Energie-Innovationszentrum / Scientific Computing Lab
Refine
Document Type
Way of publication
- Open Access (8)
Language
- English (15)
Keywords
- one-dimensional turbulence (5)
- stochastic modeling (3)
- turbulent boundary layer (2)
- turbulent mixing (2)
- AI teaching (1)
- AI transfer (1)
- Heat and mass transfer (1)
- Kelvin-Helmholtz instability (1)
- Pipe flow (1)
- Smoothed Particle Hydrodynamics (SPH) (1)
- Spanwise curvature effects (1)
- Stochastic turbulence modeling (1)
- atmospheric turbulence (1)
- boundary layer theory (1)
- electrohydrodynamic turbulence (1)
- heat transfer (1)
- hierarchical parcel swapping (HiPS) (1)
- high Reynolds number (1)
- high Schmidt number (1)
- lattice Boltzmann method (1)
- linear stability analysis (1)
- machine learning (1)
- map-based stochastic advection modeling (1)
- multiphysical boundary layers (1)
- multiple passive scalars (1)
- multiscale flow (1)
- one-dimensional turbulence (ODT) (1)
- passive scalar (1)
- periodic forcing (1)
- reduced-order stochastic modeling (1)
- rotating flow (1)
- round jet (1)
- spanwise curvature (1)
- stochastic turbulence modeling (1)
- thermal convection (1)
- turbulence modeling (1)
- turbulent coaxial pipe flow (1)
- turbulent drag enhancement (1)
- turbulent jet (1)
- two-phase flow (1)
- volatility modeling and prediction (1)
- wind energy (1)
Institute
BTU
A journal paper in Advances in Science and Research details the numerical modeling approach used to create the data. Here, the model input files, the raw data, processed data, and plot scripts are provided that support the research.
The code used to generate the data is an extended version of the one-dimensional turbulence (ODT) model. The current model implementation utilizes an adaptive grid that further increases numerical efficiency. A reduced version of the adaptive ODT code used in this work is available free of charge at:
https://github.com/BYUignite/ODT
The bash script makePlot.sh is the top-level driver and contains all additional information about the cases. Some other Details are provided by low-level README files. Python-3.8 is required to run the scripts.
A characteristic-based Off-Lattice Boltzmann Method (OLBM) and a stochastic One-Dimensional Turbulence (ODT) model is utilized for numerical simulation of turbulent thermal convection. Standalone ODT results for low-order statistics are compared with those from various eddy-viscosity-based subgrid-scale models utilized in Large-Eddy Simulations (LES) with OLBM. The predictive capabilities of both approaches are discussed by comparison with available reference Direct Numerical Simulation (DNS) results. All turbulence models are able to predicted the mean temperature, but fail to fully capture fluctuations. While the OLBM aims to represent large-scale structures, it misses some constitutional small-scale fluctuations. By contrast, the reduced-order ODT model captures small-scale fluctuations in the vicinity of the wall, but cannot resolve the organized bulk flow. Here, the modeling capabilities of both OLBM and ODT as standalone tools are discussed. On this basis, a strategy for the incorporation of ODT as wall model in OLBM is suggested.
The present study numerically investigates turbulent momentum transfer in concentric coaxial (annular) pipe flow with small radius ratios (𝜂 = 𝑅𝑖∕𝑅𝑜 = 0.1,0.04,0.02). To model the flow, a stochastic one-dimensional turbulence (ODT) model formulated for cylindrical geometry is used that provides full-scale resolution along a representative radial coordinate. The present investigation extends the model validation by Tsai et al. (PAMM,22:e202200272, 2023), to radius ratios smaller than 0.1 and addresses boundary layers with strong span-wise curvature effects. The focus is on the assessment and analysis of statistical flow features in the vicinity of the inner cylinder wall, particularly in cases with small radius ratios. Following Boersma & Breugem (Flow Turbul. Com-bust.,86:113–127, 2011), classical boundary-layer and mixing-length theory is utilized to analyze the model predictions. The results demonstrate that the ODT model captures leading-order curvature and mixing-length effects by its physics-compatible construction. Utilizing the model for extrapolation to high Reynolds numbers inaccessible to conventional high-fidelity numerical approaches shows that curvature effects persist and nonlocally affect the entire boundary layer. The model results provide support for a spanwise-curvature-modified wall function.
Modeling turbulent mixing is a standing challenge for nonpremixed chemically reacting flows. Key complications arise from the requirement to capture all relevant scales of the flow and the necessity to distinguish between turbulent advective transport and molecular diffusive transport processes. In addition, anisotropic mean shear, variable advection time scales, and the coexistence of turbulent and nonturbulent regions need to be represented. The fundamental issues at stake are addressed by investigating multi-scalar mixing in a three-stream coaxial jet with a map-based stochastic one-dimensional turbulence model. ODT provides full-scale resolution at affordable costs by a radical reduction of complexity compared to high-fidelity Navier–Stokes solvers. The approach is partly justified by an application of the boundary-layer approximation, but neglects fluctuating axial pressure gradients. It is demonstrated that low-order scalar statistics are reasonably but not fully captured. Despite this shortcoming, it is shown that the model is able to reproduce experimental state-space statistics of multi-stream multi-scalar mixing. The model therefore offers physics-compatible improvements in multi-stream mixing modeling despite some fundamental limitations that remain from unjustified assumptions.
Certifying Fully Dynamic Algorithms for Recognition and Hamiltonicity of Threshold and Chain Graphs
(2023)
Two-sided Bogoliubov inequality to estimate finite-size effects in quantum molecular simulations
(2022)
Turbulent concentric coaxial pipe flows are numerically investigated as canonical problem addressing spanwise curvature effects on heat and momentum transfer that are encountered in various engineering applications. It is demonstrated that the wall-adapting local eddy-viscosity (WALE) model within a large-eddy simulation (LES) framework, without model parameter recalibration, has limited predictive capabilities as signalized by poor representation of wall curvature effects and notable grid dependence. The identified lack in the modeling of radial transport processes is therefore addressed here by utilizing a stochastic one-dimensional turbulence (ODT) model. A standalone ODT formulation for cylindrical geometry is used in order to assess to which extent the predictability can be expected to improve by utilizing an advanced wall-modeling strategy.
The detailed numerical representation of turbulent mixing processes is a standing challenge for non-premixed chemically reacting flows. The full range of relevant flow scales needs to be captured and it is also necessary to distinguish turbulent advective from molecular diffusive processes in order to represent Reynolds and Schmidt number effects. These requirements are addressed here by utilizing two different map-based stochastic turbulence modeling strategies. The one-dimensional turbulence (ODT) model utilizes event-based turbulence modeling, whereas the hierarchical parcel-swapping (HiPS) model is a fully event-based mixing model. ODT provides full-scale resolution at affordable costs by dimensional model reduction based on the boundary-layer approximation to shear flow. HiPS is far less costly than ODT but currently limited to locally homogeneous isotropic turbulence. The physics-compatible modeling capabilities with respect to phase-space representation of turbulent mixing are demonstrated for two canonical cases using standalone model formulations.
Capturing features of turbulent Ekman–Stokes boundary layers with a stochastic modeling approach
(2023)
Atmospheric boundary layers (ABLs) exhibit transient processes on various time scales that range from a few days down to seconds, with a scale separation of the large-scale forcing and the small-scale turbulent response. One of the standing challenges in modeling and simulation of ABLs is a physically based representation of complex multiscale boundary layer dynamics. In this study, an idealized time-dependent ABL, the so-called Ekman–Stokes boundary layer (ESBL), is considered as a simple model for the near-surface flow in the mid latitudes and polar regions. The ESBL is driven by a prescribed temporal modulation of the bulk–surface velocity difference. A stochastic one-dimensional turbulence (ODT) model is applied to the ESBL as standalone tool that aims to resolve all relevant scales of the flow along a representative vertical coordinate. It is demonstrated by comparison with reference data that ODT is able to capture relevant features of the time-dependent boundary layer flow. The model predicts a parametric enhancement of the bulk–surface coupling in the event of a boundary layer resonance when the flow is forced with the local Coriolis frequency. The latter reproduces leading order effects of the critical latitudes. The model results suggest that the bulk flow decouples from the surface for high forcing frequencies due to a relative increase in detached residual turbulence.
Joint predictive modeling of hydrodynamics and electrokinetics is a standing numerical challenge but crucial for various applications in electrochemistry and power engineering. The present lack in modeling of electrohydrodynamic (EHD) turbulent flows lies in the treatment of small-scale processes and scale interactions. To overcome these limitations, a stochastic one-dimensional turbulence (ODT) model is utilized. The model aims to resolve all scales of the flow, but only on a notional line-of-sight, modeling turbulent advection by a stochastically sampled sequence of eddy events that punctuate deterministic molecular diffusive advancement. In this study, two canonical flow configurations are investigated that address different coupling strategies and flow physics. First, EHD effects in a variable-density vertical pipe flow of an ideal gas with an inner concentric electrode are investigated with a one-way coupled model formulation. Electric fields are generated by means of a corona discharge and the corresponding effect of a fixed ionic charge density field. Second, in order to reduce physical complexity, EHD effects the turbulent boundary layers in plane Couette flow of an isothermal univalent ionic liquid are investigated with a fully coupled model formulation. Both application cases demonstrate that ODT has predictive capabilities due to multiscale resolution of transport processes. Present results suggest that more expensive fully than one-way coupling of electrokinetics is crucial when charge relaxation times are significantly larger than the mean advection time scale.
The transformation of the energy sector is based on the integration of various renewable sources, such as wind and solar energy. One of the key challenges for the integration of these sources into the existing power grid is their erratic and sometimes discontinuous availability (volatility). Wind energy is one of the most relevant sources of CO2 neutral electric energy, but volatile due to fluctuating wind fields on multiple scales. This has already been realized so that senors provide real-time information on the scale of individual wind turbines. However, fore- casting remains an unresolved problem since numerical weather prediction models cannot provide the necessary level of detail. New modeling strategies are required that integrate turbine-scale and meso-scale information for accurate site-specific short-term prediction. Present and forthcoming research aims to incorporate fluctuations on multiple levels of fidelity, depending on the abstraction layer