffusion of kinetic energy.
   • Work of the body force (driving force), which acts as an effective pressure gradient.
   • Turbulent TKE flux ($q_t^k = -\rho \overline{v'' k}$), a high-order term often neglected.
2. Overlap Layer Analysis: By analyzing the energy balance in the overlap layer and invoking the mixing length hypothesis ($l_m$) and turbulent Prandtl number ($Pr_t$), the authors derive a generalized differential equation linking the temperature gradient to the heat flux.
3. Introduction of Three Parameters: The transformation kernels are defined by three parameters to correct for compressibility effects:
   • $\psi_1$: Accounts for the mixing length model ($l_m$). The authors test linear, parabolic, and enhanced models.
   • $\psi_2$: Accounts for the work of the body force and the ratio of bulk to local velocity.
   • $\psi_3$: Accounts for the turbulent TKE flux.
4. Transformation Definitions:
   • VD-type: Uses wall-scaling ($y^+$).
   • SL-type: Uses semi-local scaling ($y^*$), incorporating local density and viscosity variations.
   • Both are formulated as integrals of the temperature difference ($\theta^+$) weighted by the derived parameters.
5. Viscous Sublayer Constraint: To ensure the transformation recovers the linear law ($T^+ \approx Pr \cdot y^+$) in the viscous sublayer, the mixing length model is constrained such that $l_m \approx \kappa y$ near the wall, avoiding the need for empirical damping functions that contradict the derived physics.

3. Key Contributions

• Unified Derivation: A complete derivation of temperature transformations that simultaneously accounts for mixing length variations, body force work, and turbulent TKE flux, addressing energy imbalances that cause singularities in previous methods.
• New Transformations: Proposal of specific VD-type and SL-type transformations (Eqs. 31 and 34) that are applicable to both isothermal and adiabatic (or mixed) wall boundary conditions.
• Simplified Models: Identification of the multi-layer structure of the turbulent TKE flux ($q_t^k/q_w$). This allows for the development of simplified transformations (Eqs. 66–68) that approximate the high-order TKE flux using analytical functions ($\beta_1, \beta_2, \beta_3$), removing the need for direct high-order statistics in practical applications.
• Mixing Length Insight: Demonstration that an undamped mixing length model (specifically a parabolic or enhanced form) is required for the transformation to work correctly, contrasting with standard eddy viscosity damping practices in the viscous sublayer.

4. Results

The proposed transformations were evaluated using Direct Numerical Simulation (DNS) and Wall-Resolved Large Eddy Simulation (WRLES) data across a wide range of Mach numbers ($M_b$) and Reynolds numbers ($Re_\tau$), including:

• Classical isothermal walls.
• Mixed isothermal/adiabatic configurations (asymmetric flows).
• Adiabatic walls.

Key Findings:

• Data Collapse: The SL-type transformation significantly outperforms the VD-type, particularly in the viscous sublayer and buffer layer, successfully recovering the compressible Law of the Wall.
• Accuracy:
  • For isothermal walls, the integral mean error over the entire boundary layer is below 2% for most cases (RMS $\approx$ 1.7%).
  • The transformed profiles collapse onto the incompressible reference profile (or extended LoW) with high fidelity.
  • The logarithmic slope is accurately recovered, with the intercept $B_T \approx 3.65$ at $Re^*_\tau \approx 1000$.
• Mixing Length Impact: The parabolic mixing length model ($l_m = \kappa y \sqrt{1-y/h}$) and the enhanced model yield better logarithmic profiles than the linear model, with the enhanced model extending the log-law closer to the channel centerline.
• Adiabatic Walls: While the transformations collapse the data well, the magnitude of the transformed temperature for adiabatic walls remains slightly higher than the incompressible reference, suggesting a need for further refinement in this specific regime.
• Simplified Models: The simplified transformations (using $\beta$ models for TKE flux) perform nearly as well as the complete forms, making them suitable for practical applications where high-order statistics are unavailable.

5. Significance

• Near-Wall Modeling: The transformations provide a rigorous theoretical basis for developing wall models for Large Eddy Simulations (LES) of compressible flows, allowing for the reconstruction of mean velocity and temperature profiles from incompressible counterparts.
• Inverse Transformation: The framework enables the prediction of compressible flow statistics (velocity, temperature, density) given bulk parameters, which is valuable for design and optimization.
• Generalizability: The approach is extendable to pipe flows (demonstrated in Appendix A) and potentially to zero-pressure-gradient boundary layers, provided appropriate models for $\psi_1, \psi_2, \psi_3$ are developed.
• Physical Insight: The study highlights the critical importance of mitigating energy imbalance in temperature transformations and clarifies the role of the turbulent TKE flux, which acts as a damping mechanism in the transformation kernel.

In conclusion, this work establishes a robust, physics-based framework for compressible temperature transformations, resolving long-standing issues with data collapse and singularities, and paving the way for more accurate modeling of high-speed turbulent flows.