8th International Conference on Modelling, Identification and Control (ICMIC-2016) Algiers, Algeria- November 15-17, 2016
MPPT fuzzy logic controller of a wind energy conversion system based on a PMSG S.Marmouh,
M. Boutoubat and L. Mokrani,
LTSS Laboratory, Laghouat University, Algeria
[email protected]
LACoSERE Laboratory, Laghouat University, Algeria
[email protected];
[email protected]
Abstract—The aim of this paper is to control a Wind Energy Conversion System (WECS), equipped with a Permanent Magnet Synchronous Generator (PMSG), for maximum power generation. Maximum Power Point Tracking (MPPT) control algorithm is applied to a PMSG whose stator is connected to the grid through a back-to-back AC-DC-AC converter. The Stator Side Converter (SSC) is controlled in such a way to extract the maximum power, for a wide range of wind speed. The Grid Side Converter (GSC) is controlled in order to ensure a smooth DC link voltage between the two converters. The presented simulation results demonstrate the performance and the effectiveness of the control strategy. Keywords---DC Link Control; Fuzzy Logic Controller; Hysteresis Controller; MPPT; Wind Energy Conversion System, PMSG.
I.
INTRODUCTION
Wind energy turbine is considered today as a source of power which allows electrical production with minimum environment perturbations. For a cleaner and economical energy society, wind energy conversion system’s (WECS’s) have received increasing consideration [1]. Wind power system for electricity production is developing very fast, and it is believed that by 2020, the entry level for wind power generation will be approximately 15%–16% [2]. Wind turbines are controlled to work only in a specified range of wind speeds limited by cut-in ( ) and cut-out ( ) speeds. Outside these limits, the turbine should be stopped to protect both the generator and turbine. Fig. 1 shows the standard power curve of a wind turbine [34]. When the wind speed is low, the turbine should be stopped and disconnected from the grid to prevent it from being driven by the generator. The second is the moderate-speed region that is limited by the cut-in speed at which the turbine starts working, and the rated speed ( ), where the turbine is controlled to extract the available power from the wind. [5]. It is important that to protect it from structural overload, it must be cease operating above the cut-out speed. This work focuses on the average-speed region, where the maximum power point tracking (MPPT) algorithm is needed. Due to the instantaneous varying swiftness of the wind, it is necessary to determine the one optimal generator speed that ensures maximum energy given in.
978-0-9567157-6-0 © IEEE 2016
Fig. 1. Standard power waveform of a wind turbine. So, it is important to include a controller that can track the maximum peak independent of wind speed. Generally, MPPT methods can be mostly, classified into those that do not use sensors and those that use sensors. The method uses sensors to track the MPPT by the control of the rotor speed and torque. Basically, it is named the Tip Speed Ratio (TSR) control. The TSR control directly regulates the turbine speed or torque to keep the TSR at an optimal value by measuring wind speed and turbine speed [6]. Noting that, the turbine pitch angle is fixed at its nominal value. This method is faced to some drawbacks, such as the impossibility to adapt the speed of the wind generator in the case of fast variation of the wind speed, due to its inertia [7]. Another method is the Perturb and Observe (P&O) method which is a simple strategy to implement with a low cost and does not require prior knowledge of neither wind speed nor generator's parameters. Nevertheless, this MPPT algorithm has not a good tracking performance [8]. In [9], the authors have reviewed an MPPT strategy based on power signal feedback (PSF). It is required to have the knowledge of the wind turbine’s maximum power curve, and track this curve through its control mechanisms. Else, they studied the hill-climb searching (HCS). This method works well only for very small wind turbine inertia. In [10] hybrid methods namely Hybrid HCS (HHCS) is proposed, where the (OTC) control is jointed to the conventional HCS algorithm. It merges the benefits of two MPPT approaches to resolve the problem of the standard HCS algorithm. Among electric generators, the permanent magnet synchronous generator (PMSG) is favored due to its high efficiency, reliability, power density, light weight, and selfexcitation structures [11–13].
8th International Conference on Modelling, Identification and Control (ICMIC-2016) Algiers, Algeria- November 15-17, 2016 This paper presents the control of a WECS, equipped by a PMSG, for maximum active power generation. For this purpose, a speed Fuzzy Logic Controller (FLC1) is developed for the MPPT strategy. This strategy is applied to the SSC by using a hysteresis control and an optimal generator speed reference which is estimated from different wind speeds. Another Fuzzy Logic Controller (FLC2) is used in order to guarantee a smooth DC voltage between the SSC and the GSC to its reference value by controlling the GSC.
Where, is the rotational speed of the turbine and R is the turbine blade radius. So, to extract a maximum power, at a variable wind speed, one can use the Tip Speed Ratio (TSR) technique. TSR control method regulates the rotational speed of the generator to maintain the tip speed at its optimal value ( ), as shown in Fig. 3.
The feasibility and effectiveness of the control strategy, in terms of active power production and a unity power factor, at the grid side, have been tested by simulation. II.
MODELING OF THE WIND ENERGY SYSTEM
The scheme of the studied system is presented in Fig. 2. As shown in Fig. 2, the studied WECS is formed by the wind turbine and the PMSG, whose stator is connected to the grid through a back to back converter. Noting that, the presence of a DC-link capacitor provides a decoupling between the two converters.
Fig. 3. Power coefficient as a function of the tip speed ratio.
B. Permanent Magnet Synchronous Generator modeling The stator voltages of the PMSG, expressed in d-q axis, are given by the following equations [15]:
d vsd RS isd sd p g sq dt v R i d sq p sq S sq g sd dt
Fig. 2. Scheme of the studied system.
(4)
A. Wind turbine aerodynamic model The wind turbine converts the wind power to a mechanical power which is expressed by [14]:
1 P C ( , ) A v3 m 2 P (
(1)
Furthermore, the available turbine mechanical torque ),when the pitch angle is fixed, can be expressed as [14]:
1 Tm CP ( ) A v3 / w m 2
(2)
Where ρ is the air density (typically equal to 1.225 kg/m3), β is the pitch angle (in degree), A is the area swept by the rotor blades (m2); is the wind speed (m/s) and is the wind-turbine power coefficient. Elsewhere, the tip speed ratio is expressed as follow [14]:
wm R v
978-0-9567157-6-0 © IEEE 2016
(3)
Where and are the components of the stator flux which are given by the following expressions:
sd Ld isd S sq Lq isq Where inductances;
the permanent flux; and is the stator phase resistance.
(5)
are the d q-axis
Elsewhere, the classical expression of the electromagnetic torque is given by [15]:
Te
3 P[(Ld Lq )isd isq s isq ] 2
(6)
Where: P is number of pole pairs. Taking as state vector the components of the two stator currents ( and ) and the generator speed , we obtain the following state representation of the PMSG in the (d-q) frame:
8th International Conference on Modelling, Identification and Control (ICMIC-2016) Algiers, Algeria- November 15-17, 2016 1 1 disd ( Rs isd Lq p g isq ) L dt Ld d 1 disq dt L ( Rs isq Ld p g isd p gs ) 0 q d g 1 3 3 dt ( p( Ld Lq )isd isq ps isq f g g ) 0 2 J 2
0 1 Lq 0
0 vsd 0 vsq T g 1 J
III. STATOR SIDE CONVERTER CONTROL As can be remarked from (6) and in order to achieve a linear relationship between the q-axis current ( ) and the electromagnetic torque ( ), the d-axis stator current ( ) is regulated to be equal to zero. Hence, the electromagnetic torque can be regulated to its reference value ( ) by acting on the the q-axis current ( ) as follows:
isqref
2 Teref 3 ps
(7)
Elsewhere and as it has been known in the literature, the fuzzy logic controller is mainly used in nonlinear systems which can’t be accurately modeled and have more inputs, uncertain factors and inaccurate properties. In our case the speed fuzzy controller includes four parts: fuzzification, fuzzy rule base, reasoning and defuzzification as shown in Fig. 5. The inputs of the fuzzy controller (FLC1) are the speed error (e) and its variation (∆e) and which are expressed as follow:
e gref g
(9)
e (1 z 1 )e
(10)
The FLC1 output is the increment of the electromagnetic torque ( ). , and are the normalisation gains.For the fuzzification, the membership functions of e, ∆e and the increment of the electromagnetic torque ( ) are presented in the Fig. 5.
Moreover, to extract the maximum power from the wind and ensure the MPPT strategy, the reference generator speed is estimated by the following equation:
gref
opt R
v
With δ is the gearbox ratio and
(8) (in our case).
Fig. 5. Speed fuzzy logic controller.
Finally, the control block diagram of the stator side converter is shown in Fig. 4. As shown in Fig. 4, a reference current ( was derived by using a Fuzzy Logic Controller (FLC1), in order to regulate the generator speed ( ) to its optimal value ( and ensure the MPPT strategy.
Fig. 6. Membership functions of the inputs and the output of the FLC1.
The fuzzy rules are summarized in TABLE I. The rules are described in the following: NB: Negative-Big; NS: NegativeSmall; EZ: Equal-Zero; PS: Positive-small; PB: Positive-Big. TABLE I.
Inference matrix.
NB
NS
EZ
PS
PB
NB
NB
NB
NS
NS
EZ
NS
NB
NS
NS
EZ
PS
EZ
NB
NS
NS
EZ
PS
PS
NS
EZ
PS
PS
PB
PB
EZ
PS
PS
PB
PB
Fig. 4. Block diagram of the rotor side converter control.
Noting that, the other component of the stator current ( is taken equal to zero. Then, both d-q reference currents ( and were transformed to their natural abc components ( and used for implementing the hysteresis modulation as shown in Fig. 4.
978-0-9567157-6-0 © IEEE 2016
8th International Conference on Modelling, Identification and Control (ICMIC-2016) Algiers, Algeria- November 15-17, 2016 To obtain the output ( ) of the FLC1, the defuzzication used in this work is based on the center of gravity method. For the 25 rules of the matrix inference, ( ) is calculated by the following expression: 25
Tem K
i 1 3 25
Qg
(17)
x s
ci Gi i
i 1
3 (vdg idg vqg iqg ) 2 3 (vqg idg vdg iqg ) 2
Pg
(11)
s
ci i
Then, the reference electromagnetic torque is expressed by:
Tem (k ) Tem (k 1) Tem
(12)
IV. GRID SIDE CONVERTER CONTROL In this work, the issue which needs to be addressed by the control of the GSC is to guarantee a smooth DC voltage between the two converters. This task is achieved by using a second Fuzzy Logic Controller (FLC2), as shown in Fig .6. In this case, the inference matrix is the same as that of the speed Fuzzy Logic Controller (FLC1) by replacing e and ∆e with and respectively. The error and its variation are expressed respectively by the following equations:
edc vdc _ ref vdc
(13)
edc (1 z 1 )edc
(14)
Besides, the grid phase voltages can be expressed as follow: di vag Rg iag Lg ag vainv dt dibg (15) vbg Rg ibg Lg vbinv dt di vcg Rg icg Lg cg vcinv dt
If the d-axis is aligned with the stator voltage, one can write: vdg =us and vqg=0. Hence, the active and reactive powers expressions (equation (18)) are easily simplified as follows:
3 Pg u s idg 2 3 Qg iqg u s 2
By neglecting the GSC losses, the DC power has to be equal to the active power flowing between the grid and the GSC and one can write: 3 Vdc idc u s idg 2 dVdc C idc im dt
Finally, the control block diagram of GSC is presented in Fig. 7. As can be remarked in (18), the DC capacitor voltage ( is controlled by the current in the voltage vector-oriented reference frame. Thus, a reference current was derived from the Dc link voltage error and its variation by tuning the FLC2, as shown in Fig. 7. Also, to achieve a unity power factor at the grid side, the reactive current must be taken equal to zero. Finally and after a dq-abc transformation of these reference currents, hysteresis modulation may then be implemented to control the GSC, as shown in Fig 7
vdg Rg idg Lg
(16)
Where, are the converter voltages, are the grid currents, are the grid voltages, and are the resistance and inductance respectively of the line between the GSC and the grid. Besides, the active and reactive power, exchanged between the grid and the GSC, are given by the following equations respectively:
978-0-9567157-6-0 © IEEE 2016
(19)
So, the DC voltage (Vdc) can be controlled by acting on the d component of the line current (idg).
By using Park transformation, (15) can be expressed as follows:
didg s Lg iqg vdinv dt di vqg Rg iqg Lg qg s Lsidg vqinv dt
(18)
Fig. 7. Block diagram of the grid side converter control.
8th International Conference on Modelling, Identification and Control (ICMIC-2016) Algiers, Algeria- November 15-17, 2016 V. SIMULATION RESULTS
0
(a)
Reavtive Power -500
-1000
Active Power 2
4
6
8
10
12
14
16
18
20
12
14
16
18
20
Grid Voltage
200
(b)
In this section, the WECS is controlled to track its maximum power operating point with a unity power factor at the grid side. Figs. (8-13) show simulation results for a wind speed in form presented in Fig (8-a). As can be seen from the plots, the generator speed ( ) track its optimal value ( ) to ensure the MPPT (see Fig.8 (b)). Moreover, the power coefficient ( ) and the tip speed ratio (λ) are kept around their optimum values of 0.4993 and 6.4 respectively (see Figs .9 (ab)). The active power ( ), injected in this grid, is varied according to the MPPT strategy (see Fig. 10 (a)) and a unity power factor is ensured at the grid side. In fact, the reactive power (Qg) injected in the grid is practically equal to zero (see fig. 10 (a)). The zoom of the phase grid voltage ( ) and its corresponding current (iag) shows that the WECS produces only active power to the grid with a unity power factor (see Fig. 11(a)). The DC voltage (Vdc) is regulated practically to its reference value of 700 V by controlling the GSC, as shown in Fig .11 (b). The direct component of the stator current of the PMSG is kept around zero according to the control strategy (see Fig.12 (a)). The other stator component is regulated to its reference value (see Fig. 12 (b)), in such a way that the electromagnetic torque tracks its reference , in order to achieve the MPPT strategy (see Fig .13 (b).
Grid Current iag*100
0
-200 2
4
6
8
10
Time(S)
Fig. 10. MPPT power generation: a) Stator active power (W) and reactive power (VAr), b) the grid phase voltage ( ) (V) and the corresponding current ( ) (A).
13 12
(a)
11 10 9 8 7
2
4
6
8
10
12
14
16
18
20
140 130
Optimal Generator Speed
(b)
120 110
Fig.11. a) Zoom of the grid phase voltage Vag (V) and the corresponding
100 90
current (
Actuel Generator Speed
80 2
4
6
8
10
Time (s)
12
14
16
Fig. 8. a) Filtered wind speed (v) (m/s), b) Generator speed (
20
) (V).
5
) and its
) (rd/s). a)
reference (
18
(A)), b) DC voltage (
0
0.4993 0.4993
-5 0
(a)
0.4993
2
4
6
8
2
4
6
8
10
12
14
16
18
20
10
12
14
16
18
20
0.4993 0.4993
0
0.4992 2
4
6
8
10
12
14
16
18
-5
20
b)
-10
(b)
6.45
-15
6.4
-20
6.35
6.3
-25 0
Time (S) 2
4
6
8
10
12
14
16
Time (S) Fig. 9. a) Power coefficient ( ), b) Tip speed ratio ( ).
978-0-9567157-6-0 © IEEE 2016
18
20
Fig. 12. Performance of conventional PMSG current-loop: a) The stator daxis current (A), b) the stator q-axis current (A).
8th International Conference on Modelling, Identification and Control (ICMIC-2016) Algiers, Algeria- November 15-17, 2016 2000
a)
1500
1000
500
0
2
4
6
8
2
4
6
8
10
12
14
16
18
20
10
12
14
16
18
20
-2 -4
b)
-6 -8 -10 -12 -14 -16
Time (S) Fig. 13. a) The aerodynamic power of the turbine ( ) (W), b) The electromagnetic torque of the PMSG (N.m).
III.
CONCLUSION
This paper has discussed the control of a WECS based on a PMSG for MPPT power generation. The SSC has been controlled, using a speed fuzzy logic controller, in such a way to permit to the wind system to track its maximum power operating point for a wide range of wind speed and ensure the MPPT strategy. The GSC is controlled also by using another fuzzy logic regulator to ensure a smooth DC voltage between the two converters (SSC and GSC). As perspectives, a coordinated control, between the SSC and the GSC, can be proposed for MPPT active power production and power quality improvement, without any system over rating. Elsewhere, the WECS can be also controlled to operate in standalone mode. REFERENCES [1] Arifujjaman, M., ―Reliability comparison of power electronic converters for grid-connected 1.5kW wind energy conversion system,‖Renew.Energy, Vol. 57, pp. 348–357, September 2013. [2] Alok Pratap, Zakaria Ziadi, Naomitsu Urasaki, and Tomonobu Senjyu ―Smoothing of Wind Power Fluctuations for Permanent Magnet Synchronous G enerator-Based Wind Energy Conversion System and Fault Ride through Consideration‖ Electric Power C omponents and Systems, Vol. 43(3), pp. 271–281, 2015 [3] Bianchi F, De Battista H, and Mantz R. Wind turbine control systems: principles, modelling and gain-scheduling design (advances in industrial control). Lavoisier; 2006. [4] Feng X, Jianzhong Z, and Ming C. ―Analysis of double objectives control for wind power generation system with frequency separation‖. In: 2011 4th International Conference on Electric Utility Deregulation and Restructuring and Power Technologies (DRPT), pp. 1366–71,2011. [5] Bianchi FD, De Battista H, and Mantz RJ. Wind turbine control systems: principles, modelling and gain scheduling design.Springer-Verlag; 2007. [6] HaeGwangJeong, Ro HakSeung and KyoBeumLee,‖An Improved Maximum Power Point Tracking Method for Wind Power Systems‖, Energies,Vol. 5, pp. 1339-1354, 2012.
978-0-9567157-6-0 © IEEE 2016
[7] Quincy, W and Liuchen, C.‖ An intelligent maximum power extraction algorithm for inverter-based variable speed wind turbine systems‖. IEEE Transactions on Power Electronics, Vol. 19, n0 .5, pp. 12421249,2004. [8] M. A.Abdullah, A. H. M.Yatim, C.W.Tan, and R.Saidur, ―A review of maximum power point tracking algorithms for wind energy systems, Renewable and Sustainable Energy Reviews”, Vol. 16, pp. 3220– 3227, 2012. [9] K. Belmokhtar, M.L. Doumbia* and K. Agbossou ―Novel fuzzy logic based sensor less maximum power point tracking strategy for wind turbine systems driven DFIG‖, ScienceDirect,pp. 0360-5442,2014 [10] S. Lalouni1 and D.Rekioua1 and K.Idjdarene1,and A.M.Tounzi2‖ An improved MPPT algorithm for wind energy conversion system‖ J. Electrical Systems. Vol. 10-4, pp. 484-494, 2014. [11] Li S, Haskew TA and Xu L. ―Conventional and novel control designs for direct driven PMSG wind turbines‖. Electric Power Systems Research Vol. 38, pp. 80-328, 2010. [12] Zhipeng Q, Keliang Z and Yingtao L. ―Modeling and control of diode rectifier fed PMSG based wind turbine‖. In: 2011 4th international conference on electric utility deregulation and restructuring and power technologies (DRPT). pp.1384–8,2011 [13] Muyeen SM, Takahashi R, Murata T and Tamura J. ―A variable speed wind turbine control strategy to meet wind farm grid code requirements‖. IEEE Transactions on Power Systems. Vol. 25, n0 1:331– 40, february 2010. [14] Djamila Rekioua, ―Wind Power Electric Systems, Modeling, Simulation and Control‖,Springer-Verlag London 2014 [15] M. Singh, V. Khadkikar, and A. Chandra, ―Grid synchronisation with harmonics and reactive power compensationcapability of a permanent magnet synchronous generator-based variable speed wind energy conversion system‖ IET Power Electron. Vol. 4, no. 1, pp. 122-130, 2011.
