Data-driven RANS closures for wind turbine wakes under neutral conditions

Steiner, Julia; Dwight, Richard P.; Viré, Axelle

doi:10.1016/j.compfluid.2021.105213

Cited by 25 publications

(27 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the LES we could potentially have used actuator-line models, but for consistency with RANS we prefer disc models. Details of the numerical setup such as boundary conditions and mesh resolution, as well as a validation of the CFD models with respect to wind tunnel measurements, can all be found in [12].…”

Section: Ground-truth Data Generation (Les)mentioning

confidence: 99%

“…Note that the evolution of the incoming ABL does not match exactly between RANS and LES, and hence an additional correction is required (a function of wall-distnace only), the details of which are not relevant here, see [12] for more information.…”

Section: Specification Of the Classifiermentioning

confidence: 99%

“…A model without a classifier was derived in [12] using the same elastic-net methodology as used in this paper. To evaluate whether the addition of a classifier is beneficial, this model will be used as a baseline here.…”

Section: Training Of a Correction Model Without Classifier (Reference)mentioning

confidence: 99%

“…The most successful modification proposed so far is the k − ε − f P model from van den Laan et al [11], who uses an eddy-viscosity limiter that is only active in regions with high-velocity gradients. A summary of possible modifications can be found in our previous publication [12]. The perspective for further improvements by expert modellers is in our opinion limited, given the complexity of the modelling task -especially when stratification is considered.…”

Section: Introductionmentioning

confidence: 99%

“…It is for this reason that we propose data-driven modelling as an tool for devising more accurate closures. In our previous work [13,12] we extended the data-driven framework Sparse Regression of Turbulent Stress Anisotropy (SpaRTA) first introduced by Schmelzer et al [14]. The framework introduces two nonlinear corrections to the baseline k − ε or k − ω equations: (i) an anisotropy correction and (ii) an additive correction to the transport equation for the turbulent kinetic energy.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Classifying regions of high model error within a data-driven RANS closure: Application to wind turbine wakes

Steiner¹,

Dwight²,

Viré³

2021

Preprint

View full text Add to dashboard Cite

Data-driven Reynolds-Averaged Navier-Stokes (RANS) turbulence closures are increasing seen as a viable alternative to general-purpose RANS closures, when LES reference data is available -also in wind-energy. Parsimonious closures with few, simple terms have advantages in terms of stability, interpret-ability, and execution speed. However experience suggests that closure model corrections need be made only in limited regions -e.g. in the near-wake of wind turbines and not in the majority of the flow. A parsimonious model therefore must find a middle ground between precise corrections in the wake, and zero corrections elsewhere. We attempt to resolve this impasse by introducing a classifier to identify regions needing correction, and only fit and apply our model correction there. We observe that such a classifier (which must be computed only from RANS-available quantities) is straightforward to construct, and accurate in operation. We further observe that the correction models are significantly simpler (with fewer terms) when limited to the identified regions than otherwise, but have similar or better accuracy. We apply our framework to three flows consisting of multiple wind-turbines in neutral conditions with interacting wakes.

show abstract

Section: Ground-truth Data Generation (Les)mentioning

confidence: 99%

Section: Specification Of the Classifiermentioning

confidence: 99%

Section: Training Of a Correction Model Without Classifier (Reference)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Classifying regions of high model error within a data-driven RANS closure: Application to wind turbine wakes

Steiner¹,

Dwight²,

Viré³

2021

Preprint

View full text Add to dashboard Cite

show abstract

waLBerla‐wind: A lattice‐Boltzmann‐based high‐performance flow solver for wind energy applications

Schottenhamml,

Anciaux Sedrakian,

Blondel

et al. 2024

Concurrency and Computation

View full text Add to dashboard Cite

SummaryThis article presents the development of a new wind turbine simulation software to study wake flow physics. To this end, the design and development of waLBerla‐wind, a new simulator based on the lattice‐Boltzmann method that is known for its excellent performance and scaling properties, will be presented. Here it will be used for large eddy simulations (LES) coupled with actuator wind turbine models. Due to its modular software design, waLBerla‐wind is flexible and extensible with regard to turbine configurations. Additionally it is performance portable across different hardware architectures, another critical design goal. The new solver is validated by presenting force distributions and velocity profiles and comparing them with experimental data and a vortex solver. Furthermore, waLBerla‐wind's performance is compared to a theoretical peak performance, and analyzed with weak and strong scaling benchmarks on CPU and GPU systems. This analysis demonstrates the suitability for large‐scale applications and future cost‐effective full wind farm simulations.

show abstract

Classifying Regions of High Model Error Within a Data-Driven RANS Closure: Application to Wind Turbine Wakes

Steiner¹,

Dwight²,

Viré³

2022

Flow Turbulence Combust

Self Cite

View full text Add to dashboard Cite

Data-driven Reynolds-averaged Navier–Stokes (RANS) turbulence closures are increasing seen as a viable alternative to general-purpose RANS closures, when LES reference data is available—also in wind-energy. Parsimonious closures with few, simple terms have advantages in terms of stability, interpret-ability, and execution speed. However experience suggests that closure model corrections need be made only in limited regions—e.g. in the near-wake of wind turbines and not in the majority of the flow. A parsimonious model therefore must find a middle ground between precise corrections in the wake, and zero corrections elsewhere. We attempt to resolve this impasse by introducing a classifier to identify regions needing correction, and only fit and apply our model correction there. We observe that such classifier-based models are significantly simpler (with fewer terms) than models without a classifier, and have similar accuracy, but are more prone to instability. We apply our framework to three flows consisting of multiple wind-turbines in neutral conditions with interacting wakes.

show abstract

Data-driven RANS closures for wind turbine wakes under neutral conditions

Cited by 25 publications

References 48 publications

Classifying regions of high model error within a data-driven RANS closure: Application to wind turbine wakes

Classifying regions of high model error within a data-driven RANS closure: Application to wind turbine wakes

waLBerla‐wind: A lattice‐Boltzmann‐based high‐performance flow solver for wind energy applications

Classifying Regions of High Model Error Within a Data-Driven RANS Closure: Application to Wind Turbine Wakes

Contact Info

Product

Resources

About