Eigenmode decomposition of turbulent velocity fields behind a swept, backward-facing step

Jürgens, Werner; Kaltenbach, Hans-Jakob

doi:10.1088/1468-5248/4/1/018

Cited by 9 publications

(9 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Notice that the POD of such a step flow is presented in [16] for a Reynolds number about twice smaller and that a low compactness of the kinetic energy distribution within the POD basis is observed. Three facts make it possible to understand this difference.…”

Section: Three-dimensional Turbulent Flowmentioning

confidence: 99%

“…Three facts make it possible to understand this difference. Firstly, the database of [16] was computed by direct numerical simulation so contains a larger range of structures than our LES database. Secondly, the POD is combined with a Fourier decomposition which is not strictly equivalent to a full POD (which is optimal) even if a spectrum close to the POD spectrum is generally expected.…”

Section: Three-dimensional Turbulent Flowmentioning

confidence: 99%

See 1 more Smart Citation

Calibrated reduced-order POD-Galerkin system for fluid flow modelling

Couplet

Basdevant

Sagaut

2005

Journal of Computational Physics

227

183

View full text Add to dashboard Cite

Section: Three-dimensional Turbulent Flowmentioning

confidence: 99%

Section: Three-dimensional Turbulent Flowmentioning

confidence: 99%

Calibrated reduced-order POD-Galerkin system for fluid flow modelling

Couplet

Basdevant

Sagaut

2005

Journal of Computational Physics

227

183

View full text Add to dashboard Cite

“…6), which results into a cycle-independent averaged flow field u and cycle-dependent fluctuations u 0 ðnÞ, a subset-dependent averaged flow fieldûðsÞ and cycle-dependent fluctuations u Ã ðnÞ appear in Equation (9). Averaging the resulting subsetdependent flow fields yields the ensemble average:…”

Section: Principles Of Conditional Averagingmentioning

confidence: 99%

“…whereby the subset specific small-scale fluctuations u Ã i ðnÞ are calculated by Equation (9). Please note that each cycle n is associated with one subset s. The root mean square of the coherent structure fluctuations is:…”

Section: Principles Of Conditional Averagingmentioning

confidence: 99%

“…Furthermore, the remaining fluctuations are expected not to be associated with a variation of a coherent motion but rather being smaller scale turbulent fluctuations. To achieve this, our method combines POD, which has been used previously for the identification and investigation of velocity field fluctuations based both on experimental and numerical data [3][4][5][6][7][8][9], and conditional averaging [10].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identification of Large-Scale Structure Fluctuations in IC Engines using POD-Based Conditional Averaging

Buhl

Hartmann

Hasse

2015

Oil Gas Sci. Technol. – Rev. IFP Energies nouvelles

View full text Add to dashboard Cite

-Cycle-to-Cycle Variations (CCV) in IC engines is a well-known phenomenon and the definition and quantification is well-established for global quantities such as the mean pressure. On the other hand, the definition of CCV for local quantities, e.g. the velocity or the mixture distribution, is less straightforward. This paper proposes a new method to identify and calculate cyclic variations of the flow field in IC engines emphasizing the different contributions from largescale energetic (coherent) structures, identified by a combination of Proper Orthogonal Decomposition (POD) and conditional averaging, and small-scale fluctuations. Suitable subsets required for the conditional averaging are derived from combinations of the the POD coefficients of the second and third mode. Within each subset, the velocity is averaged and these averages are compared to the ensemble-averaged velocity field, which is based on all cycles. The resulting difference of the subset-average and the global-average is identified as a cyclic fluctuation of the coherent structures. Then, within each subset, remaining fluctuations are obtained from the difference between the instantaneous fields and the corresponding subset average. The proposed methodology is tested for two data sets obtained from scale resolving engine simulations. For the first test case, the numerical database consists of 208 independent samples of a simplified engine geometry. For the second case, 120 cycles for the well-established Transparent Combustion Chamber (TCC) benchmark engine are considered. For both applications, the suitability of the method to identify the two contributions to CCV is discussed and the results are directly linked to the observed flow field structures.Résumé -Identification de fluctuations de structure aux grandes échelles dans les moteurs à combustion interne utilisant des moyennes conditionnelles basées sur la POD -Les variations cycle à cycle (VCC) dans les moteurs à combustion interne sont un phénomène bien connu, et leur définition et quantification sont bien établies pour des quantités globales telles que la pression moyenne. Toutefois, la définition des VCC pour des quantités locales, par exemple la vitesse ou le mélange, est moins simple. Le présent article propose une nouvelle méthode visant à identifier et à calculer des variations cycliques de l'écoulement dans les moteurs à combustion interne, en mettant l'accent sur les différentes contributions provenant de structures énergétiques aux grandes échelles (cohérentes), identifiées en combinant la détermination d'une moyenne conditionnelle et par décomposition orthogonale aux valeurs propres (Proper Orthogonal Decomposition, POD) et de fluctuations aux petites échelles. Des sous-ensembles adaptés requis pour l'établissement d'une moyenne conditionnelle sont tirés de la combinaison de coefficients POD des second et troisième modes. Dans chaque sous-ensemble, la vitesse est moyennée et ces moyennes sont comparées à la moyenne d'ensemble du champ de vitesse basée sur l'ensemble d...

show abstract

Application of proper orthogonal decomposition methods in reactive pore diffusion simulations

et al. 2014

View full text Add to dashboard Cite

Reactive pore diffusion is an important process in automotive exhaust‐gas aftertreatment modelling the overall conversion of pollutants. It features highly nonlinear source terms from chemical reactions coupled with transport processes. This work examines the application of model reduction by proper orthogonal decomposition. It is shown that this technique can be successfully applied to the system using separate bases for each species. Using a basis obtained for baseline conditions, predictions can be made for species profiles within a pore system for different conditions, potentially leading to significantly reduced computational requirements.

show abstract

Eigenmode decomposition of turbulent velocity fields behind a swept, backward-facing step

Cited by 9 publications

References 31 publications

Calibrated reduced-order POD-Galerkin system for fluid flow modelling

Calibrated reduced-order POD-Galerkin system for fluid flow modelling

Identification of Large-Scale Structure Fluctuations in IC Engines using POD-Based Conditional Averaging

Application of proper orthogonal decomposition methods in reactive pore diffusion simulations

Contact Info

Product

Resources

About