Evaluation of Statistical Consistency in Synthetic Turbulence under Wavenumber Bounds

Basic Information

• Paper ID: 2510.13089
• Title: Evaluation of Statistical Consistency in Synthetic Turbulence under Wavenumber Bounds
• Authors: Hongyuan Lin, Yi Liu, Shizhao Wang, Chun-Hian Lee
• Classification: physics.flu-dyn
• Publication Date: October 15, 2025
• Paper Link: https://arxiv.org/abs/2510.13089

Abstract

The Random Fourier Method (RFM) is widely employed for synthetic turbulence generation due to its mathematical clarity and simplicity. However, deviations persist between preset inputs and synthesized results, and the sources of these errors have not been adequately assessed. This study aims to systematically evaluate the effects of spectral coefficient calibration, grid constraints, and wavenumber bounds on the accuracy of turbulence generated by RFM. Results demonstrate that recalibration of spectral coefficients is crucial for ensuring turbulent kinetic energy (TKE) consistency. The upper wavenumber bound, determined by grid resolution, controls the overall TKE level, while the lower bound, set by computational domain size, governs the fidelity of the energy spectrum in the low-wavenumber range. Furthermore, extending the wavenumber range beyond grid-constrained boundaries may improve TKE accuracy but simultaneously amplifies spectral deviations.

Research Background and Motivation

Problem Definition

Synthetic turbulence is of paramount importance in computational fluid dynamics (CFD) and computational aeroacoustics (CAA), with applications including broadband noise research, turbulent inflow generation for high-fidelity simulations, and particle diffusion studies. Insufficient accuracy in synthetic turbulence may introduce artificial noise or distort the physical mechanisms under investigation.

Limitations of Existing Methods

Although the Random Fourier Method (RFM) is widely adopted due to its conceptual simplicity and convenience, it exhibits two primary deficiencies in statistical consistency:

1. Energy-level errors: Manifested as deviations in total turbulent kinetic energy (TKE)
2. Spectral-level errors: Manifested as distortions in the turbulent energy spectrum E(κ) within the discretized wavenumber range

Research Motivation

While previous studies have addressed coefficient determination or proposed empirical wavenumber criteria, few have focused on error sources, which may lead to unexpected deviations in practical applications. This study systematically evaluates the effects of spectral coefficient calibration, grid constraints, and wavenumber bounds on the statistical consistency of turbulence generated by RFM.

Core Contributions

1. Systematic Error Analysis: First systematic evaluation of the effects of spectral coefficient calibration, grid constraints, and wavenumber bounds on synthetic turbulence accuracy
2. Error Source Identification: Clarification that TKE errors primarily originate from high-wavenumber truncation, while E(κ) errors stem from insufficient low-wavenumber coverage
3. Theoretical Framework Enhancement: Establishment of a theoretical framework demonstrating that wavenumber bound selection must account for grid constraints
4. Practical Guidance Principles: Provision of practical guidelines for parameter selection in CFD and CAA applications

Methodology Details

Task Definition

Construction of fluctuating velocity fields satisfying preset statistical constraints, enabling high-fidelity reproduction of target energy spectra and turbulent kinetic energy.

Model Architecture

Random Fourier Method Foundation

The fluctuating velocity component u' at arbitrary spatial location x can be expressed as:

u'(x) = 2∑(m=1 to M) u_m cos(k_m·x + ψ_m)σ_m

Where:

• M: Total number of modes
• u_m: Amplitude of the m-th mode
• k_m: Wave vector of the m-th mode
• ψ_m: Random phase angle
• σ_m: Unit directional vector

von Kármán-Pao Spectrum

The widely-used von Kármán-Pao spectral form is adopted:

E(κ) = αu'²/κ_e · (κ/κ_e)⁴/[1+(κ/κ_e)²]^(17/6) · exp[-2(κ/κ_η)²]

Where α is the spectral coefficient, κ_e is the energy-containing wavenumber, and κ_η is the Kolmogorov wavenumber.

Spectral Coefficient Calibration

To account for finite Reynolds number effects, the local Reynolds number is introduced:

Re_L = TKE²/(εν)

The calibrated spectral coefficient α and energy-containing wavenumber κ_e are determined by solving equations involving confluent hypergeometric functions.

Technical Innovations

1. Systematic Analysis Framework: Establishment of a comprehensive analytical framework considering both spectral coefficient calibration and wavenumber bounds
2. Grid Constraint Theory: Clarification that grid-constrained wavenumber bounds are κ_min = 2π/L and κ_max = π/Δx
3. Error Separation Mechanism: Separation of TKE and spectral errors with individual analysis of their contributing factors

Experimental Setup

Benchmark Parameters

• Velocity fluctuation: u' = 0.25 m/s
• Dissipation rate: ε = 0.5402 m²/s³
• Preset turbulent kinetic energy: TKE_pre = 0.09375 m²/s²

Grid Configuration

• Baseline grid: 128³ points
• Computational domain length: 0.18π m
• Number of modes: 10³ (linearly distributed)

Evaluation Metrics

TKE Relative Error

Err_TKE = |TKE_syn - TKE_pre|/TKE_pre

Spectral Relative Error

Err_E(κ_j) = |E_syn(κ_j) - E_pre(κ_j)|/E_pre(κ_j)

Experimental Results

Spectral Coefficient Calibration Effects

Calibrated spectral coefficients reduce TKE error from 17.28% to 10.14%, validating the importance of coefficient calibration.

Wavenumber Bound Extension Effects

When the upper bound is extended to 10κ_max:

• TKE error significantly reduces to 1.239%
• Spectral error dramatically increases to 28.50% (mean) and 41.15% (RMS)
• Demonstrates that exceeding grid-constrained boundaries induces aliasing effects

Upper Wavenumber Bound Impact

As grid resolution increases from 32³ to 512³:

• Err_TKE decreases from 32.8% to 0.64%
• Spectral error remains within 5-8% range
• Confirms that the upper bound primarily controls TKE consistency

Lower Wavenumber Bound Impact

Effects of increasing computational domain (lowering κ_min):

• Err_TKE decreases from 36.12% to approximately 8%
• Spectral error decreases from 17% to below 5%
• Confirms that the lower bound primarily controls low-wavenumber spectral fidelity

Related Work

Historical Development

• Kraichnan (1970): First proposal of the Random Fourier Method
• Karweit et al. (1991): Wavenumber range selection based on experimental data
• Béchara et al. (1994): Introduction of integral length scale and Kolmogorov wavenumber
• Bailly & Juvé (1999): Connection between computational domain and grid resolution
• Saad et al. (2017): Association of wavenumber range with sampling theory

Comparison with Existing Work

In contrast to existing research, this study provides the first systematic analysis of error sources rather than merely empirical criteria, establishing a theoretical foundation for parameter selection.

Conclusions and Discussion

Main Conclusions

1. Spectral Coefficient Calibration is Essential for TKE Consistency: With unrestricted wavenumber ranges, recalibrated coefficients can eliminate deviations between computed and preset TKE values
2. Upper Wavenumber Bound Primarily Controls TKE Consistency: Increasing grid resolution to elevate the theoretical upper bound can significantly reduce TKE error
3. Lower Wavenumber Bound Controls Low-Wavenumber Spectral Fidelity: Expanding the computational domain to lower the bound improves E(κ) consistency in the low-wavenumber range
4. Wavenumber Range Selection Must Account for Grid Constraints: When bounds exceed grid-constrained limits, aliasing effects occur, causing substantial increases in E(κ) error

Limitations

1. Single Spectral Form: Research is primarily based on the von Kármán-Pao spectrum; applicability to other spectral forms requires further verification
2. Isotropy Assumption: Study assumes isotropic turbulence; anisotropic cases in practical applications are not addressed
3. Fixed Modal Distribution: Linear modal distribution is employed; effects of alternative distributions warrant investigation

Future Directions

1. Modal Number and Distribution Optimization: Investigation of modal number and distribution effects within fixed wavenumber bounds
2. Anisotropic Turbulence Extension: Extension of methodology to anisotropic turbulence scenarios
3. Practical Application Validation: Verification of parameter selection guidelines in specific CFD and CAA applications

In-Depth Evaluation

Strengths

1. Strong Systematicity: First systematic separation and analysis of different factors' effects on synthetic turbulence accuracy
2. Solid Theoretical Foundation: Establishment of a complete theoretical framework linking error sources to physical mechanisms
3. High Practical Value: Provision of explicit parameter selection guidelines with significant engineering application importance
4. Reasonable Experimental Design: Systematic analysis of individual factor effects through controlled variable methodology

Weaknesses

1. Missing Computational Cost Analysis: Lack of discussion on computational efficiency implications of different parameter choices
2. Limited Validation Cases: Analysis based on single benchmark case with insufficient diverse verification
3. Insufficient Comparison with Actual Turbulence: Lack of direct comparative validation against DNS or experimental data

Impact

1. Theoretical Contribution: Provision of important theoretical insights for the synthetic turbulence field
2. Application Value: Provision of practical parameter selection tools for CFD and CAA engineers
3. Methodological Significance: Established analytical framework generalizable to other synthetic turbulence methods

Applicable Scenarios

1. Computational Aeroacoustics: Broadband noise prediction and acoustic design
2. Turbulent Inflow Generation: Boundary condition specification for high-fidelity simulations
3. Particle Diffusion Research: Diffusion modeling in environmental and industrial applications

References

This paper cites 21 important references covering the historical development and recent advances in synthetic turbulence methods, providing a solid theoretical foundation for the research. Key references include Kraichnan's pioneering work, sampling theory analysis by Saad et al., and recent coefficient calibration methods by Guglielmi et al.