evaluation of voltage flicker emissions of variable speed dfig-based

EVALUATION OF VOLTAGE FLICKER EMISSIONS OF VARIABLE SPEED DFIG-BASED WIND TURBINES Walid S. E. Abdellatif 1, Ali H. Kasem Alaboudy 1 and Ahmed M. Azmy 2 1

Faculty of Industrial Education, Suez University, Suez, Egypt 2 Faculty of Engineering, Tanta University, Tanta, Egypt

Wind-energy systems cause fluctuations more than any other sources and hence, voltage fluctuations are expected with these units. This paper addresses the problem of voltage fluctuation caused by doubly fed induction generators (DFIG)based variable-speed wind turbines. Flicker emission of these units is investigated during continuous operation. The effects of grid strength, angle of grid impedance and site parameters, such as wind speed and wind turbulence, on voltage fluctuation are investigated. A flickermeter model, implemented according to IEC61000-4-15, and all system components are implemented in MATLAB/SIMULINK environment. Finally, a comparison between variablespeed and constant-speed technologies, related to short-term flicker severity is introduced. Keywords: Variable-speed wind turbines, DFIG, Maximum power point tracking (MPPT), Voltage flicker.

1. INTRODUCTION The growing demand for electric energy throughout the world has highly motivated the use of renewable sources of energy. Among the unconventional renewable-based energy sources that have been intensively studied, wind energy seems to have a vital role in the near future. Over the last ten years, the global wind energy capacity has increased rapidly and became the fastest developing renewable energy technology. Basically, there are many reasons for using more wind energy within power grids. For instance, wind generation represents a clean and renewable source with minimal running cost requirements [1]. Variable speed wind turbine (VSWT) topologies include many different generator/converter configurations, based on cost, efficiency, annual energy capturing, and control complexity of the overall system. The doubly-fed induction generator (DFIG) is the most commonly used device for wind power generation, The DFIG consists of a wound rotor induction generator (WRIG) with the stator directly connected to the grid, whereas the rotor is connected through a power electronic converter. The power converter controls the rotor frequency and thus the rotor speed. This concept supports a wide speed range operation, depending on the size of the frequency converter. Typically, the variable speed range is about 30% around the synchronous speed. The rating of the power electronic converter is only 25– 30% of the generator capacity; which allows a variable-speed operation over a large, but restricted range. Generally, there are two converters, the rotor-side converter and the grid-side

converter [2]. The grid side converter is connected to the grid by using a coupling inductor. The three-phase rotor-winding is connected to the rotor side converter by slip rings and brushes and the three-phase stator-windings are directly connected to the grid [3]. Flicker is generally attributed to voltage fluctuations caused by load flow changes in the grid. The high power fluctuations due to grid-connected wind turbines depend on the utilized wind power generation technology. The wind speed changes over moments, hours, days and seasons, consequently, due to wind speed variations, wind gradient and tower shadow effect, the output power from wind turbine exhibits high fluctuations, which cause flicker emission produced by grid connected wind turbines during continuous operation [4]. One of the most important wind power quality considerations is the effect of voltage fluctuations, which disturb the sensitive electric and electronic equipment's. This may lead to a great reduction in the life span of most equipment. Flicker has widely been considered as a serious drawback and may limit the maximum amount of wind power generation that can be connected to the grid [5]. For the above mentioned reasons, it is essential to investigate the integration of wind turbines (WT) and its influence on the power quality due to the uncontrollable nature of distributed wind energy. Among the influences of wind energy on the power quality aspects, flicker, dips and harmonics mitigation have especial importance [6]. There are numerous factors that affect flicker emission of grid-connected wind turbines during continuous operation, such as wind characteristics (e.g. mean wind speed, turbulence intensity) and grid conditions (e.g. short circuit capacity, grid impedance angle, load type) [710]. The type of wind turbine also has an influence on flicker emission. Variable speed wind turbines have shown better performance related to flicker emission in comparison with constant speed wind turbines [8], [11]. The lighting flicker level is generally used to measure voltage fluctuation. Voltage fluctuation disturbs the sensitive electric and electronic equipment. And, hence, the stander IEC 61400-21 is developed to provide procedures for determining the power quality characteristics of wind turbines especially flicker emissions [12]. In this paper, detailed models of variable speed DFIG-based wind turbines, control actions and the standard flickermeter according to IEC 61000-4-15 are developed to accurately evaluate the flicker emissions under different operating conditions. The DFIG is connected to the grid at the point of common coupling (PCC) via an ac-dc-ac back-to-back converter set. Two control schemes are developed for rotor- and grid-side converters. The control of the rotor side converter is developed to achieve maximum power point tracking (MPPT), while, the control scheme of the grid side converter is designed to operate at unity power factor and stabilize the dc link voltage to its nominal value. The target is to investigate the fluctuations caused by variable speed wind turbine and the effect of site parameters (mean wind speed and turbulence intensity) and grid parameters (grid short circuit ratio and grid impedance angle) on voltage fluctuation. In addition, the performances of variable speed wind turbines and constant speed units are compared regarding the short-term flicker severity, Pst.

2. VARIABLE SPEED WIND TURBINE MODELING Variable speed wind turbine system is shown in Fig. 1. The system under study comprises a variable speed wind turbine, equipped with a gearbox, a doubly-fed induction generator (DFIG) and back-to-back IGBT voltage source converter. The DFIG is the most commonly used device for wind power generation.

Figure 1. Variable speed wind turbine system.

The output power Pm as a function of the power coefficient Cp is given by: [13]. Pm 

1 AV 3C P (, ) 2

(1)

On the other hand the tip-speed ratio is defined as: R  r V

(2)

where ρ is the air density (kg/m3), R is radius of the turbine blade (m), V is the wind speed (m/s), ωr is turbine angular speed, Cp is the power coefficient and β is pitch angle. A typical Cp-λ characteristic, for different values of the pitch angle β, is illustrated in Fig. 2. The maximum value of Cp is assumed to be: Cp-max = 0.48. This value is achieved for β = 0 degree and for λ = 8.1. This particular value of λ will be termed the nominal value (λ _nom).

Figure 2. Cp -λ characteristics, for different values of the pitch angle β.

M echanical output pow er (pu)

From (1), to extract maximum available power at the available wind speed, V, the wind generator has to run at a particular speed that corresponds to wind speed in such a way that the maximum power point is tracked as shown in Fig.3. The objective of operating in the maximum power point tracking (MPPT) mode is to maximize power extraction at low to medium wind speeds. This is accomplished by following the maximum value of the wind power coefficient (Cp_max) as depicted in Fig. 3. 1

12 m/s

Maximum power curve

0.8 10.8 m/s

Maximum power point

0.6 9.6 m/s 0.4

8.4 m/s

0.2

7.2 m/s 6 m/s

0 0

0.2

0.4

0.6 0.8 1 Generator speed (pu)

1.2 pu 1.2 1.4

Figure 3. Wind turbine mechanical power output rotor speed (dotted line shows the MPPT operation points).

2.1 Doubly–fed induction generator system A significant advantage with using DFIG is the ability to produce power more than its rated value without overheating. DFIG is able to transfer maximum power over a wide speed range in both sub and super synchronous modes. A DFIG-based wind turbine can transmit power to the network through the generator stator and rotor. Recently, DFIG as a variable speed generator has attracted a wide interest for application with wind energy [14]. The d-q equivalent circuit diagram of a DFIG is shown in Fig. 4. The electrical equations of the DFIG based on the equivalent circuit are found in [15]. 2.2 Control system The control system is an important item for wind turbine performance. It maximizes the extracted power from the wind through all the components and also makes sure that the delivered power to the grid complies with the interconnection requirements. The control systems at the grid-side converter and the rotor-side converter are based on PI controller. 2.2.1 Rotor-side converter The rotor-side converter is used to control the active power of the DFIG and the voltage level measured at the stator terminals. It is important to control the DFIG output active power in order to adjust the angular speed of the turbine rotor so as to maximize the extracted power for a certain wind speed [16].

Figure 4. The d-q equivalent circuit of DFIG.

The current mode control is used to achieve the optimum wind power tracking. The optimum mechanical power is given according to the rotor speed. The reference reactive current, Iqr-ref reflects the PI controller output, which depends on the error signal produced from comparing the actual and the required values of the reactive power. The rotor reference current has to be limited according the rotor-side converter rating. The reference components are compared with their actual components and the error signal is tuned through PI controller to represent the control contribution to correct the rotor voltage signal as illustrated in Fig. 5. The corrected rotor voltage waveform is applied to the pulse width modulation (PWM) as a reference waveform to create the corresponding switching signals for the rotor-side converter.

Figure 5. Structure of control system for rotor-side converter.

2.2.2 Grid-side converter The grid side converter is responsible for controlling the voltage level at the DC capacitor and the reactive power exchange with the grid. The control system for the grid side converter is presented in Fig. 6. The necessary measurements for the control system are AC voltages and currents at the grid side as well as the value of the DC voltage, while the AC-signals are transformed in the dq system. As shown in Fig. 6, the control scheme consists of the following units and loops:  A phase-locked loop (PLL) which is responsible for synchronizing the positive-sequence components of the three-phase primary voltage V. The PLL defines the angle θ, which is used to compute the direct-axis and quadrature-axis components of the AC three-phase voltages and currents denoted by Vd, Vq or Id, Iq on the diagram.  The AC voltage regulator loop: whose output is the reference current Iqref for the current regulator, where Iq is the current in quadrature with voltage that controls reactive power flow.  The DC voltage regulator loop: whose output is the reference current Idref for the current regulator, where Id is the current in phase with the voltage that controls active power flow.  The current regulator loop: that is responsible for controlling the magnitude and phase of the voltage generated by the PWM converter, i.e. V'd and V'q. This loop depends on Idref and Iqref reference currents produced respectively by the DC voltage and AC voltage regulators.

Figure 6. Structure of control system for grid-side converter.

3. FLICKER EMISSION OF DFIG-BASED GRID CONNECTED WIND TURBINES The contribution of wind turbines to voltage fluctuations is divided into continuous operation and switching operation. This paper focuses only on flicker emission during continuous operation. A wind turbine with DFIG is integrated to external power system

represented by constant voltage source connected in series with its Thevenin’s equivalent impedance as in Fig.7. The active and reactive power exchange at the point of common coupling (PCC) is varying according to the incident wind speed and; as a result, the voltage at this point will fluctuate. Voltage fluctuation caused by these wind turbines raises certain limitation regarding the penetration level of the wind generation since it can result in serious electrical disturbances on utility power grids [17].

Figure 7. A simple diagram for wind turbine generator connected to grid.

Fig. 7 shows a simplified diagram of a wind turbine generator connected to the utility grid. If the variation of the active power injected to the grid is ΔP and the corresponding variation of the reactive power delivered to the grid is ΔQ, then voltage fluctuation, at PCC, ΔV/V, is given as follows:

V  P.R  Q. X V

(3)

V  S .Z . cos(   ) V

(4)

This can be rewritten as follows;

where R is the resistance of the grid impedance (pu), X is the reactance of the grid impedance (pu). ΔS is the apparent power variation, S  P 2  Q 2 (pu), Z is the grid impedance X R

amplitude (pu), θ is the grid impedance angle,   tan 1( ) , and φ is tan-1(ΔQ/ΔP). From (3), it is obvious that the voltage fluctuation is not only dependent on the reactive power variation, ΔQ, but also on the active power variation ΔP. Voltage flicker is typically assessed by a standard flickermeter. The nominal voltage is assumed to be 1 pu.

4. FLICKERMETER MODEL Flicker, which is one of the important power quality aspects, is defined as “an impression of unsteadiness of visual sensation induced by a light stimulus, whose luminance or spectral distribution fluctuates with time” [18]. Flicker level is measured by the use of a flickermeter described in IEC 61000-4-15. The flicker level depends on the amplitude, shape and

repetition frequency of voltage fluctuation. The flickermeter takes voltage as an input and gives the short-term flicker index, Pst, as an output. The flickermeter can be divided into two main parts: 1) A simulator for the response of the lamp–eye–brain chain. 2) An on-line statistical analyzer of the flicker signal and a demonstrator of the results. Blocks 1, 2, 3, and 4 represent the first part, while the second part is symbolized by block 5, as illustrated in Fig. 8. More details about flickermeter description are presented in [19], [20].

Figure 8. Block diagram of flickermeter according to IEC 61000-4-15.

5.

SIMULATION RESULTS AND DISCUSSION

Owing to the characteristics of direct-connected constant-speed wind turbines, the active and reactive power exchange at the point of common coupling (PCC) is varying according to the incident wind speed; as a result, the voltage at this point will be fluctuated. Voltage fluctuation caused by these wind turbines raises certain limitation regarding the penetration level of the wind generation since it can result in serious electrical disturbances on utility power grids. Variable-speed wind turbines have shown better performance related to voltage fluctuation in comparison to constant-speed wind turbines. The better performance of Variable-speed wind turbines is due to the buffering effect of the back-to-back converter set, and the rotor speed flexibility that converts power spikes into speed variations. Although the variable-speed wind turbine produces lower flicker levels. The impacts of different factors on the short-term flicker index Pst, are presented in the following subsections. 5.1 The effect of short circuit ratio (SCR) Fig. 9 shows the effect of SCR on the short-term flicker index Pst. This simulation test is carried out, where the mean wind speed at hub level is maintained at 10 m/sec, the site turbulence is 12% and the X/R ratio equals ten. As shown in Fig. 9, it can be seen that the flicker level decreases with the increase of short circuit ratio. In addition, variable speed DFIG based wind turbines have shown better performance related to voltage fluctuation in comparison to constant speed squirrel cage induction generator (SCIG)-based wind turbines for all values of SCR. 5.2 The effect of grid impedance angle Fig. 10 shows the variation of the flicker emission with the impedance angle θ, (θ=tan1 (X/R)). The SCR is maintained at 50 pu. The mean wind speed at hub level is 10 m/sec and site turbulence is 12%. From Fig. 10, it can be seen for constant speed operation that the

flicker decreases with the increase of θ to a certain minimum point and, then, the flicker increases again with increase of θ. The minimum point of voltage fluctuation occurs when θ +φ =90°. The minimum point here is recognized at θ ≈ 45°. On the other hand, the voltage flicker effect on DFIG is lower than on SCIG, where the Pst is almost constant at a very low level as shown in Fig. 10. 2.4 2.2 Constant speed (SCIG) Variable speed (DFIG)

Short-term flicker index, Pst

2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 10

15

20

25 30 35 Short circuit ratio (pu)

40

45

50

Figure 9. The effect of SCR on the short-term flicker index Pst. 0.7 Constant speed (SCIG) Variable speed (DFIG)

Short-term flicker index, Pst

0.68 0.66 0.64 0.62 0.6 0.58 0.56 0.54 0

10

20

30

40

50

60

70

80

90

Grid impedance angle, arctan(X/R),deg.

Figure 10. Variation of Pst with the grid impedance angle.

5.3 The effect of mean wind speed at hub level Fig. 11 shows the effect of mean wind speed with the short-term flicker index, Pst at X/R ratio equals one and turbulence intensity equal to 12%. The flicker emission increases with increasing wind speed. Saturation or flat variation is recognized when the mean wind speed exceeds the rated value. The flicker emission effect on DFIG increases slightly with the increase of the mean wind speed and consequently, the DFIG shows better performance than SCIG as shown in Fig.11.

0.8

Short-termflicker index, Pst

0.75 Constant speed (SCIG) Variable speed (DFIG)

0.7

0.65

0.6

0.55 10

11

12

13

14

15

16

17

18

Mean wind speed (m/sec)

Figure 11. Variation of Pst with the mean wind speed.

5.4 The effect of turbulence intensity In this case, the mean wind speed at the hub level is maintained at 10 m/sec., the fault current level is 10 pu, and the grid impedance angle is 45°. Fig. 12. Shows the variation of Pst with the turbulence intensity. The increase in the wind speed turbulence increases the power variability; then the flicker emission increases with the increase of the turbulence intensity. In all cases, the DFIG is better than SCIG regarding the flicker emission with the increase of the turbulence intensity as shown in Fig. 12. 1.6

Short-termflicker index, Pst

1.5 1.4 Constant speed (SCIG) Variable speed (DFIG)

1.3 1.2 1.1 1 0.9 0.8 0.7 5

10

15

20

25

Turbulence intensity (%)

Figure 12. Variation of Pst with wind speed turbulence.

6. CONCLUSION This paper tackles the voltage fluctuation caused by variable speed DFIG based on wind turbines. For constant-speed operation, the voltage flicker decreases with the increase of X/R

ratio of the grid impedance until a certain minimum point is reached. Beyond this point, the flicker emission increases with increasing X/R ratio of the grid impedance. For variable-speed operation, the voltage flicker is almost constant at a very low level. The flicker level decreases with the increase of fault level, while the flicker emission increases with the increase of wind speed (in the region below the rated wind speed). Generally, variable-speed wind turbines have shown better performance related to voltage fluctuation in comparison to constant-speed wind turbines.

7. APPENDIX DFIG Based Wind Turbine System Data 1. Induction Generator Parameters: Nominal power: Voltage: Frequency: Stator resistance: Stator leakage inductance: Rotor resistance: Rotor leakage inductance: Magnetizing inductance: No. Of pole pairs: Inertia of the whole system: Friction factor: Rated wind speed: Turbulence intensity:

Pnom=6*1.5/0.9MW; Vnom=575V; Fnom=60Hz; Rs=0.0071 pu; Ls=0.1710 pu; Rr=0.0050 pu; Lr=0.1560 pu; Lm=2.9Pu Pm=3pair pole J=0.089k.g.m^2 B=0.05479N.m VW=12m/s; Iu = 12%;

2. DC bus and Grid Parameters: DC-Link voltage: Vdc=1200V Capacitance of the dc link: C=0.06F Grid frequency: F=60Hz -The grid is represented by voltage source of 1 pu behind the

R, L impedance.

3. PI parameter: Rotor side converter current loop: Kp=0.3 & Ki=8 Grid side converter current loop: Kp=2.5 & Ki=500 Dc-link loop: Kp=0.002 & Ki=0.05

8. REFERENCES [1] Ackermann T., “Wind power in power systems”, John Willey and Sons, 2005. [2] Qu L. and Qiao W. “Constant power control of DFIG wind turbines with supercapacitor energy

storage” IEEE Transactions on Industry Applications, vol. 47, no. 1, January/February 2011.

[3] Biswas S. “Application of unified power flow controller in DFIG based wind turbine”

International Journal of Science and Research (IJSR), Vol. 3, No. 6, June 2014. [4] El-Tamaly H. H., Wahab M. A. A. and Kasem A. H., “Voltage fluctuation produced from wind

turbines directly connected to grid,” 10th International Middle East Power Systems Conf., Mepcon‟10th , Port-Said, Egypt, 13-15 December 2005. [5] Machmoum M., Hatoum A., Bouaouiche T., "Flicker mitigation in a doubly fed induction generator wind turbine system," Math. Comput. Simul, 2010. [6] Tamaly H. H. El-, Wahab M.A. A., and Kasem A. H.," Simulation of Directly Grid-Connected Wind Turbines for Voltage Fluctuation Evaluation, "International Journal of Applied Engineering Research ,vol.2, no.1, pp. 15–30, 2007. [7] Larsson Å., “Flicker Emission of Wind Turbines during Continuous Operation,” IEEE Trans. Energy Conversion, vol. 17, no. 1, pp. 114-118, Mar. 2002. [8] Sun T., Chen Z., and Blaabjerg F., “Flicker Study on Variable Speed Wind turbines with Doubly Fed Induction Generators,” IEEE Trans. Energy Conversion., vol. 20, no. 4, pp. 896–905, Dec. 2005. [9] Papadopoulos M. P., S. Papathanassiou A., Tentzerakis S. T., and Boulaxis N. G., “Investigation of the Flicker Emission by Grid Connected Wind Turbines,” in Proc. 8th Int. Conf. Harmonics Quality of Power, vol. 2, pp. 1152–1157,Athens, Greece, Oct. 14–16, 1998. [10] Abdel-Latif W. S. E., Alaboudy A. H. K., Mostafa H. E. and Fekry M. Y., " Evaluation and mitigation of voltage flicker caused by constant speed wind turbines," IEEE PES ISGT Middle East Conference, Saudi Arabia, Dec. 2011. [11] Chen Z., Guerrero J. M., and Blaabjerg F., "A review of the state of the art of power electronics for wind turbines," IEEE Trans. Power Electronics, vol. 24, no. 8, Aug. 2009. [12] Carrillo C., Diaz-Dorado E., and Cidras J. “PSCAD/EMTDC- based modling and flicker estimation for wind turbines” Europe Premier Wind Energy Event. EWEC, Marseille. 2009. [13] Elshiekh M. E., Mansour D. A. and Azmy A. M. “Improving fault ride-through capability of DFIG-based wind turbine using superconducting fault current limiter” IEEE Trans. on Applied Superconductivity, vol. 23, no. 3, June 2013. [14] Unchim T. and Oonsivilai A., "A Study of wind speed characteristic in PI controller based DFIG wind turbine" World Academy of Science, Engineering and Technology 60, 2011. [15] Sang Ko H., Yoon G., Kyung N. H. and Hong W. P. “Modeling and control of DFIG-based variable-speed wind-turbine” Electric Power Systems Research, no. 78, pp. 1841–1849, 2008. [16] Li H., and Chen Z., "Overview of different wind generator systems and their comparisons," IET Renew. Power Geneerators, vol. 2, no. 2, pp. 123–138, 2008. [17] Jedut L., Rosolowski E., and Rudion K., "Investigation of DFIG based wind turbine influence on the utility grid" Modern electric power systems, Wroclaw, Poland, 2010. [18] Kim Y., Marathe A. and Won D. “ Comparison of various methods to mitigate the flicker level of DFIG in considering the effect of grid conditions” Journal of Power Electronics, Vol. 9, No. 4, July 2009. [19] IEC61000-3-3, 1994, Electromagnetic compatibility (EMC).part 3: limitation of voltage fluctuations and flicker in low-voltage supply system for equipment with rated current 16 A. [20] IEC61000-4-15: ‘Electromagnetic compatibility (EMC). Part 4: testing and measurements techniques – Section 15: flickermeter, functional and design specifications’ (1997, 1st edition).