The Fluctuating Reduction Method for Generation Power of Wind-PV Hybrid System

Copyright © IFAC Power Plants and Power Systems Control. Seoul, Korea, 2003

ELSEVIER

IFAC PUBLICATIONS www.elsevier.com/locale/ifac

THE FLUCTUATING REDUCTION METHOD FOR GENERATION POWER OF WIND-PV HYBRID SYSTEM

Jin-Seok Oh*, Sang-Tae Lee*, Seng-Hwan Jun*, Byeong-Duck Ye*, Ji-Yeong Lee**

* Korea Maritime University ** Graduate school ofKorea Maritime University

Abstract: This paper reports the performance of a CB(Circuit Breaker) and converter for Wind-PV (Photovoltaic) system with battery. For this purpose, a fluctuating reduction controller for an electric generation hybrid(wind+PV) system is suggested. The system operates a wind turbine, PV, CB, converter and a battery. Integration of wind and PV sources, which are generally complementary, usually reduce the capacity of the battery. Also, CB controls the overvoltage of the generation system. The objective is to control the operation of the converter and the CB for reducing the power fluctuation. This paper includes discussion on system performance, power quality, fluctuation and effect of the randomness of the wind. Copyright © 2003 IFAC Keywords: Circuit Breaker, Converter, Hybrid, Fluctuation, Potovoltaic, Wind.

1. INTRODUCTION

battery while the generator supplies power to the converter. This paper describes the CB control algorithm for overvoltage control and AC-DC converter algorithm for boosting the generation power as booster for generation power applications. The algorithm is developed from an understanding of the energy flow within the system. The purposed scheme can also function as a battery charger and a power supply without any additional power circuit components. The proposed AC-DC converter with the network is addressed using case studies of voltage fluctuation and harmonics elimination. In this paper, the impacts on voltage fluctuation can be minimized and furthermore, the network voltage control could also be improved by the proposed operating algorithm.

Wind energy conversion system is based upon the well-known DC link Converter. The wind generator provides current at a variable frequency. This current is rectified onto the DC link using a converter with six diodes. Generally, the wind turbine generation system has to use a AC-DC converters for an uninterruptible power supply. Fig. 1 shows the schematic diagram of the proposed system for the wind turbine generation which consist of a converter, CB and battery. The proposed power conversion scheme has a variety pattern of applications in the hybrid generation systems. Wind Generator

~onverte:Battery I Load it i L : SW 1 1 r--~-HrfT1f'-1+++--r---<:t"oi-- -I I

Rectifier

0,

0,

o,!

i

i

i I

04

05

I

!

To AC Bus

6! El

is

if)

I

!

I

I

I

I

I

I

,

I

!Cs ,

2. BASIC STRUCTURE AND OPERATING PRINCIPLE

i

i

i

I

l

The wind generator provides voltage and current at a frequency directly correlates to the operating speed of the wind turbine. A converter, typically containing six devices, rectifies voltage and current onto the DC link. A capacitor bank provides filtering, resulting in a voltage stiff DC link characteristic. The proposed power conversion system converts energy from a variable frequency wind-power to a fixed frequency output as shown in Fig. I. The system consists of a wind generator, rectifier, converter and CB. These elements discussed below.

Ballery!

i

I

:

j

!

I I

I

i

I I

I

DC

Fig. I. Schematic diagram of the proposed converter system. It is also function as an AC to DC converter thus transfelTing power from the hybrid generator to the 127

2.1 Wind generator

Fig. 2. As the wind speed fluctuates about some mean value the power will fluctuate also so that the operating point moves, at varying rates, backwards and forwards along the curve_ When the mean wind speed is near either cut-in speed or the rated wind speed, the true operating point will lie off the curve. Near cut-in it will lie above the curve and near the rated wind speed it will lie below. The power curve in actual operation will thus be a distortion of the theoretical curve. The amount of the distortion depends upon the amplitude of the wind speed fluctuations. The actual power curve for wind speeds above the rated value may thus lie between the limits indicated by the P-curve and Q-curve in Fig. 2. The wind speed and power curves are shown in Fig. 3. Fig. 3 shows that the generating electric power is according to the wind speed. It is also noted that the varying wind power can result in bus voltage fluctuation.

The wind generator is linked to the battery through CB and AC-DC converter. Fig.1 presents wind generating power load to the bus or AC-DC converter according to CB condition. The power captured by the wind turbine may be written as the equation 1. (1)

where, p is air density (kg/m'), A is swept area(m'),

Cp is coefficient of wind turbine, V is wind velocity(nVs). The torque produced by the wind generator can be written as follows: Twe=3EwscI>wm

1_(Wwe(lim»2

20JwmLws

(2)

Wwe

where, P is number of poles, OJwe is (p12) . OJwm

,

OJwm is angular shaft speed, <1> wm is flux linked by the stator windings. It is interesting to note that for a given constant voltage E ws , whose value can be obtained through the equation (3) in the terminals of wind generator.

16 ~

/\

14

(/

E ~ 12

g

10

'"c

8

.~

(3)

:rJ'J

I' .'"

f\ \ 'I

~

~

V

The rotor speed can be described in the following equation:

20

"'"

/

~

['I

'\

/ AI'!

\I 60

40

80

100

120

140

160

tlme(seconds)

dOJwm = '!-J(T J wo dt

- T,w'- )dt

(a) Wind speed

(4) 60

is aerodynamic torque

50

captured from the wind and TwL is the electric torque load. The rotor speed acceleration and deceleration can be described as equation (4). The rate of rotor speed is proportional to the inverse of the inertia and difference between Two from T wL . The aerodynamic torque is affected by the operating Cp. The voltage fluctuation problem is closer to a steady state problem such as load varying, which is well defined by the real and reactive power distribution.

~ 40

Where, J is the inertia,

Two

Q;

~ 30

/\

'\)

20

I

~

~/

10

o

o

20

40

60

80

100

120

140

160

time(seconds)

(b) Output power Fig. 3 Wind speed (a) and Output power(b)

2.2 The PV modules

------------p

Rated wind speed

/\

c.

"~~-------------.Q

Cut-in speed

/\

/'\ / \

The PV module manufacturer rating is 6OW. The PV module power can be calculated by equation (5)

Wind speed_

Where I ph is the cell photo current, Vp is the PV output voltage, 10 is the saturation current, q is the charge of electron, K is Boltzman's constant and T is the cell temperature. The relation between Vp and

Fig. 2. Relationship between power and wind speed To understand the effect of fluctuating winds more precisely, consider the theoretical (power/wind speed) curves, for steady winds, given in full line in

derivative ( dP / dVp ) of output power wit output voltage can be expressed as

128

V

i] = ~[I- exp(-t IT)] + 1] exp( -t / r) R

Fig. 4 shows the characteristic curves of equations (5) and (6), and their relationship.

iz

=

vs -vB [1-exp(-t/r)]+1 R

(7)

z exp(-tIT)(8)

.----------.---------,80

Where T = L / R is the time constant, R is the resistance, Vs is the generating voltage and VB is the battery voltage. Equation (7) and (8) are valid for T = ton + t a for

70

:?

6 r----------.---7"'~_..

60

3

a:

50 40

continuous current and ton + t a < T for discontinuous current. T is the switching period and t d is the time at which the current becomes zero. From equation (8), ta is as follow;

30

\

20 10

1

2

4

10

8

i

12

'3

14

16

18

V,J.Vl

(9)

Fig. 4. Characteristic of PV module It is shown that the MPP(Maximum Power Point) must correspond to such a point where the dP / dVp

Where A = Vs / VB . The relation of ta IT. can be shown Fig. 7.

is equal to zero. The PV module output voltage can be adjusted by this concept. In this paper, the MPPT is designed by MPC algorithm with dP / dVp = O.

A = 0.9

1.0

2.3 Converter

A = 0.5

0.1

The converter regulates the DC output of wind generator-rectifier unit and PV by switching ratio controller as shown in Fig. 5.

0.01 0.00] IL-_ _---+-

0.01

Source

I

i

PV output

~

L

11=1

MPPT

I

~B_BLV_BI- ~i~J

j hi

Fig. 7. Relationship between ton and r From Fig. 7, these two cases ton / r((l and ton / .. »1 should be different as equation (10) and equation (11).

-,--,',,-IP_ft---L_S_W_D_ _

J

td::I~Aton Fig. 5. Block diagram of the generator and converter

I

t a =r·ln-I-A

We will suppose that the system switches from one state to another. Its equivalent circuits for on-state and off-state are given in Fig. 6. L

"/

~

..

L

R

R

r

(tonlT((I) (tonlT»l)

(10)

(ll)

The current iL for continuous and discontinuous condition is given i] and i z , the system circuit of Fig. 6 can be modeled by the following set of linear differential equations: L

~J

~~

= Vs - RiL

. di L L-=V -V -RI dt s B L

-------

(a) on-state

_

1.0

To load

L

.

;--_ _-+

0.1

(on state)

(12)

(off state)

(13)

This converter has a function of DC-DC boost switching regulator. The converter switching strategy studied in this paper is a PWM algorithm. The PWM waveform for switching is generated by a modulating wave and a carrier wave. The generation of modulating signal VM is generated by a difference voltage (Vs - VB )

(b) off-state

Fig. 6 Equivalent circuits for the two states In Fig. 6, the currents i] and i z are given by

129

for the switching condition. The carrier signal Vs is the T-periodic signal. The switching wavefonns of converter are shown in Fig. 8. OFF

ON

equation h - (i L - i B)} x 100 /i L . Wind-PV hybrid system fonns a complementary system and this complementary feature is favorable to system reliability. The switching action of converter is according to the generation power as shown in Fig. 9 (d). The charge efficiency can be increased by converter and the charging power fluctuation has been greatly reduced. From Fig. 9 (e), the battery voltage keeps a constant voltage with CB and converter. The overvoltage was controlled by using CB for the control output of wind system. Also, the DC link fluctuation can be reduced by switching strategy of converter. 16

r-----------------, (Volt)

Fig. 8. Switching wavefonns of converter The converter switches from one state to another, whenever the generating voltage Vs is boosted under

-===\--------:~-':"..-----~i------t------

12

1 ~

10

1

\~\7

8

the control condition VBM ::; Vs ::; VB. Where VBM is a minimum battery voltage for switching condition. The converter has the three operating condition such as OFF, PWM and ON configuration. As can be seen in Fig. 8, the operating condition of ON and OFF is fixed by Vs and VB .

t==

60

5

180

120

time(sec)

(a)

The switching condition is Vc - VM = 0 . The converter switches from on( VM )VC ) and

8 with MPPT

off( VM (Vc ) in the PWM area. When the inductor current becomes zero during the off-state, the time is

6

( (

~

>4 a..

diL = 0). This converter algorithm includes the dt PWM strategy and the condition of td .

td

L-----------"--+----115

t

(

-

/-.-'

"

, ,--"'-, PVoutput

2

0

2.4 CB operation

60

The CB close signal for connecting wind generator to bus is generated by the synchronizing condition such as voltage, frequency and phase. For this purpose, the wind generator voltage is maintained within ±3% of bus Voltage. The output voltage would be 12Volt with llmlsec wind speed. The CB control signal can be generated between lOmlsec to 14m1sec. The synchronizing is perfonned with the wind generator frequency kept at a high level compared with the bus frequency. In this paper, the CB signal for close is produced with the above condition.

120

180 time(sec)

(b) 125

t;

C 100

u'"

"t)

E

c:

'"

~

/'

"---../

----------

~

'--...........-

f:; 75

'" CD

"C

'"

.Cl

-.:

~ ~ 50 u '0 25

o 60

120

180 time(sec)

(c)

3. Experimental Results

From Fig. 9 (a), (b) and (c), the time evolution of the environmental condition can be observed. Fig. 9 (a) shows the wind speed signal varying between 6m1sec and 14m1sec, with a mean of lO.5m1sec. This wavefonn is presented the turbulent part of the wind and its variation. The control signal of CB is generated above llmlsec. Fig. 9 (b) displays the current supply in the converter by the PV system. It can be seen that the PV power is a more static source than wind. Fig. 9 (c) shows the charge efficiency of wind generator and PV that it is estimated with the

~

Q;

> c: 0 u

"0 c

.2 <3

"c

Cl

:c

-"

.~

(f)

60

120

180 time(sec)

(d) 130

system. Energy Conversion, Vol. 16, pp. 134139. Po S. Dokopoulos, A. C. Saramourtsis, A. G. Bakirtzis (1996). Prediction and evaluation of the performance of wind-diesel energy systems. Energy Conversion, Vol. 11, pp. 385-393. Z. Chen and E. Spooner(2001). Grid power quality with variable speed wind turbines. Energy Conversion, Vol. 16, pp. 148-153.

125 :s! .0 g100 ~ "" ~ '0 >

""
C

'"

~

Q;

-

~

'" 75 c 0


~

~

50

E Q)

'"Cl ;;; en 25 ~

U

""

0 120

60

180 lime(sec)

(e) Fig. 9. Experimental results 4. Conclusion The wind-PV hybrid system forms a complementary system. Generally, the solar is a more static source than wind, but the wind provides energy during periods of no sunshine. This hybrid system has a high efficiency and reliability. This paper describes an power control system with CB and converter. The overvoltage of wind system is reduced by CB and DC link fluctuation is controlled by converter. So, the DC bus fluctuation can be reduced with CB and converter. Also, the high efficiency power is achieved by the PWM strategy for converter.

REFERENCES Abdelali El Aroudi and Ramon Leyva (2001). Quasi-periodic route to Chaos in a PWM voltage -controlled DC-DC boost converter. Circutts and systems- 1, VoL 48, pp. 967-977. Edurd Muljadi, Hetbert L. Hess and Kim Thomas (2001). Zero sequence method for energy recovery form a variable-speed wind turbine generator. Energy Conversion, Vol. 16, pp. 99103. Francois Giraud and Zyiad M. Salameh (2001). Steady-state performance of a grid-connected rooftop hybrid wind-photovoltaic power system with battery storage. Energy Conversion, Vol. 16, pp. 1-7. F. Valenciaga, P. F. puleston, P. E. Battaiotto and R. 1. Mantz (2001). Passivity/sliding mode control of a stand-alone hybrid generation system. Proc.-Control Theory, Vol. 147, pp. 680-685. Gyu Bum Joung and Jae-Dong Choi, Design of Solar Array Simulator for Spacecrsft, KIEE International Transactions on Electrical Machinery and Energy Conversion Systems, Vot. 12, No. 2, pp. 52-56, June, 2002 Malinowski, Marian Po Kazmierkowski, Steffan Hansen, Frede Blaabjerg, G.D. Marques(2oo1). Virtual-flux-based direct power control of threephase PWM rectifiers. Industry Applications, Vol. 37, pp. 1010-1027. Peng Wang and Roy Billinton (2001). Reliability beneft analysis of adding WTG to a distribution

131