scheme of grid-connected wind power generation system using cage-type induction generators, which is based on a fuzzy logic control. The induction generator ...
2004 35th Annual IEEE Power Electronics Specialists Conference
Aachen, Germany, ZOOQ
Variable Speed Wind Power Generation System Based on Fuzzy Logic Control for Maximum Output Power Tracking Ahmed G Abo-Khalil
Dong-Choon Lee* Jul-Ki Seok Yeungnam University 2 14-I , Daedong, Gyungsan, Gyeongbuk, 7 12-749, Korea Email: dclee(ii,w.ac.kr
Abstract . This paper proposes a variable speed control scheme of grid-connected wind power generation system using cage-type induction generators, which is based on a fuzzy logic control. The induction generator is operated in indirect vector control mode, where the d-axis current controls the excitation level and the q-axis current controls the generator torque, by which the speed of the induction generator is controlled according to the variation of the wind speed in order to produce the maximum output power. The generated power flows into the utility grid through the hacktu-hack PWM converter. The grid-side converter controls the de link voltage and the line-side power factor by the q-axis and the d-axis current control, respectively. Experimental results are shown to verify the validity of the proposed scheme.
approach leading to an independent control of active and reactive power flow into the grid. The power is therefore injected into the grid with low distortion currents and at unity power factor. Figure I show the schematic diagram of the system which is studied in this paper. The generator is a squirrel-cage induction machine whose stator is connected to the grid side through the back-to-back PWM converter
covener1
covener2
I. INTRODUCTION Fig. I Schematicdiagram ofthe overall system
Recently, interests of renewable energy such as wind and solar energy, fuel cell, tidal power, and geothermal power have been attracted since the fossil energy will be exhausted and its use causes the environmental issue. Among them, wind power generation is more economic and so has been developed at commercial level. The doubly-fed type induction generator is being popularly used for wind power generation operating at constant rotor speed since the rotor circuit can be controlled in order to give the constant output voltage of the stator. However, this system has a disadvantage of a slip ring and the operating range is rather limited for a constant speed operation [I], [2]. On the other hand, the PM(permanent magnet) synchronous generator is presently too expensive for the high power level even though it has been already designed for commercial purpose. However, it is preferable to the low power generation system since no external excitation is required. Squirrel cage induction motors are the most commonly used electrical machine in AC drives, because they are robust, cheap and have low maintenance cost. These advantages make the induction machine very attractive for wind power applications both for fixed and variable speed operation. To take advantage of higher energy capture and increase in the system compliance resulting from variable speed operation a power electronics interface must be provided between the machine terminals and the grid. The back-to-back PWM converter-based power electronic interface is suitable option for the cage induction machine in wind power applications [3]. Vector control techniques are used to control the machine torque and flux separately, which gives a high dynamic performance. The grid-side converter currents are also controlled using a vector control
0-7803-8399-0/04/$20.00 02004 IEEE.
2039
Due to the nonlinear characteristics of wind turbine, there is a particular operating point at which the output power of the system is maximum for a given wind speed. This has led to the development of control techniques, to control the power conditions, in such a way to extract the maximum available wind energy at any given time. There were several papers dealing with the maximum output power control using the machine parameters and the wind speed measurement [4]. It can be called a model-based method [SI-[6].Instead, we can use a search-based method to fmd the generator speed which gives the maximum output power by perturbing the generator speed reference according to the wind speed [7]-[I I]. In this paper, an on-line maximum power point tracking control based on a fuzzy logic control is proposed for the indirect vector-controlled induction generator system. The fuzzy logic controller tracks the maximum power point and extracts the maximum output power from the wind generator system under varying wind conditions. Two realtime measurements 6P, and 6 q ' are used as the control input signal and the output of the controller is the new speed reference variation n6w,'. The proposed algorithm is experimentally verified for a small-scale laboratory setup.
U. TURBINE CHARACTERISTICS Wind turbine simulator is needed as a prime-mover to drive the squirrel cage induction generator, which can be performed by a dc motor torque control. Here, the wind turbine blade characteristics are described briefly. The kinetic energy available from the wind is transferred as shaft torque input to the induction generator, which in
Aachen, Germany, 2004
2004 35lh Annual IEEE Power Electronics Specklisrs Conference
turn converts it to electrical energy [2]. The theoretical power which can be obtained from the wind passing through a circular area is given by
ci,=O.S.p.Av .......................................... 3
(1)
back PWM converter connected by a dc-link capacitor is used as shown in Fig. 4. The generator-side converter works as an inverter operating in regenerative mode and the grid-side convcirter functions as an ac/dc boost converter with reverse power flow.
where p, A and U are the specific density of air [kg/m3],the area swept by the blades [m'] and the wind speed [m/s], respectively. The power conversion coefficient of the rotor (C,) is defined as the ratio between the mechanical power available at the turbine shaft and the power available in wind. Neglecting the friction losses, the mechanical power available to be converted by the generator is given by
P =o.s.p.c,v 3......................................... "l
r
I,."
.
I,
Om.r*l,.*,b conrrm
Lm%+-,dr CcmCM
LW
Fig 3 Back-to-backPWM converter system for grid connection
(2)
A. Control of induction generator
The generator confroller controls the rotational speed in order to produce the maximum output power, where the indirect vector contrcil is used. The control part consists of a speed controller arid the d-q current controllers. The daxis current component is generally set to maintain the rated field flux in the whole range of speed, while the speed loop will gemrate the q-axis current component to control the generator torque and speed at different wind speed as shown in Fig. 4. The fuzzy logic controller is to find the reference sp:ed along which tracks the maximum power point, which will be described in the following section.
Fig. 2 Variation of power conversion coefficient CO with tip-speed ratio A
The power conversion coefficient is a function of the tipspeed ratio h which is defined as
a = -0,R .................................................. -0
(3)
wherewmand R are the rotor speed and the radius of the turbine blade, respectively. Fig. 2(a) shows that the power converted from the turbine blade is a function of the rotational speed and that the power converted is maximum at the particular rotational speed. Fig. 2(b) shows that the average power conversion coefficient C , is maximum at a particular,?. Hence, to fully utilize the available wind energy to obtain the maximum output power, A should be maintained at its optimum value ,lop,.In this case the maximum available power at any wind speed is given by P, = o.s.p.c,,,,.v 3
..............................
Fig. 4 Control Uock diagram o f inductiongenerator
B. Control of grid-side convener
.(4)
To achieve the full control of the grid-side current, the dc-link voltage must be boosted to a level higher than the amplitude of the grid line-line voltage. The power flow of the grid-side converter is controlled in order to keep the dclink voltage constant. Since the power increasing causes a decrease of the dc-link voltage, the dc-link voitage must be regulated to keep the voltage variation within small tolerance band. To maintain the dc-link voltage constant and to ensure the reactive power flowing into the grid at null, the supply side (convertercurrents are controlled using the d-q vector control approach. The dc-link voltage is controlled to the desired value by using a PI controller and
And the reference speed in the control scheme is determined according to the optimum tip-speed ratio
'
4>N ...............................................
0, =--0
R
(5)
111. CONTROL OF GRID-CONNECTED WIND
POWER GENERATION SYSTEM In order to control the induction generator, the back-to-
2040
2004 35th Annual IEEE Power Electronics Specialists Conference
Aachen, Germany, 2004
the change in the dc-link voltage represents a change in the q-axis (Q current component. When operating at uniy power factor, the demand for the reactive current (&) will be zero. Fig. 5 shows a control block diagram of the gridside converter.
U
Fig. 6 Input and output of fuwy controller
Input membership tinclions
Fig. 5 Control block diagram ofgrid-side converter. output membership Cnulons
Fig. 7 Membership functions of fuzzy controller
IV. FUZZY LOGIC CONTROLLER FOR MPPT The fuzzy logic control is applicable to search the generator speed reference which tracks the maximum output power point(MPPT) at varying wind speeds. The FLC(fuzzy logic controller) block diagram is shown in Fig. 6. The FLC doesn't require any detailed mathematical model of the system and its operation is governed simply by a set of rules. The principle of the FLC is to perturb the generator reference speed y' and to estimate the corresponding change of output power P,. If the output power increases with the last increment, the searching process continues in the same direction. On the other hand, if the speed increment reduces the output power, the direction of the searching is reversed. The fuzzy logic controller is efficient to track the maximum power point, especially in case of frequently changing wind conditions. The input(SP, and 6q' ) and output(nSw,') membership functions are shown in Fig. 7. Triangular symmetrical membership functions are suitable for the input and output, which give more sensitivity especially as variables approach to zero value. The width of variation can be adjusted according to the system parameter. The input signals are first fuzzified and expressed in fuzzy set notation using linguistic labels which are characterized by membership functions before it is processed by the FLC Using a set of rules and a fuzzy set theory, the output of the FLC is obtained [lo]. This output, expressed as a fuzzy set using linguistic labels characterized by membership functions, is defuzzified and then produces the controller output. Figure 7 shows the input and output membership functions and Table I lists the control rule for the input and output variable. The next fuzzy levels are chosen for controlling the inputs and output of the fuzzy logic controller. The variation step of the power and the reference speed may vary depending on the system. In Fig. 7, the variation step in the speed reference is from O.S[rad/s] to O.S[rad/s] for power variation ranging over
2041
TABLE I:Rules of fuzzy logic controller
from -IO[W] to IO[W]. The membership definitions are given as follows: n(negative), n++(very big negative), nb(negative big), nm(negative medium), ns (negative small), z(zero), p(positive), ps@ositive small), pm(positive medium, pb(positive big), and p++ ( very big positive ).
V. EXPERIMENT AND DISCUSSION Fig. 8 shows the experimental setup of reduced-scale at laboratory. The characteristics of the wind turbine are simulated using a torque-controlled dc motor. The ratings and parameters of the cage-type induction generator are listed in Table 11 in Appendix. Also, the specification of the wind turbine blade modeled for the simulator is given in Table 111. The inverter switching frequency is S[kHz], and the current and the speed control respectively sampling periods are I O O [ p ] , I [ms],
A
QFig. 8 Experimental setup
Ill
i Aachen, Germany, 2004
2004 35rh A n n u l IEEE Power Electronics Specialists Conference
. .
...
i Generaurspd !.cen%!%!!PutP!e!
..
lOO[rpm~dii
*
L
~
.-.
...i....i....,.....i....!.
i
!
I
~
I 4OO[Wl/div I ................................................................................................. mcrou!qutpower
.................................. 4OO[ W]/div..~!
8?0[Wb-
i ~
,
i 4oo[w]/dN i P
Converteroulputpower L
...
~
~
L . 1
I
.......
_t. ~..
1 1-. ..... ~,~...
...
I.
4
i
L 0IAk-c-i
:
-y
kord.aiscurrent
+---7 S[A~~N
.............................................................
1
~
-,.., :
!
Converterd-adscurrent
[
A
X A Y ~N
0,5[S]/dN Fig. 9 Different measured waveforms in case of wind speed variation in steps between 6 [ d s ] and 8 [ d s l .
Fig. 10 Different measured waveforms in case ofwind speed varialion in saw-tooth waveform beween 6 [ d s ] and 8[mJs].
. The output signal of the function generator is used as a wind speed, which is read into the DSP through the D/A converter. In fact, the fuzzy controller doesn't require the wind speed measurement. However, it is needed for the turbine simulator using the dc motor torque control. Fig. 9 shows the different variables in case of the stepwise wind speed variation between 6[ms] and 8[m/s]. From the top, wind speed(a), generator speed(b), generator output power(c), converter output power(d), dc link voltage(e), generator q-axis current(f), generator d-axis current(g), converter q-axis current(h), and converter d-axis current(i) are shown. It is seen that according to the wind speed variation the generator speed varies and that its output power is produced corresponding to the wind speed
variation. Due to the vector control of the induction generator, its d-axis current is kept constant and only the qaxis current is varied for the speed control. The dc link voltage is controlled at 34O[V]. For the grid-side converter, the d-axis current is controlled to be zero for unity power factor and the q-axis current is controlled so as to deliver the power flowing from the dc link to the grid-side. Fig. IO shows the different variables in case of the wind speed varying in a szw-tooth waveform, which correspond to those of in Fig. 9. Even though the wind speed varies continuously, the F1.C works well and it gives the good tracking performance for the maximum output power point. Fig. I I shows thr: current control performance of the grid-side converter, .where (a) shows the three-phase line
2042
~
2004 35th Annual I E E E P o w e r Elecrronics Specialists Conference
Aachen, G e r m a n y , 2004
current and (b) shows the unity power factor control Fig. 12 shows the trajectory of the output power corresponding to the maximum at the given wind speed. When the generator speed reference doesn't result i?om the optimal tip-speed ratio, the output power decreases below the maximum value.
@I
4oaldirl
Parameters stator resistance rotor resistance
0.533 [Q]
moment of inertia
0.0071 kgm2
Parameters
Value
max. power conv. coeff. optimal tip-speed ratio cut-in speed rated wind speed
0.45 5
Value
0.93 C I 21
4 [Ids] 12 [Ids]
ACKNOWLEDGMENT
Fig. I 1 Control ofgrid-side wnverter (a) three-phase ac ourput current (b) unity power factor control
This work has been supported by KESRI(R-2003-B362), which is funded by MOCIE(Ministry of Commerce, Industry and Energy) in Korea.
I
REFERENCES [I]
L. H. Hansen, L. Helle, and F. Blaabjerg, "Conceptual sulvey of
generators and power electronics for wind Nrbines", Technical Repon. Riso National Lobunrtory, Roskilde, Denmark, Dec. 2001. [2] Siegfried Heier. Grid Integration of Wind Energy Conversion Sysrem, John Wiely & Sons, 1998. [31 R. Peria R. Cardenas, R. Blasco, G. Asher, and J. Clare, "A cage induetian generator using back to back PWM converters far variable speed grid connected wind energy system", IEEE IECON Rec., pp. 1376-1381,2001.
1-1
Fig. I 2 Output power C U N ~ S for different wind speeds
VI. CONCLUSIONS
A fuzzy logic algorithm for the maximum output power of the grid-connected wind power generation system using a cage-type induction generator has been proposed and implemented. The induction generator was controlled in indirect-vector control method and its speed reference was determined by fuzzy rules. The back-to-back PWM converter was used to connect the generator and the grid. The grid-side converter controls the dc-link voltage and the ac current. The validity of the proposed algorithm has been verified by experimental results for a small-scale system. APPENDIX The induction machine used for test is 230V, three-phase, four poles, 50Hz, 3kW, 1435rpm, ofwhich parameters are
2043
[41 H. G. Kim, D. C. Lee, J. K. Seok, and G. M. Lee, "Stand-alone wind power generation system using vector-controlled cage-type induction generators," ICEMS Proc., in Beijing, pp. 289-292,2003, [SI G. 0. Garcia J. C. Mendes Luis, R. M.Stephan, and E. H. Waranabe, "An efficient conmller for an adjustable speed induction motor drive," IEEE Trans. on /E, vol. 4, pp. 533-539, 1994. [6] Pedersen, P. Z. Grabowski, and P. Thagersen ,"On the energy optimized control of standard and high-efficiency induction motors in CT and HVAC applications," IEEE Trans. on /A, vol. 34, no. 4, pp. 822-831, 1998. [7] D.S . Kirschen, D. W. Novotny, and T,A,. Lipo, "On-line efficiency optimization of a variable frequency induction motor drive," IEEE Tram. on IA, vol. 21, pp. 610-616, 1985. [SI Q. Wang and L. Chang, "An independent maximum power extraction strategy for wind energy wnversian systems," IEEE Canadian Con$ Proc. on ElecandCompr. En&, pp. 1142-1147, 1999. [9l M. G. Simces, Bimal k. Bose , "Design and performance evaluation of fuzzy-logic-based variable-speed wind generation system", IEEE Trans.on 1ndrrstryApplic"tions. vol. 33, pp.956-965, 1997. [lo] G. C. Sousa and B. K. Bose, "Fuuy logic-based on-line efficiency optimization control of an indirect vector wntmlled induction motor drive': IEEE-IECON, Maui, Hawaii, Nov. 1993, pp.1168-1174. [Ill R. M. Hilloowala and A. M. Sharaf , " A rule-bared fuzzy logic wntroller for a PWM inverter in a stand alone wind energy wnvenion scheme" , IEEE /AS Annu. Meering Con5 Rec., Toronto , October 1991993, pp. 2066-2073.
