Experimental,Measurements,and

Experimental Measurements and Numerical Simula4ons in a 3-‐Turbine Array of 45:1 Scale DOE RM1 Turbines

Alberto Aliseda!Danny Sale!

Teymour Javaherchi!Nick Stelzenmuller!

WHAT IS TIDAL ENERGY? Background Design Experiment Single Turbine Results Array Results

LABORATORY-SCALE ROTOR GEOMETRY

I Maximize chord-based Reynoldsnumber

I Choose foil to minimize Reynoldsnumber effects

I Match performance and optimum tipspeed ratio with blade-elementmomentum design code

I Attempt to match power extractionand wake characteristics at scale, notgeometry

0.2 0.4 0.6 0.8 10.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2x 10

5

r/R Normalized spanwise coordinate

Chord

−bas

ed R

eynold

s num

ber

U=0.5 m/s

U=0.7 m/s

U=0.9 m/s

U=1.1 m/s

Transition?

Solid lines represent DOE RM 1Dotted lines represent lab−scale rotor

DOE RM 1 Turbine Scaling down process!







6 7 8 9 10 110

0.1

0.2

0.3

0.4

0.5

TSR

Eff

icie

ncy

Full−scale DOE RM 1 efficiency(predicted from BEMT)

Lab−scale DOE RM 1 experimental efficiency

Geometrically-scaled DOE RM 1 Turbine Performance!

6 7 8 9 10 110

0.1

0.2

0.3

0.4

0.5

TSR

Eff

icie

ncy

Full−scale DOE RM 1 efficiency(predicted from BEMT)

Lab−scale DOE RM 1 experimental efficiency







Performance-scaled !DOE RM 1 Turbine!

DOE RM1 @ 45:1 scale Similarity based on performance curves


LABORATORY-SCALE TURBINE45 cm diameter rotor

Experimental CondiAons Rechord ~ 105 Tip Speed Ra4o (TSR) = 4.5-‐8

Blockage Ra4o = 20%

30

Figure 4.1: Photograph of the Bamfield Marine Science Centre flume

4.1.1 Flume dimensions and specifications

The BMSC flume has a width of 2 m, a depth of up to 1 m, and 12.3 m test section

length with full optical access. The pumps that drive the flow are capable of a

volumetric flow rate of approximately 1 m3/s, which results in a freestream flow speed

of 0.5 m/s. This flow speed was judged to be too slow to produce adequate Reynolds

numbers, which are shown to be Re ⇠ 60, 000 at this flow speed in Figure 3.3. To

increase the flow speed, and thus the Reynolds number, a partition was constructed

in the flume, which can be seen in Figure 4.1. This partition halved the flume width

from 2 m to 1 m, doubled the maximum flow speed, and doubled the blockage ratio.

Three Different Array ConfiguraAons 1.  Array of two coaxial turbines.

2.  Array of three coaxial turbines.

3.  Array of three turbines with lateral offset.

Measurement LocaAons: Two Turbines Coaxially Mounted


TWO CO-AXIAL TURBINES AT VARIOUS SPACINGS

Performance for Two Coaxial Turbines: Experimental Measurements


TWO CO-AXIAL TURBINES PERFORMANCE

5.5 6 6.5 7 7.5 8 8.5 90

0.1

0.2

0.3

0.4

0.5

TSR

Eff

icie

ncy

Performance curves for two co−axially arranged turbinesat various separation distances

Upstream turbine

5D downstream turbine




Performance of Two Coaxial Turbines: Comparison of Experiments and SimulaAons


TURBINE ARRAYS: THREE CO-AXIAL TURBINES

Measurement LocaAons: Three Turbines Mounted Coaxially

3-‐Turbine Coaxial Array Performance Comparison of Experiments and SimulaAons

Downstream separaAon 5D

EvoluAon of the available KineAc Energy Flux a 3-‐Turbine Coaxial Array


Evolution of TKE contours in a 3-Turbine Coaxial Array!


3-‐Turbine Offset Array Performance Comparison of Experiments and SimulaAons

3-‐Turbine Offset Array Performance LES SimulaAons

InvesAgaAon of the Flume Blockage Effect

ε = 20 % ε = 10 % ε = 5 % *

•  Increase in blockage leads into increase in efficiency (verAcal shi[ of Cp curve).

•  Increase in blockage shi[s the peak of efficiency toward higher TSR values

(horizontal shi[ of Cp curve).

Summary and Conclusions!

•  Three Turbines Array present non-monotonic performance: third turbine has higher efficiency than middle turbine!

•  Confinement plays a increasingly important role for higher number of turbines and lateral offset in the Array.!

•  Agreement between experimental and numerical results is best for single turbine and optimum TSR. !

•  Angular velocity fluctuations in the experiments, and enhanced wake recovery, not captured by simulations, leads to numerical/experimental divergence with lower TSRs, larger arrays and higher confinement.!

RMS of Normalized Rota4onal Velocity Temporal Evolu4on (TSR = 6.15, 7.16)

Array of Three Turbines with Lateral Offset

•  ObservaAon of similar physics compared to results from array of two & three coaxial turbines.

•  Downstream turbines’ efficiency increase monotonically with the TSR value.

3-‐Turbine 1/4D Lateral Offset Array Performance

Reynolds-‐number Dependent Performance

5 6 7 8 9 10 11 120.25

0.3

0.35

0.4

0.45

0.5

TSR

Eff

icie

ncy

Single turbine efficiency at various flow speeds, averaged over one minute

0.52 m/s flowspeed0.61 m/s flowspeed0.65 m/s flowspeed0.71 m/s flowspeed0.75 m/s flowspeed0.90 m/s flowspeed

Turbine Comparison for Performance

5 6 7 8 9 10 11 12 130

0.1

0.2

0.3

0.4

0.5

TSR

Eff

icie

ncy

Turbine 1Turbine 2Turbine 3

PIV Velocity Profiles in the Wake turbine

0.4 0.5 0.6 0.7 0.8 0.9 1 1.10

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

Norm

aliz

ed v

erti

cal

dis

tance

fro

m c

ente

rlin

e

Normalized streamwise velocity

Rotor tip

Free surface

Streamwise velocity profiles for TSR 7

2D up

2D down

3D down

5D down

7D down

Results of BEM study (3/3) Effect of blockage on efficiency

•  Blockage increases the efficiency. •  Blockage shi[s the peak of efficiency to the right. •  Blockage delays the peak of efficiency of the two downstream turbines.

Source : S. J. Miley’s catalog of airfoils

€

α(r)= arctanVinc(r )rω

+ β(r)

Angle of adack :

3-‐Turbine 1/4D Lateral Offset Array Performance


Documents

Experimental,Measurements,and