The VIVACE Converter

Enhancing Flow Induced Motions to Harness Hydrokinetic Energy

in an Environmentally Compatible Way

Michael M Bernitsas, Ph.D.

Mortimer E. Cooley Professor of NA&ME and ME
Director Marine Renewable Energy Laboratory

Marine and Hydrokinetic Device Modeling Workshop

National Wind Technology Center, Boulder CO
March 1, 2011

1/40
 Outline
I. Concept
Enhancement of flow induced motions:
VIV, galloping, buffeting

II. Development of VIVACE

Stage 1: Channel – scale 2 Scales
Stage 2: Towing tank – scale 3 Stage 3: Open-water - scale 3

III. Research Advances

Virtual m-c-k Turbulence stimulation
Galloping vs. VIV Flow transition
PTC to FIM map Multiple cylinders
CFD Fish-tail

2/40
I.1. Concept: Enhance flow induced motions

VIV (Vortex Induced Vibration)

Galloping
Soft
Hard
Wake galloping
Proximity galloping
Interference galloping

Flutter
Buffeting

3/40
 I.2. VIV
Stationary cylinder Vortex Induced Vibration
• Elastic cylindrical body
• Rigid cylinder on elastic
support

Features
• Vortex synchronization
Forced oscillation • Synchronization lock-in
at fn +/- 50%-60%
• Self - limiting amplitude
(forced oscillations)
• Initial, upper, and lower
synchronization branches
• Vortex structures
• Hysteresis
Lab picture animation by Williamson • Correlation length 4/40
 I.3. High damping VIV at 8×103<Re<1.5×105

Smooth cylinder results * A/D=1.9

Skop-Griffin Plot

Typical VIV tests are:

• Lab based with low Re

and low damping

Amax/D
• Field based with high Re
trying to suppress VIV

Few of these results are relevant VIVACE tests

to energy harness through VIV
5/40
 I.4. Oscillators: Linear and nonlinear
Linear oscillator VIVACE, VIV high-Re oscillator

VIV, low-Re oscillator VIVACE, VIV+galloping oscillator

6/40
 II.1. Development of VIVACE
• Stage 1: The concept
• Scales
• Stage 2: Proof of concept, channel tests
• Stage 3: Field tests
Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
Tidal-current Subsystem testing Subsystem testing Full scale Commercial
energy conversion at intermediate at large scale prototype testing demonstrator
concept scale testing
U of Michigan U of Michigan St. Clair River St. Clair River TBD
2005 to 2009 2009 to Present Summer 2010 Summer 2011 Summer 2011
 II.2. Stage 1: The concept
1940: Tacoma Narrows bridge collapsed 1965: Ferrybridge cooling towers
due to wind-induced vibrations collapsed due to VIV

FIM can be controlled to generate energy! 8/40

 II.3. How VIVACE works
Objective: Capture the abundant hydro-kinetic energy in even low-
speed ocean/river currents without using dams or turbines

Approach: Develop technology that mimics and enhances natural

phenomena: VIV, galloping
VIV: Enhance vortex shedding, Harness VIV energy
Galloping: Enhance instability, Harness VIV energy
Fish biomimetics: Surface roughness; cylinder proximity; passive fish tail

Concept: FIM converts hydrokinetic energy to transverse

mechanical motion.

VIV concept Flow Lines Movement Shedding

Current

Bluff Body Movement

Vortices Shedding 9/40
 II.4. Stage 2: Proof of concept lab tests

Flow
Velocity
U=
1.6knots
(0.8m/s)

Synchronization
U=[0.56-1.05]m/s
at high damping,
K=2*518 N/m,
m*=1.45 10/40
 II.5. VIVACE scalability & modularity
Small Scale
Water flow Large Scale

Object: Cross-
section of a cylinder Alternating Vortices

Lab Scale

Farm Scale
11/40
VIVACE scales

Scale 1: P≤10W 1,000≤Re≤20,000

Scale 2: P≤200W 20,000≤Re≤150,000

Scale 3: P≤5kw 150,000≤Re≤500,000

Scale 4: P≤100kw 500,000≤Re

4-cylinder VIVACE module 12/40

II.6. Enhance VIV & galloping

Fish biomimetics:
Passive turbulence
control A/D vs. U*

13/40
37/86
II.7. Stage 3: Prototype testing
Univ. of Michigan towing tank:
Sept. 2009

St. Clair river: Summer 2010

14/40
Lab tests: 1 cylinder, 1.9 knots

15/40
Lab tests: 1 cylinder with PTC, 2kn

16/40
Lab tests: 1 cylinder with PTC, 2 knots

17/40
Lab tests: 2 cylinders with PTC, 2 kn

18/40
 II.8. River deployment: 2 cylinders with PTC
The St. Clair River, Blue Water Bridge. Beta 1 Prototype at dock

Beta 1 being tested in St. Clair River Underwater view of VIVACE cylinders

19/40
River deployment: 2 cylinders with PTC
Open-water 2-cylinder testing

20/40
 III. Research: Vision and goals
research goals

G1 Function like a school of fish, i.e. a 3-D device with component

synergy stemming only from hydrodynamic interaction
G2 Operate efficiently at four scales with speeds as low as 0.5knots
G3 Be environmentally compatible.
G4 Generate electricity at a competitive cost.
21/40
Objective #1: Integrated PTO & Vmck
Virtual m-c-k model

2 3

1 LTFSW channel
22/40
 Physical & Virtual VIVACE
Schematic of physical VIVACE converter Schematic of virtual VIVACE converter

y + cbearing y = f fluid (t) + Fmotor

m
23 23/40
 Objective #2: Hydrokinetic to Mechanical
Expand the high lift regime TrSL3

Drag and lift coefficient vs. Re (Zdravkovich 1997)

Name of
Reynolds number
the Characteristic feature
lower limit range < Re < upper limit range
regime
1×l03- 2×l03 < Re < 2×l04- 4×l04 TrSL2 Formation of transition vortices in free shear layer
2×l04- 4×l04 < Re < l×l05- 2×l05 TrSL3 Fully turbulent shear layer
1×l05 - 2×l05< Re < 3.5×l05- 6×l06 TrBL

24/40
 Expand synchronization range Ur=Ucur/(fn,wD)
2.8 VIV GLP
2.6
2.4 P60 @ ±10°~
2.2 ±26°
2.0 P60 @ ±15°~
1.8 ±31°
A/D

1.6 P60 @ ±20°~

1.4 ±36°
1.2 P60 @ ±30°~
1.0
0.8 ±46°
P60 @ ±40°~
0.6
0.4 ±56°
0.2 P60 @ ±50°~
0.0 ±66°
2 3 4 5 6 7 8 9 U*
10 11 12 13 14 15 16 17 18 19

1.4

1.2

1.0
fosc/fn,w

0.8 P60 @
±10°~±26°
P60 @
0.6 ±15°~±31°
P60 @
0.4 ±20°~±36°
3 4 5 6 7 8 9 10 11 12 13 14 15 25/40
U
*
Maintain VIV in the transition region

26/40
 Objective #3: Passive turbulence control
Mechanics of PTC Major Parameters of PTC
• Trip the boundary layer. • αPTC, placement angle.
• Set the correlation length. • Area coverage.
• Introduce turbulence. • k, Roughness grit size.
• T=k+p, PTC total height.

Placement PTC
Coverage Area
angle (αPTC) (Roughness) k
(width of strip) T
Double-sided PTC p
tape

Stagnation Cylinder Surface

point Cylinder (Front) 27/40
Map of PTC to FIM (P180)

 Half inch width, P180

 6 Zones –WS1, HG1,
SG, HG2, SS, WS2

28/40
Objective #4: Enhance vorticity or instability?

29/40
 Objective #5: Improve cylinder interaction

Four in the
channel
Two in the towing tank

Two in the St. Clair River 30/40

Improve cylinder interaction (cont.)

Four cylinders in the channel

Center to Center distances:
1 to 2: 1.95 Diameters
2 to 3: 3.95 Diameters
3 to 4: 1.63 Diameters

Cylinder spacing robustness

31/40
62/86
Improve cylinder interaction (cont.)

Four cylinders in the channel 32/40

 Objective #6: Increase power density
This is a hydrodynamic design issue: complexity vs. power density
From: 51 W/m3 at 3 knots 5˝ cylinder
To: 239 W/m3 at 3 knots 5˝ cylinder
To: 341 W/m3 at 3 knots 3.5˝ cylinder
To: 2,728 W/m3 at 6 knots 3.5˝ cylinder
Diesel engines: 25,000 W/m3

Smooth cylinder
Lee (2009)

PTC

Smooth

33/40
 Benchmark: Power density
VIVACE
> 14,600
Wind

Footprint Volume = Footprint * water depth

Footprint Volume Density (kWatts/m3)

Current Velocity
= 3knots

Early New Pelamis Power Energetech Verdant Lunar MCT

VIVACE VIVACE Buoy Power 34/40
 Objective #7: Fish-tail kinematics
Passive fish tail
Tails and splitter plates

2.2 3.5" cylinder with k=413 N/m (A/D vs ReD)

2
1.8
K=413N/m
1.6
1.4 1,5D tail(-)
1.2
A/D

1
0.8
0.6
0.4
0.2
0
4.00E+00 2.00E+04 4.00E+04 6.00E+04 ReD
8.00E+04 1.00E+05 1.20E+05 1.40E+05

Powerful but not a research priority of MRELab

35/40
Objective #8: Improve research tools
Measurements:
•Channel •Towing tank •St. Clair River
To identify new phenomena and their parametric dependence
Increase test section depth from 80cm to 120+25cm
Increase A/D limit from 3 to 5.5 for D=3.5”

Flow visualization: Large FOV (von Karman-scale)

To describe new phenomena and their wake/vortex structures.
To identify source of oscillatory forces.

Flow visualization: Small FOV (Boundary layer-scale)

To understand the formation of the vortex structures and shear
layers that cause these new phenomena.

CFD simulations:
For comparison and possibly complementary data only.
36/40
Visualization: Large FOV

Single-body with broad-

wake FOV: about 6*D;
magnified on the right.

Multi-body with broad-

wake FOV: about 15*D

Wake-structure scale with 32 frames/sec 37/40

Visualization: Small FOV

Boundary layer scale with 1,000 frames/sec for PIV

38/40
 CFD simulations
Cylinder with PTC in FIM at high Reynolds numbers
2P: initial branch Re=50,000 P+S+S+P+S+S: upper branch Re=95,000

S+P+S+S+P+S: transition upper

branch to galloping Re=110,000 2P+8S: galloping Re=130,000
THANK YOU for your attention
Acknowledgements

DOE
DOD

Office of
Technology Transfer

Detroit/Wayne County
Port Authority
Marine Renewable
Energy Laboratory
University of Michigan