Turbulence and Wake Vortex Analysis

Our new paper led by Kerry Klemmer, published in Journal of Renewable and Sustainable Energy (Editors’ Pick!), develops a new fast-running wake model for both the wake velocity deficit and the wake added turbulence generated by a wind turbine, accounting for the stability of the atmospheric boundary layer. We validate the model, showing lower error than existing wake models, using multiple independent large eddy simulations under diverse stability conditions. Open-access paper: https://lnkd.in/e8rBn22D Open-source model code: https://lnkd.in/e_8gGRwM *The wake model is already integrated with the Unified Momentum Model for wind turbine rotors (Liew, Heck, & Howland Nature Commun. 2024 https://lnkd.in/eTQuvS5x) Presently, state-of-practice engineering models of mean wake momentum utilize highly empirical turbulence models that do not explicitly account for atmospheric boundary layer stability and also typically neglect the interaction between the wake momentum deficit and the turbulence kinetic energy added by the wake. We develop a novel turbulence model that predicts the wake-added turbulence kinetic energy and we couple it with a wake model based on the fast-running, parabolized Reynolds-Averaged Navier-Stokes (RANS) equations. I’m especially excited to share this paper since it is published in JRSE’s Special Collection: Flow, Turbulence, and Wind Energy that is organized in honor of Prof. Charles Meneveau for his seminal contributions to our field and his unparalleled mentorship and leadership.

For decades, we’ve relied on Eulerian frames to model vortex stretching in CFD, assuming instantaneous alignment between strain and vorticity. But RANS and LES often struggle to capture the "history" of deformation that leads to extreme dissipation. My latest work challenges the Eulerian snapshot. Using the JHTD's DNS (1.073 billion cells grid, Reλ≈433), I tracked the Lagrangian history of fluid parcels and found a "systematic phase lag" (τ>0). Read the preprint: arxiv.org/abs/2601.08862 So what is the main finding? A specific pressure topology, the Pressure Hessian saddle point (λp min<0), forms before the enstrophy burst occurs. The pressure field is, therefore, believed to be deterministic geometric precursor. Why this changes the game for many fluid dynamics applications: 1. Predictive Control: We can now detect the "geometric shadow" of turbulence before it manifests physically. This enables "look-ahead" Active Flow Control for UAVs and rotors, replacing reactive feedback loops with predictive logic. 2. MHD "Locking": In Magnetohydrodynamics, the Lorentz force suppresses this hysteresis loop, forcing the flow into "simultaneous locking." This mechanism explains how to stabilize plasma sheaths for hypersonic stealth and optimize energy bypass in scramjets. 3. Better Solvers: The work proposes a Jacobi-style deviation equation (D2ω/Dt2+K(t)ω=0) to predict the onset of the inertial range more accurately than strain-alignment models. Read the preprint to learn more: Lagrangian Phase-Lag and Geometric Precedence in Turbulent Vortex Stretching. URL: arxiv.org/abs/2601.08862 #FluidDynamics #CFD #TurbulenceModeling #DirectNumericalSimulation #Hypersonics #LagrangianDynamics #MHD #AerospaceEngineering #ScientificComputing

LES data visualisations using the Q-criterion for flow past a circular cylinder for Re =200. The Q-criterion itself is a scalar quantity derived from the velocity gradient tensor, specifically Q = (1/2)(||Ω||² - ||S||²), where Ω is the rotation rate tensor, and S is the strain rate tensor. Positive Q values identify regions where rotation dominates strain, effectively highlighting coherent vortex structures, such as those in the cylinder wake. Re=200 is the critical range where the wake transitions from 2D to 3D (typically starting around Re=190). Spanwise disturbances emerge due to secondary instabilities, like "mode A" (with a spanwise wavelength of about 3-4D) involving vortex dislocations and waviness in the primary vortices. As Re increases slightly beyond this, "mode B" (shorter wavelength ~1D) can appear, leading to finer-scale streamwise vortices. The Q-criterion in LES data reveals the onset of these 3D effects: initial spanwise undulations in the vortex cores, some rib-like structures connecting primary vortices, and mild breakup of uniformity. This marks the shift from laminar to transitional flow, where energy cascades begin to introduce weak turbulence precursors, making the wake more complex but not yet fully chaotic.

🌪️ Do you know what happens behind a wind turbine? It’s not just empty air — it’s a turbulent zone called the Wake, where wind slows down and becomes chaotic after hitting the rotor blades. This is known as the Wake Effect, and it plays a crucial role in how wind turbines are distributed across a wind farm. 🔍 Why is it so important? When a turbine extracts energy from the wind, it creates a trail of lower-speed, turbulent air behind it. Any turbine placed in this wake will: 🔻 Generate less power — up to 40% reduction in some cases. ⚙️ Experience higher fatigue loads due to wind turbulence. 📉 Have a shorter operational life if not properly maintained or spaced. 📐 How does this affect turbine layout? ✅ Spacing matters: Turbines are typically spaced 5 to 9 rotor diameters apart in the prevailing wind direction. In crosswind directions, spacing can be smaller (3 to 5 diameters), but careful modeling is needed. ✅ Wind direction matters: In regions with stable, unidirectional winds, linear rows can be optimized. In variable-wind areas, layouts may be staggered or offset to reduce overlapping wakes. ✅ Topography and turbulence modeling: Modern wind farms use CFD simulations and LiDAR data to predict wake behavior before installation. Wake steering and yaw control strategies are now being used to redirect wakes and maximize farm output. 💡 A well-designed wind farm isn’t just about installing turbines — it’s about understanding the invisible dynamics of the wind. 🌬️ The Wake Effect may be hidden, but its impact is powerful. #WindEnergy #WakeEffect #WindFarmDesign #WindTurbines #SustainableEngineering #RenewableEnergy #Turbulence #CFD #CleanPower #WindOptimization