Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Studies of Transient and Pulsating flows with application to Hydropower
Luleå University of Technology, Department of Engineering Sciences and Mathematics, Fluid and Experimental Mechanics.ORCID iD: 0000-0002-3349-601X
2018 (English)Doctoral thesis, comprehensive summary (Other academic)Alternative title
Studier av transient och pulserande strömning med applikation inom vattenkraft (Swedish)
Abstract [en]

The rotational motion of a hydraulic turbine runner makes pulsating flows ubiquitous in different locations of the machine. The cyclic loading thus induced may generate large pressure forces acting periodically on both stationary and rotating parts. In addition to the presence of pulsating flows in a turbine runner, transient flows are encountered at an increasingly higher rate due to the continual installation of intermittent sources of renewable energy, such as wind and solar power. To mitigate the imbalance that these unpredictable sources induce on the frequency of the electrical grid, hydropower turbines are enforced to regulate their power production, and consequently flow rate, thus leaving them to operate under transient conditions. In terms of wear and fatigue, a startup or shutdown of a hydraulic turbine corresponds to 10-20 hours of steady state operation at the design point. Transient operation of a hydraulic machine can, however, also be used in favour for measuring the discharge through the turbine using the pressure-time method. A better understanding of pulsating and transient flows thus has the potential both to mitigate problems associated with them, and to increase the accuracy with which the turbine flow rate can be measured; two great merits for the hydropower community. In light of this observation, the following work constitutes a fundamental investigation of transient and pulsating flows performed in a straight pipe.Studies have been performed experimentally using particle image velocimetry, hot-film anemometry, laser Doppler velocimetry and pressure sensors.

A chief finding is that the time-development of the wall shear stress and near-wall turbulence fields exhibit significant similarity between transient and pulsating flows, despite the different conditions of the mean flow. Whereas the former is initiated from a statistically steady state, the latter is constantly subjected to a time-varying forcing. Both types of unsteady flows have previously been investigated in detail; however, any potential similarity between them has, largely, been unexplored. An important implication of this finding, then, is that knowledge acquired in one type of unsteady flow can be used, if not interchangeably, at least as a guidance for the expected behaviour in the other type of flow. An example is the development of unsteady turbulence models. Another important finding is that the frictional losses arising during the late stage of a pressure-time flow rate measurement can be accurately modelled using an analytical laminar formulation of the wall shear stress, despite the bulk of the flow being turbulent. The formulation of the wall shear stress has potential to be further improved by incorporating a damping-function.

Place, publisher, year, edition, pages
Luleå: Luleå University of Technology, 2018.
Series
Doctoral thesis / Luleå University of Technology 1 jan 1997 → …, ISSN 1402-1544
Keywords [en]
Turbulent flows, Pipe flow, Unsteadiness, Friction modeling
National Category
Fluid Mechanics and Acoustics
Research subject
Fluid Mechanics
Identifiers
URN: urn:nbn:se:ltu:diva-67865ISBN: 978-91-7790-065-8 (print)ISBN: 978-91-7790-066-5 (electronic)OAI: oai:DiVA.org:ltu-67865DiVA, id: diva2:1188187
Public defence
2018-04-20, E231, Luleå, 09:30 (English)
Opponent
Supervisors
Available from: 2018-03-07 Created: 2018-03-06 Last updated: 2018-04-05Bibliographically approved
List of papers
1. Wall friction and velocity measurements in a double-frequency pulsating turbulent flow
Open this publication in new window or tab >>Wall friction and velocity measurements in a double-frequency pulsating turbulent flow
2016 (English)In: Journal of Fluid Mechanics, ISSN 0022-1120, E-ISSN 1469-7645, Vol. 788, p. 521-548Article in journal (Refereed) Published
Abstract [en]

Wall shear stress measurements employing a hot-film sensor along with laser Doppler velocimetry measurements of the axial and tangential velocity and turbulence profiles in a pulsating turbulent pipe flow are presented. Time-mean and phase-averaged results are derived from measurements performed at pulsation frequencies ω+ = ων/u¯τ 2 over the range of 0.003-0.03, covering the low-frequency, intermediate and quasi-laminar regimes. In addition to the base case of a single pulsation imposed on the mean flow, the study also investigates the flow response when two pulsations are superimposed simultaneously. The measurements from the base case show that, when the pulsation belongs to the quasi-laminar regime, the oscillating flow tends towards a laminar state in which the velocity approaches the purely viscous Stokes solution with a low level of turbulence. For ω+ < 0.006, the oscillating flow is turbulent and exhibits a region with a logarithmic velocity distribution and a collapse of the turbulence intensities, similar to the time-averaged counterparts. In the low-frequency regime, the oscillating wall shear stress is shown to be directly proportional to the Stokes length normalized in wall units ls + (=√2/ω+), as predicted by quasi-steady theory. The base case measurements are used as a reference when evaluating the data from the double-frequency case and the oscillating quantities are shown to be close to superpositions from the base case. The previously established view that the time-averaged quantities are unaffected by the imposition of small-amplitude pulsed unsteadiness is shown to hold also when two pulsations are superposed on the mean flow

National Category
Fluid Mechanics and Acoustics
Research subject
Fluid Mechanics
Identifiers
urn:nbn:se:ltu:diva-3505 (URN)10.1017/jfm.2015.722 (DOI)000368413600014 ()2-s2.0-84953775346 (Scopus ID)155fb13a-51af-4173-9c93-79be3ef6f44f (Local ID)155fb13a-51af-4173-9c93-79be3ef6f44f (Archive number)155fb13a-51af-4173-9c93-79be3ef6f44f (OAI)
Note
Validerad; 2016; Nivå 2; 20160119 (andbra)Available from: 2016-09-29 Created: 2016-09-29 Last updated: 2018-07-10Bibliographically approved
2. On the Similarity of Pulsating and Accelerating Turbulent Pipe Flows
Open this publication in new window or tab >>On the Similarity of Pulsating and Accelerating Turbulent Pipe Flows
2018 (English)In: Flow Turbulence and Combustion, ISSN 1386-6184, E-ISSN 1573-1987, Vol. 100, no 2, p. 417-436Article in journal (Refereed) Published
Abstract [en]

The near-wall region of an unsteady turbulent pipe flow has been investigated experimentally using hot-film anemometry and two-component particle image velocimetry. The imposed unsteadiness has been pulsating, i.e., when a non-zero mean turbulent flow is perturbed by sinusoidal oscillations, and near-uniformly accelerating in which the mean flow ramped monotonically between two turbulent states. Previous studies of accelerating flows have shown that the time evolution between the two turbulent states occurs in three stages. The first stage is associated with a minimal response of the Reynolds shear stress and the ensemble-averaged mean flow evolves essentially akin to a laminar flow undergoing the same change in flow rate. During the second stage, the turbulence responds rapidly to the new flow conditions set by the acceleration and the laminar-like behavior rapidly disappears. During the final stage, the flow adapts to the conditions set by the final Reynolds number. In here, it is shown that the time-development of the ensemble-averaged wall shear stress and turbulence during the accelerating phase of a pulsating flow bears marked similarity to the first two stages of time-development exhibited by a near-uniformly accelerating flow. The stage-like time-development is observed even for a very low forcing frequency; &#x03C9;+=&#x03C9;&#x03BD;/u&#x00AF;&#x03C4;2=0.00073" role="presentation">ω+=ων/u¯¯¯2τ=0.00073 (or equivalently, ls+=2/&#x03C9;+=52" role="presentation">l+s=2/ω+−−−−√=52), at an amplitude of pulsation of 0.5. Some previous studies have considered the flow to be quasi-steady at ls+=52" role="presentation">l+s=52; however, the forcing amplitude has been smaller in those studies. The importance of the forcing amplitude is reinforced by the time-development of the ensemble-averaged turbulence field. For, the near-wall response of the Reynolds stresses showed a dependence on the amplitude of pulsation. Thus, it appears to exist a need to seek alternative similarity parameters, taking the amplitude of pulsation into account, if the response of different flow quantities in a pulsating flow are to be classified correctly.

Place, publisher, year, edition, pages
Springer, 2018
National Category
Fluid Mechanics and Acoustics
Research subject
Fluid Mechanics
Identifiers
urn:nbn:se:ltu:diva-65766 (URN)10.1007/s10494-017-9855-5 (DOI)000424665400005 ()30069140 (PubMedID)
Note

Validerad;2018;Nivå 2;2018-02-20 (rokbeg)

Available from: 2017-09-22 Created: 2017-09-22 Last updated: 2018-08-06Bibliographically approved
3. The self-similarity of wall-bounded temporally accelerating turbulent flows
Open this publication in new window or tab >>The self-similarity of wall-bounded temporally accelerating turbulent flows
2018 (English)In: Journal of turbulence, ISSN 1468-5248, E-ISSN 1468-5248, Vol. 19, no 1, p. 49-60Article in journal (Refereed) Published
Abstract [en]

From the study of viscous flow it is known that certain time-dependent laminar problems, such as the impulsively started flat plate and the diffusion of a vortex sheet, possess self-similar solutions. Previous studies of turbulent channel and pipe flows accelerating between two steady states have shown that the flow field evolves in three distinct stages. Furthermore, recent direct numerical simulations have shown that the perturbation velocity, i.e. the surplus velocity from the initial value, in an impulsively accelerating turbulent channel and pipe flow also possesses a self-similar distribution during the initial stage. In here, these results are developed analytically and it is shown that accelerating flows in which the centreline velocity develops as Uc(t) = U0(t/t0)m will possess a self-similar velocity distribution during the initial stage. The displacement thickness of the perturbation velocity is shown to be dependent only on the type of acceleration, and not on the initial Reynolds number, the acceleration rate or the change in Reynolds number. The derived formulas are verified with good agreement against measurements performed in a linearly accelerating turbulent pipe flow and with data from channel flow simulations.

Place, publisher, year, edition, pages
Taylor & Francis, 2018
National Category
Fluid Mechanics and Acoustics
Research subject
Fluid Mechanics
Identifiers
urn:nbn:se:ltu:diva-66785 (URN)10.1080/14685248.2017.1390239 (DOI)000427894200003 ()
Note

Validerad;2018;Nivå 2;2017-12-15 (andbra)

Available from: 2017-11-28 Created: 2017-11-28 Last updated: 2018-04-06Bibliographically approved
4. Transient wall shear stress measurements and estimates at high Reynolds numbers
Open this publication in new window or tab >>Transient wall shear stress measurements and estimates at high Reynolds numbers
2017 (English)In: Flow Measurement and Instrumentation, ISSN 0955-5986, E-ISSN 1873-6998, Vol. 58, p. 112-119Article in journal (Refereed) Published
Abstract [en]

Transient wall shear stress measurements using hot-film anemometry have been performed in a large-scale laboratory setup at high Reynolds numbers. Starting from Reynolds numbers 1.7×1061.7×106 and 0.7×1060.7×106, the flow was brought to a complete rest by closing a knife gate thus replicating a pressure-time (also know as Gibson) flow rate measurement in a hydropower plant. Ensemble-averaged mean wall shear stresses obtained from 22 repeated runs have been compared with estimates obtained using the pressure-time method. The objective of the work has been to assess the accuracy of the frictional formulation entering the pressure-time integral. It is shown that both the standard method, a quasi-steady approach as well as the recently introduced unsteady method all reproduce the measured wall shear stresses quantitatively during most of the transient. The last phase, following the complete closure of the gate, which is characterized by a slow decay towards zero shear stress at the wall is, however, not captured by the available methods. In general, the unsteady formulation produces the smallest flow rate estimation error, which in turn, implies the best modeling of the frictional losses.

Place, publisher, year, edition, pages
Elsevier, 2017
National Category
Fluid Mechanics and Acoustics
Research subject
Fluid Mechanics
Identifiers
urn:nbn:se:ltu:diva-66149 (URN)10.1016/j.flowmeasinst.2017.10.003 (DOI)000419419600014 ()2-s2.0-85032350533 (Scopus ID)
Note

Validerad;2017;Nivå 2;2017-11-06 (andbra)

Available from: 2017-10-16 Created: 2017-10-16 Last updated: 2018-03-06Bibliographically approved
5. Characteristics of the wall shear stress in pulsating wall-bounded turbulent flows
Open this publication in new window or tab >>Characteristics of the wall shear stress in pulsating wall-bounded turbulent flows
2018 (English)In: Experimental Thermal and Fluid Science, ISSN 0894-1777, E-ISSN 1879-2286, Vol. 96, p. 257-265Article in journal (Refereed) Published
Abstract [en]

A pulsating turbulent pipe flow has been investigated experimentally using hot-film anemometry and particle image velocimetry. Particularly, a `paradoxical phenomenon' that is known to occur for a range of forcing frequencies and time-averaged Reynolds numbers has been investigated in detail. The paradoxical phenomenon is that the oscillating component of the wall shear stress exhibits a smaller amplitude in a turbulent flow compared to in a laminar flow exposed to the same oscillation in the pressure gradient. In here, the phenomenon is explained by splitting the response of the wall shear stress into one contribution resulting from the imposed pressure gradient , and a second contribution resulting from the oscillating Reynolds shear stress, At the conditions of maximum reduction of the wall shear stress amplitude,  and  are nearly 136 degrees out of phase. The contributions are thus interfering destructively, this being the ultimate reason for the reduced amplitude. It is also shown that the level of reduction is dependent on the imposed forcing amplitude, this in turn residing from a dependence of the time-development of the oscillating Reynolds shear stress on the forcing amplitude.

Place, publisher, year, edition, pages
Elsevier, 2018
National Category
Fluid Mechanics and Acoustics
Research subject
Fluid Mechanics
Identifiers
urn:nbn:se:ltu:diva-67866 (URN)10.1016/j.expthermflusci.2018.02.036 (DOI)000432235300025 ()2-s2.0-85044007965 (Scopus ID)
Note

Validerad;2018;Nivå 2;2018-03-23 (rokbeg)

Available from: 2018-03-06 Created: 2018-03-06 Last updated: 2018-06-11Bibliographically approved
6. Laminar similarities between accelerating and decelerating turbulent flows
Open this publication in new window or tab >>Laminar similarities between accelerating and decelerating turbulent flows
2018 (English)In: International Journal of Heat and Fluid Flow, ISSN 0142-727X, E-ISSN 1879-2278, Vol. 71, p. 13-26Article in journal (Refereed) Published
Abstract [en]

An experimental study of a pipe flow ramping monotonically between two turbulent states has been undertaken. Ensemble-averaged mean and turbulent flow quantities obtained from two-component particle image velocimetry and hot-film anemometry measurements have been presented. It is shown that the initial developments of the mean and turbulence quantities in linearly as well as impulsively accelerating and decelerating flows are similar. Specifically, the mean perturbation velocity (defined as the surplus/deficit from the initial value) can be described using self-similar expressions. The duration of this initial stage, when normalized appropriately, is shown to be approximately invariant of the type of transient imposed on the bulk flow. Data from studies of linearly accelerating and decelerating flows as well as impulsively accelerating and decelerating flows have been used to validate the results, covering four orders of magnitude of the dimensionless parameter . The highest initial Reynolds number (31,000) is, however, relatively low thus requiring further studies at high Reynolds numbers to assure the universality of the results. We have also shown that the time-development of the mean and turbulent quantities between an accelerating and a decelerating flow looses their similarity as the transient proceeds beyond the initial stage. The departure was explained by the time-evolvement of the production of turbulence kinetic energy, which exhibit differences between the two types of transients.

Place, publisher, year, edition, pages
Elsevier, 2018
National Category
Fluid Mechanics and Acoustics
Research subject
Fluid Mechanics
Identifiers
urn:nbn:se:ltu:diva-67867 (URN)10.1016/j.ijheatfluidflow.2018.03.005 (DOI)000435428900002 ()2-s2.0-85044133898 (Scopus ID)
Note

Validerad;2018;Nivå 2;2018-03-26 (andbra)

Available from: 2018-03-06 Created: 2018-03-06 Last updated: 2018-07-26Bibliographically approved
7. Improved frictional modelling for the pressure-time metho
Open this publication in new window or tab >>Improved frictional modelling for the pressure-time metho
(English)Manuscript (preprint) (Other academic)
Abstract [en]

The pressure-time method is classified as a primary method for measuring discharge in hydraulic machinery. The uncertainty in the discharge determined using the pressure-time method is typically around %; however, despite dating back almost one hundred years in time, there still exists potential to reduce this uncertainty. In this paper, an improvement of the pressure-time method is suggested by implementing a novel formulation to model the frictional losses arising in the evaluation procedure. By analyzing previously obtained data from CFD, laboratory and full-scale pressure-time measurements it is shown that the new friction model improves the accuracy of the flow rate calculation by approximately 0.1-0.2 percentage points, compared to currently utilized friction models. Despite being a small absolute improvement, the new friction model presents an important development of the pressure-time method because the relative improvement is significant.

National Category
Fluid Mechanics and Acoustics
Identifiers
urn:nbn:se:ltu:diva-67870 (URN)
Available from: 2018-03-06 Created: 2018-03-06 Last updated: 2018-03-06

Open Access in DiVA

fulltext(1262 kB)68 downloads
File information
File name FULLTEXT02.pdfFile size 1262 kBChecksum SHA-512
6054e62d705c6023f3831e2e0fad7ddf02ab9307da67c47fd934917759267045f18cfd7b44185473a3aa32adf5dfd8b0a545c9b312a96d21fd448cd1d53260fa
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Sundström, Joel
By organisation
Fluid and Experimental Mechanics
Fluid Mechanics and Acoustics

Search outside of DiVA

GoogleGoogle Scholar
Total: 73 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 1087 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf