Tong W. Wind Power Generation and Wind Turbine Design

Подождите немного. Документ загружается.

Intelligent Wind Power Unit with Tandem Wind Rotors 345

that both torques show the same absolute value irrespective of the relative rota-

tional speed and the output/load, because the rotational torque should be always

dynamically counter-balanced between the inner and the outer armatures.

Figure 19 shows the rotational torques T

and T

at the various relative speeds

against the induced electric current I

, where these torques include the mechan-

ical torques in the generator and T

is given by the absolute value. That is, these

are the shaft torques in practical use. The rotational torques of both shafts T

, T

are directly proportional to the induced electric current I

irrespective of the rota-

tional speeds and/or the external load but the shaft torque T

of the outer armature

is slightly larger than T

of the inner armature due to the mechanical torque of the

larger bearings.

The output P and the induced voltage E against the induced electric current I

are determined by the relative rotational speed N

(= N

− N

), as shown in Fig. 20 .

The output increases with the increase of the induced voltage E at the same I

while E is proportional to the relative rotational speed N

and I

determines the

rotational torque as presented in Fig. 19 .

5 Demonstration of intelligent wind power unit

5.1 Preparation of the tentative tandem wind rotors

5.1.1 Front blade design

Figure 21 shows the drag and the lift coefﬁ cients C

, C

against the angle of attack

a for MEL012 aerofoil [ 10 ]. The angle giving the maximum C

is a = 11°, but the

angle a = 8° was chosen as the design dimension irrespective of the radial position,

so as to avoid the ﬂ ow stall on a large scale at the lower rotational speed. The chord

of the blade was designed taking the traditional blade element theory [ 11 ] into

02468101214

, -T

N.m

min

-1

min

-1

min

-1

300 150 -150

900 150 -750

900 750 -150

1500 750 -750

-T

Figure 19: The output and the induced voltage against the induced electric current.

346 Wind Power Generation and Wind Turbine Design

-0.5

0.5

1.5

-5 0 5 10 15 20

, C

deg.

Figure 21 : Drag and lift coefﬁ cient against the attack angle.

0.2

0.4

0.6

0.8

100

0246810

P kW

E V

-150

-350

-550

-750

= 50 min

-1

P E

Figure 20: The output and the induced voltage against the induced electric current.

account. The ﬂ ow through the wind rotor is contributed by the axial and the tan-

gential induction factors (wind velocity deceleration ratio), and the relative ﬂ ow

angle at the inlet of the front wind rotor

measured from the tangential direction

and the chord C at the radius r are given as follows [ 12 ], where l

is the speed ratio

at the radius r [=(rotational speed at r )/wind velocity].

tan (1/ ),KKl

−

==f

(1)

[]

Fin

1cos( )

=−f

(2 )

Figure 22 shows the blade proﬁ le with MEL012 aerofoil elements [ 10 ] which is

Intelligent Wind Power Unit with Tandem Wind Rotors 347

5.1.2 Rear blade design

The counter-rotational torque of the rear wind rotor must coincide with the rotational

torque of the front wind rotor designed above, but the convenient design procedures

for the rear blade interacting with the front wind rotor is not yet known. Accord-

ingly, the ﬂ ow around the front wind rotor was simulated by the commercial CFD

code CFX-10 with k – ε turbulent model, to get the inlet ﬂ ow condition of the rear

wind rotor. In the simulation, the circularly cylindrical ﬁ eld with 20 d

radius ( d

the diameter of the front wind rotor) was set, and was divided into two domains.

One is the rotational domain surrounding the wind rotor with 80,668 tetra-type ele-

ments (mesh size is 0.5 mm in close to the blade surface) and 6294 prism-type

elements in the radial direction with 2 d

, while 1 d

upstream and 3 d

downstream

of the rotor. The other is the stationary domain with 799,060 prism-type elements

surrounding the rotational domain, and both domains were joined with the Frozen-

Rotor interface keeping the relative blade position of the tandem wind rotors. The

ﬂ ow around the tandem wind rotors with two-dimensional blades was simulated

in a trial by the above code and is shown in Fig. 23 accompanied with the plotted

experimental results, where V

Mtm

and V

θ tm

are the axial and the swirling velocity

components divided by the wind velocity V at the wind tunnel outlet, R is the dimen-

sionless radius divided by d

/2, Z

and Z

are the blade numbers of the front and the

rear wind rotors, and the measurement sections M1–M3 and the blade proﬁ le are

also given in the ﬁ gure. It was conﬁ rmed that the above code can predict roughly the

actual axial velocity, and the ﬂ ow conditions are discussed in [ 13 ].

The ﬂ ow around the front wind rotor equipped with three blades shown in Fig. 22

was simulated by the above code. Figure 24 shows the relative ﬂ ow angles measured

from the tangential direction, where

Fin

and

Fout

are the simulated relative ﬂ ow

angles averaged in the tangential directions at the inlet and the outlet sections of the

Front blade Rear blade

Figure 22 : Blade proﬁ les of the tandem wind rotors.

348 Wind Power Generation and Wind Turbine Design

0 0.2 0.4 0.6 0.8 1

Fin

, f

Fout

Fin

Fout

= 3

by eq. (1)

V = 8 m/s

= 400 min

-1

Figure 24 : Relative ﬂ ow angles measured from the tangential direction.

Figure 23: Flow around the tandem wind rotors composed of two-dimensional

blades.

-0.4

-0.2

0.0

0.2

0.4

0 0.2 0.4 0.6 0.8 1 1.2

Tip of

rear rotor

Tip of

front rotor

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Mtm

=850 [min

-1

] V=11 [m/s]

=-640 [min

-1

] Z

=3 Z

Intelligent Wind Power Unit with Tandem Wind Rotors 349

front wind rotor,

is the outlet ﬂ ow angle determined by eqn (1) at each radial

position of the front blade. The angle

is close to

Fout

, that is,

may give the

outlet ﬂ ow angle. Accordingly, the rotational torque of the rear wind rotor, namely

the angular momentum change, was determined with the simulated outlet ﬂ ow angle

Fout

of the front wind rotor, which is corresponding to the inlet ﬂ ow angle of the rear

wind rotor, and the outlet ﬂ ow angle of the rear wind rotor predicted by eqn (1).

Figure 22 also shows the designed rear blade with MEL012 aerofoil elements

[ 10 ], where the chord is also derived from eqn (2) but the height was determined

tentatively.

5.2 Preparation of the model unit and operations on the vehicle

The synchronous generator ( Fig. 16 ) equipped with the tentative tandem wind

rotors ( Fig. 22 ) were mounted on the circularly cylindrical tower as the down-

wind type, and boarded on the pick-up type truck as shown in Fig. 25 . The wind

velocity can be controlled by the driving speed of the truck, and also measured

by the cup-type anemometer. The rotational speeds of the wind rotors, the output,

the induced voltage and the induced electric current were automatically accumu-

lated and stored in the personal computer every one second, in accompanying

with the data of the wind velocity. The output was consumed by many bulbs, but

these resistances depend mainly on the induced voltage, namely the rotational

speed, and the induced electric current affecting the temperature of the ﬁ lament.

Therefore, the resistance was replaced by the speciﬁ ed power of the bulb at 100 V,

P , called “bulb load” as the indication for the external load.

The driving site is at the seashore of Wakamatsu, Kitakyushu, Japan, where there are

two straight roads of about 1.5 km in the north–south and the east–west directions. The

wind is calm and its direction is nearly constant, that is, the site is suitable for making

the experiments using the truck. The truck is used to be driven in three modes, which

are the acceleration, steady state, then the deceleration. Figure 26 shows one of the

rotational speeds of the front and the rear wind rotors N

, N

, and the wind velocity V

Anemometer

Power mete

Bulb load

Tachometer

Figure 25: Field experiment on the pick-up type truck.

350 Wind Power Generation and Wind Turbine Design

accumulated from start to stop of the truck. The data corresponding to the

acceleration and deceleration of the truck were excluded for the evaluations of the

experimental data, in order to get the results at the steady-state ﬂ ow conditions.

5.3 Performances of the tandem wind rotors

The effects of the rear blade setting angle b

on the performances are shown

in Fig. 27 , in keeping the front blade setting angle b

= 20° and the bulb load

7.5

22.5

-600

-300

300

600

020406080100

V m/s

, N

min

-1

Measurement time Sec

Acceleration

DecelerationSteady State

Figure 26 : Accumulated rotational speeds while driving the truck.

-1000

-500

500

1000

, N

min

-1

deg. 10 20 30

, P

300

600

900

048121620

P W

V m/s

= 20deg.

bulb

= 900W

Figure 27: Effect of the blade setting on the performance.

Intelligent Wind Power Unit with Tandem Wind Rotors 351

bulb

= 900 W, where the blade settings are deﬁ ned as shown in Fig. 28 at the

blade tips. The rear wind rotor never counter-rotates against the front wind rotor

at the lower wind velocity V because not only the rear wind rotor cannot generate

the sufﬁ ciently counter-rotational torque T

corresponding to T

of the front wind

rotor but also the outer armature pulls the inner armature by the magnetic force.

With the increase of the wind velocity, the rear wind rotor with b

larger than 20°

begins to counter-rotate successfully. On the contrary, the rear wind rotor with b

10° never change the rotational direction and is in the blowing mode because the

rear blade with the excessively larger angle of attack cannot generate the fruitful

counter-rotational torque due to the ﬂ ow stall on larger scale. Judging from the

counter-rotation expected to this type unit, the rear wind rotor with b

= 20° may

be acceptable within the measured data.

Figure 29 shows the effect of the bulb load on the operating conditions at the blade

setting angles b

= b

= 20°. The load P

bulb

affects obviously the rotational speed of

Wind direction

Front Rotor

Rear Rotor

Figure 28: Blade setting angles.

-900

-600

-300

300

600

0 5 10 15 20

, N

min

-1

V m/s

=20deg.

423

1023

bulb

150

300

450

600

750

0 5 10 15 20

423

1023

P W

V m/s

bulb

423

1023

E V

=20deg.

bulb

423

1023

bulb

Figure 29 : Effect of the bulb load on the operation conditions.

352 Wind Power Generation and Wind Turbine Design

the rear wind rotor N

in the lower wind velocity than 12 m/s, but scarcely affects the

speed of the front wind rotor N

regardless of the wind velocity V . Besides, the rear

wind rotor starts to counter-rotate at the slower wind velocity with the decrease of

the bulb load. The relative rotational speed N

= N

− N

contributes proportionally

to the induced voltage E , the induced electric current I

depends on the load, and the

output P is obtained electrically from √ 3 EI , while given mechanically by Tw

( T =

= − T

, w

: the relative angular speed). Within the measured data, the operating

condition with the bulb load 1023 W gives the higher output P .

5.4 Trial of the reasonable operation

As recognized in the above discussions, the performances are affected not only by

the wind velocity V but also by the blade setting angles b

, b

and the bulb load.

Then, this wind power unit was operated impractically but reasonably, in trail, by

changing the rear blade setting angle b

and the bulb load, while keeping the front

blade setting angle b

= 20°. The output was kept constant at P = 430 W in the

wind velocity higher than V = 12.6 m/s, and the unit was operated at the optimum

rotational speed giving the maximum output in the slower wind velocity. Figure 30

shows the reasonable operation, where the adjusted rear blade setting angle and

bulb load are denoted with ( b

(°), P

bulb

(W)), and the experimental data are plot-

ted and represented by the curve. The bulb load P

bulb

is not related directly to the

output P because the bulb resistances are affected by the ﬁ lament temperature

150

300

450

600

P W

(30,600)

(30,723)

(20,1023)

(30,1023)

(20,725)

deg., P

bulb

=20deg.

P=430 W

(30,900)

V=12.6 m/s

(20,900)

-800

-400

400

800

048121620

, N

min

-1

V m/s

Figure 30 : Trial of the reasonable operations.

Intelligent Wind Power Unit with Tandem Wind Rotors 353

as mentioned before. This type unit may be successfully operated by adjusting

suitably the blade setting angle and the external load. Such operations, however,

are not expected and not acceptable for the proposed wind power unit because the

rotational speeds must be adjusted automatically in response to the wind velocity.

The ﬁ nal target of the serial researches is to get smartly the above operation by the

superior works of the tandem wind rotors in cooperation with the double arma-

tures of the generator presented before. Consequently, it is necessary to improve

more and more the blade proﬁ les suitable for the front and the rear wind rotors.

6 Optimizing the proﬁ les of tandem wind rotors

The tandem wind rotors have been investigated as reviewed before. In the proposed

tandem wind rotorts, however, the rear wind rotors must counter-rotate against the

front wind rotor, stop, and rotate in the same direction of the front wind rotor, with

the increase of the wind velocity. This chapter is the status report for optimizing

the wind rotor proﬁ les. The authors pay attention, here, not only to the wind rotor

proﬁ les but also to the arrangements between both wind rotors.

6.1 Experiments in the wind tunnel

The diameter of the front wind rotor was kept constant at d

= 500 mm, and the

diameter of the rear wind rotor was changed from d

= 260 to 560 mm, for the

experiments in the wind tunnel. The axial distance between the front and the rear

wind rotors (the twist centres of the blade) was also changed from l = 40 to 200 mm.

The blade proﬁ les of the front and the rear wind rotors are shown in Fig. 31 .

The blade E is supplied to the front and/or the rear wind rotors, and has not the

camber and the twist. The front blade G is formed with MEL002 aerofoil elements

[ 10 ] and has the twist to get the desirable angle of attack a = 7°, as the single wind

Rear blade G

Unit: mm

Front blade G

Front and/or Rear

blade E

Figure 31: Blade proﬁ les of model tandem wind rotors.

354 Wind Power Generation and Wind Turbine Design

rotor with the blade tip speed ratio l = 6 [=(blade tip speed)/(wind velocity)]. The

rear blade G is also formed with MEL002 and is twisted taking account of the ﬂ ow

behind the front wind rotor. The front and the rear blade numbers are Z

= 3, Z

= 5

which are optimized in the previous paper [ 13 ]. Hereafter, the tandem wind rotors

are called “Tandem wind Rotor EE” which is composed of the front blade E and

the rear blade E, where the front and the rear blade setting angles are b

= b

= 11°

(see Fig. 28 ). “Tandem Wind Rotor GE” is composed of the front blade G and the

rear blade E, while b

= 10°, b

= 16°. “Tandem Wind Rotor GG” is composed of

the front blade G and the rear blade G, while b

= 5°, b

= 15°. “Single Wind Rotor

G” consists of with the front blade G, while b

= 0°. These blade setting angles

were optimized at the preliminary experiments [ 14 , 15 ].

The model wind rotors shown in Fig. 32 were set perpendicular to the wind

direction, at 185 mm downstream of the wind tunnel outlet with the nozzle diam-

eter of 800 mm. The front and the rear wind rotors were connected directly and

respectively to the isolated motor with the inverter, in place of the peculiar generator.

The rotational torques of the front and the rear wind rotors were counter-balanced

by the rotational speed control. The performances of the model wind rotor in the

following discussions are evaluated without the mechanical torques of the bear-

ings and the pulley system. The ﬂ ow conditions discharged from the wind tunnel

outlet, namely the inlet ﬂ ow conditions of the wind rotors, are almost uniform and

axis-symmetry without the swirl.

The outputs of the model wind rotors may be smaller than those of the proto-

types, as the Reynolds number estimated at the front blade tip is (0.8–2.2) × 10

the experiments.

6.2 Optimum diameter ratio of front and rear wind rotors

To know the optimum diameter ratio D

[= d

/ d

], the performances of the tan-

dem wind rotors were investigated by changing the diameter of the rear wind rotor,

and are shown in Fig. 33 . The output coefﬁ cient is deﬁ ned by C

= P /( r AV

/2) and

(C)

(B)

(A)

(A): Torque meter

(B): Tachometer

(C): Motor

= 500 mm

Figure 32: Model tandem wind rotors for performance experiment.