Upload
quest-global
View
221
Download
0
Embed Size (px)
Citation preview
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
1/24
Fluid Structure Interaction study inluid Structure Interaction study inFrancis Turbine
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
2/24
Introduction
About QuEST Global
QuEST Global Engineering is a diversified Engineering Services company, employing 2000.
engine, Aerospace & Defense, Power Generation, Oil & Gas, and other verticals to cutproduct development and project costs, shorten lead times, extend capacity and maximizeengineering resources availability by providing support across the complete product life cycle
, , , ,instrumentation and controls, embedded systems development, manufacturing support,vendor management, and in-house precision machining. We leverage our local presence
and global reach to support engineering cost reduction initiatives for our customers.
About Hitachi
Hitachi, Ltd., (NYSE: HIT / TSE: 6501), headquartered in Tokyo, Japan, is a leading globaldiversified product & services company with approximately 400,000 employees worldwide.
e company o ers a w e range o systems, pro ucts an serv ces n mar et sectorsincluding information systems, electronic devices, power and industrial systems, consumerproducts, materials, logistics and financial services.
QuEST Global Hitachi Ltd. Confidential
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
3/24
Typical Hydropower Plant Setup
Francis Turbine
QuEST Global Hitachi Ltd. Confidential 3
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
4/24
Basic ConceptsHigh head turbines are subjected to dynamic forces.Some significant sources are
1. Interaction between runner and guide
2. Vortex shedding at the trailing edge of vane
3. Cavitation at the trailing edge of the runner vanes
Testing of hydraulic turbine are expensiveh d h
http://en.wikipedia.org/wiki
/File:Turbine_Francis_Worn.JPG
QuEST Global Hitachi Ltd. Confidential 4
Is there any way to reduce the testing cause?YES, through CFD
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
5/24
Basic ConceptsPhysical Problems
Analytical Solutions Experimental
CFD
What is CFD?It is the science (art) to replace the partial differential equations (which in this casedescribe the flow of fluids) with numbers and simple algebraic expressions, and tosolve them through iteration in time and/or space to obtain a final numericaldescription of the total flow field under consideration.
CFD works by dividing the region of interest into a largenumber of cells or control volumes (the mesh or grid).
It solves the following equations for any problem
QuEST Global Hitachi Ltd. Confidential 5
Conservation of Mass Conservation of Momentum Conservation of Energy
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
6/24
Basic ConceptsAdvantages1. Accuracy2. Ease-o -Use
3. Speed4. Powerful Visualizations
Disadvantages or Limitationsisadvantages or Limitations1. CFD solutions and accuracy rely upon physical models (e.g. turbulence,
compressibility, chemistry, multiphase flow, etc.).2. Solving equations on a computer invariably introduces numerical errors. Round-off error Truncation error
CFD can be used in:1. Conceptual studies of new designs2. Detailed product development3. Troubleshooting4. Redesign
QuEST Global Hitachi Ltd. Confidential 6
CFD analysis complements testing and experimentation. It reduces the total effortrequired in the laboratory.
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
7/24
OverviewHydro turbines are desired to work over a wide range of operating conditions due toload variations.
During dynamic loading, structures fails below its yield strength.
In some cases, when fluid interacts with a solid structure, exerting pressure that may causedeformation in the structure and, thus, alter the flow of the fluid itself.
When the flow induces significant stresses in the solid; however, since the resulting deformation ofthe solid is small, the flow field is not greatly affected
QuEST Global Hitachi Ltd. Confidential 7
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
8/24
Background of the study1. In the present study, Experimental testing of a Francis Turbine had shown that there is a crack
developed at the root of the runner blade after some hours of operation, and this was the
.
When looked at this problem, the flow induces stress in the turbine and the deformation isvery small and will not alter flow.
. ,
vortex shedding from the vane during operation, and this was the probable cause for the
observed structural failure in the vane
In this case, the vortex shedding frequency when match the resonance of the vane, the vane
vibrates violently and will alter the flow.
These two problems has to solved using computational methods with different approaches .
Such problems can be solved using Fluid Structure Interaction (FSI) approach.
FSI simulations can be broadly categorized as one-way or two-way coupled.
One way FSI analysis (Interaction of runner blades with guide vanes)
Two ways FSI analysis (Vortex shedding)
QuEST Global Hitachi Ltd. Confidential 8
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
9/24
Case study1. In Vortex shedding was one of the causes proposed for the failure of the original Tacoma
Narrows Bridge (Galloping Gertie) in 1940, but was rejected because the frequency of the vortex
. .
2. A thrill ride "Vertigo" at Cedar Point in Sandusky, Ohio Suffered the fate of vortex sheddingduring the winter of 2001, one of the three towers collapsed. The ride was closed for the winter
at the time.
7th November1940; 11:00 AM
QuEST Global Hitachi Ltd. Confidential 9
http://www.mae.cornell.edu/images/etc/fluids
/structure/tacoma.gif
http://www.cigital.com/justiceleague/wp-content/uploads
/2007/07/bridge.gif
1940; 11:00 AM
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
10/24
Contents1. Executive Summary
a Objective. Objectiveb. Approach
2 Model and Assumptions. Model and Assumptions3. Boundary Conditions4 R lt d Di i. Results and Discussions5. Conclusions
QuEST Global Hitachi Ltd. Confidential 10
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
11/24
Executive SummaryObjective of the studyOne Way FSI AnalysisTo develop the numerical methodology of Fluid Structure Interaction to find the intensity of stresslevels at root of the Francis Runner at operation condition
Two Way FSI AnalysisTo develop the numerical methodology of Fluid Structure Interaction to evaluate the ability of anunsteady numerical simulation to accurately reproduce the vortex shedding frequency on thena ura requency o e s ay vane.
QuEST Global Hitachi Ltd. Confidential 11
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
12/24
Executive SummaryApproachOne Way FSI Analysisne Way FSI Analysis1. Perform the unsteady CFD simulation of Francis Turbine and save the result file2. Map the results (Pressure) on the structure and perform the FEA simulation
.
Two Way FSI Analysis1. 2D Flow field Analysis to Determine Required Mesh Density
2. 3D Modal Analysis of the Vane to Determine Natural Frequencies
3. 3D Flow field Analysis of the Vane to Determine Span wise Variation in Shedding Frequency
4. 2D Flow field Analysis of the Vane to Determine Shedding Frequency at Various Flow rates
.
the stay vane
QuEST Global Hitachi Ltd. Confidential 12
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
13/24
Model and AssumptionsOne Way FSI A design geometry was used for creating a CFD and FEA model
It should be noted that the design model does not include the fillets
Cavitation effect are neglected do to higher computational time
Draft tube
Fluid
Domain
Solid
oma n
QuEST Global Hitachi Ltd. Confidential 13eometry Mesh
Guide vaneSolid region
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
14/24
Model and AssumptionsTwo Way FSIA design geometry was used for creating a CFD and FEA model
It should be noted that the design model does not include the fillets
Cavitation effect are neglected do to higher computational time
uDomain
FluidMesh
QuEST Global Hitachi Ltd. Confidential 14
SolidDomain
SolidMesh
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
15/24
Boundary ConditionsWorking Fluid : WaterRunner
Upper
Cavity
Inlet
Vanesra t
TubeLower
Cavity
Outlet
One Way FSI
QuEST Global Hitachi Ltd. Confidential 15
Turbine Material :
Structural Steel
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
16/24
Boundary ConditionsOutlet
Fluid Structure
Interaction
Blade
Frictionless Support
(Symmetry)Inlet
QuEST Global Hitachi Ltd. Confidential 16
Two Way FSI
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
17/24
Results and DiscussionsObservationsOne Way FSI Analysisy1. Form this analysis, it was seen that the stress were maximum at the root of the runner leading
edge.
2. Form the stress harmonic levels is seen at the root that the cause due the interaction of Guideane an unner a e.
3. This analysis doesn't conclude the failure of the runner at root. Fatigue analysis has to becarried our to confirm.
Variation of Nomalized Von Mises Stress
Hub0.98
1.00
1.02
ax
mm
0.92
0.94
0.96
/
QuEST Global Hitachi Ltd. Confidential 17
values were monitored 0.90
0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035
Time (s)
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
18/24
Results and DiscussionsOne Way FSIThe Peak stresses in the runner were
Normalized Pressure Fluctuation at
Trailing Edge of Guide Vane
1.04
predicted to occur at the LE root near the
Hub region.These unsteady pressure effects are quiteconsiderable ~ 4.5% variation durin each 0.96
0.98
1.00
.
Ps/Ps(Max)
Normalized Von-Mises Stress
vane passing. This results in huge variationsin stress levels.
0.92
0.94
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
0.8
1.0
1.2
ss
LimitationsNode to node connectivity of fluid
me s
0.4
0.6
NormilizedS
tre
the accuracy of the stress prediction.
QuEST Global Hitachi Ltd. Confidential 18
0.0
0.2
Hub Leading Edge Hub Trailing Edge Shroud Leading Edge Shroud Trailing Edge
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
19/24
Results and DiscussionsObservationsTwo Way FSI Analysisy1. Analysis was carried our with 3 Mesh configurations 13,000 cells, 16,000 cells and 3,50,000
cells. It was seen that 16,000 and 3,50,000 cell had a close match of results and for furtheranalysis 16,000 cells mesh were used.
Cases Mesh Size First NaturalFrequencySecond Natural
Frequency1 13,000 91.55 (Hz) 183.1 (Hz)2 16,000 85.45 (Hz) 170.9 (Hz)3 3 50 000 85.45 Hz 164.8 Hz
QuEST Global Hitachi Ltd. Confidential 19
2. It was seen from 3D Modal analysis, the major frequencies (Bending: 73 Hz & Torsional: 155 Hz)
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
20/24
Results and DiscussionsObservationsTwo Way FSI Analysisy3. From CFD results, FFT calculations for different points were done and when plotted, it was seen
that the frequency had very little effect over the span. This concludes this analysis can be donewith 2D to determine the vortex shedding frequency.
Frequency Distribution behind the Trailing
1
2
3
4
5
6
2.0
2.5
de
Point 1 Point 2 Point 3
Point 4 Point 5 Point 6
7
8
9
Symmetry0.5
1.0
.
Amplitu
Point 7 Point 8 Point 9
QuEST Global Hitachi Ltd. Confidential 20
Points just behind the
hydrofoil at different spans
0.0
0 200 400 600 800 1000 1200 1400
Frequency (Hz)
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
21/24
Results and DiscussionsObservationsTwo Way FSI Analysisy4. For 2D CFD analysis with different inlet velocities,
a. From the FFT analysis it was seen that at inlet velocity of 12m/s, the vortex sheddingfrequency is 73 Hz
b. This concludes that for inlet velocity of 12m/s, the vortex shedding frequency matches thenatural frequency of the model
c. So, FSI analysis can be done with a inlet velocity of 12m/s
Velocity MaximumFrequency(m/s) (Hz)
Frequency plot @ R1P1
2
2.5
73
12 73
13 78 1
1.5
Amplitude
QuEST Global Hitachi Ltd. Confidential 21
0
0.5
0 100 200 300 400 500 600 700 800 900
Frequency (Hz)
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
22/24
Results and DiscussionsObservationsTwo Way FSI Analysisy5. The Stress were monitored at the root of the blade. The values were around 10 times less then
the yield stress of the material.
Von Mises Stress (MPa)Displacement of the Hydrofoil at Trailing Edge in Mid Span
4
5
0
1
2
3
ement(mm)
-4
-3
-2
-1
Displa
QuEST Global Hitachi Ltd. Confidential 22
-5
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Time (s)
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
23/24
ConclusionsOne Way FSI From this analysis, it was seen that the stresses were developed at the root of the blade which
developed as a crack of the turbine due to runner interaction with guide vane.
Improvement in the results would have been achieved if the there would have been node tonode connectivity and using same geometry (Fluid and Surface in common)
Two Way FSI Form this analysis, it was seen that the vortex shedding frequency derived from flow induced
vibrations match the resonance (bending mode) of the vane. For such lockup conditions,Karman vortices exhibit a strong instability and substantial increase in vibration level.
The Stress at the root of the blade were around 10 times less then the yield stress of thematerial. So, cycle calculation would help in determining the fatigue characteristics and numberof cycles the vane can sustain.
Coupled analysis did not show the resonant interaction between fluid and solid that wasexpected. This is due the oscillations have not reached a constant phase and recommended torun from some more time
QuEST Global Hitachi Ltd. Confidential 23
8/9/2019 FSI Paper Powergen India Asia Read-Only Compatibility Mode
24/24
THANK YOUHANK YOU