Some considerations on the accuracy and frequency response of some derivative filters applied to particle image velocimetry vector fields

Foucaut, Jean-Marc; Stanislas, Michel

doi:10.1088/0957-0233/13/7/313

Cited by 138 publications

(156 citation statements)

References 10 publications

Supporting

Mentioning

153

Contrasting

Unclassified

Order By: Relevance

“…33 Gaussian smoothing, characterized by a filter width of σ ≈ 10 y + , of the velocity field is performed after which the available velocity gradients are calculated using a second order least squares difference scheme. 34 Care is taken to ensure that the filter width is smaller than the Taylor micro-scale λ T ≈ 180 y + , to prevent affecting the relevant flow scales. Subsequent conditional sampling is performed on the raw unfiltered velocity fields.…”

Section: A T/nt Interfacementioning

confidence: 99%

Interfaces and internal layers in a turbulent boundary layer

et al. 2015

View full text Add to dashboard Cite

New experimental research is presented on the characteristics of interfaces and internal shear layers that are present in a turbulent boundary layer (TBL). The turbulent/non-turbulent (T/NT) interface at the outer boundary of the TBL shows the presence of a finite jump in streamwise velocity and is characterised by a thin shear layer. It appears that similar layers of high shear occur also within the TBL which separate regions of almost uniform momentum. It turns out that they exhibit similar characteristics as the external T/NT interface. Furthermore, the spatial growth rate of the TBL, that is derived from theoretical analysis, can be correctly predicted from a momentum balance near the external T/NT interface. Similarly, the entrainment velocities for the average internal layers have been determined. Results indicate that internal layers move slower in the vicinity of the wall, whereas they move faster than the large scale boundary layer growth rate in the outer region of the TBL. It is believed that shear layers bound large scale flow regions of approximately uniform momentum. Hence, the entrainment velocities of these internal layers may be interpreted as growth rates of the large scale motions in a TBL. C 2015 AIP Publishing LLC. [http://dx

show abstract

Section: A T/nt Interfacementioning

confidence: 99%

Interfaces and internal layers in a turbulent boundary layer

et al. 2015

View full text Add to dashboard Cite

show abstract

“…It must be noted that the vorticity sampling window size determines the vorticity computation based on a particular vortex scale. [14][15][16] There is a need to obtain the optimal vorticity calculation and mapping, which produces vorticity values with the same range in each localized spatial group such that a vortex can be visually observed when the values are mapped onto a color scale. Typically, setting a large vorticity sampling size can remove noise fluctuations and obtain vorticity maps with minimum intra-class and maximum inter-class vorticity group values.…”

Section: Flow Grid Representationmentioning

confidence: 99%

Cardiac Flow Analysis Applied to Phase Contrast Magnetic Resonance Imaging of the Heart

et al. 2009

View full text Add to dashboard Cite

Abstract-Phase contrast magnetic resonance imaging is performed to produce flow fields of blood in the heart. The aim of this study is to demonstrate the state of change in swirling blood flow within cardiac chambers and to quantify it for clinical analysis. Velocity fields based on the projection of the three dimensional blood flow onto multiple planes are scanned. The flow patterns can be illustrated using streamlines and vector plots to show the blood dynamical behavior at every cardiac phase. Large-scale vortices can be observed in the heart chambers, and we have developed a technique for characterizing their locations and strength. From our results, we are able to acquire an indication of the changes in blood swirls over one cardiac cycle by using temporal vorticity fields of the cardiac flow. This can improve our understanding of blood dynamics within the heart that may have implications in blood circulation efficiency. The results presented in this paper can establish a set of reference data to compare with unusual flow patterns due to cardiac abnormalities. The calibration of other flow-imaging modalities can also be achieved using this well-established velocityencoding standard.

show abstract

“…3 top line). This is expected since, for EU, an increase of time separation, dt, decreases the precision error, ε 2 u 2dt 2 , associated with the uncertainty on the velocity field, but intensifies the truncation error, dt 2 6 ∂ 3 u ∂t , stemming from the discretisation scheme (see van Oudheusden, 2013), a behaviour also observed for derivative filters (Foucaut and Stanislas, 2002). Therefore, for good quality data, as is the case of zero noise, it is the truncation error that is more prominent and dictates small time separations to be used.…”

Section: Convection Velocity and Frame Time-separation Dependencementioning

confidence: 86%

Full-field pressure from snapshot and time-resolved volumetric PIV

2016

View full text Add to dashboard Cite

This paper deals with pressure estimation from snapshot and time-resolved three component (3C) volumetric PIV data using Taylor's hypothesis, an Eulerian and a pseudo-Lagrangian approach. The Taylor's hypothesis approach has been shown to provide accurate results for pressure in the case of 3C planar PIV data with an appropriate choice of convection velocity (de Kat and Ganapathisubramani, 2013) and here we extend its use on 3C volumetric velocity snapshots. Application of the techniques to synthetic data shows that the Taylor's hypothesis approach performs best using the streamwise mean as the convection velocity and is affected the least by noise, while the Eulerian approach suffers the most. In terms of resolution, the pseudoLagrangian approach is the most sensitive. Its accuracy can be improved by increasing the frame time-separation when computing the material derivative, at the expense of volume loss from fluid parcels leaving the FOV. Comparison of the techniques on turbulent boundary layer data with DNS supports these observations and shows that the Taylor's hypothesis approach is the only way we can get pressure when time information is not present.

show abstract

Some considerations on the accuracy and frequency response of some derivative filters applied to particle image velocimetry vector fields

Cited by 138 publications

References 10 publications

Interfaces and internal layers in a turbulent boundary layer

Interfaces and internal layers in a turbulent boundary layer

Cardiac Flow Analysis Applied to Phase Contrast Magnetic Resonance Imaging of the Heart

Full-field pressure from snapshot and time-resolved volumetric PIV

Contact Info

Product

Resources

About