Study of Dynamic Response of Doubly Fed Induction Generator during Grid Interconnection & Fault Condition

DOI : 10.17577/IJERTV3IS030583

Download Full-Text PDF Cite this Publication

Text Only Version

Study of Dynamic Response of Doubly Fed Induction Generator during Grid Interconnection & Fault Condition

Ajay Kushwaha

Electrical & Instrumentation Engineering. Department, Thapar University,

Patiala, Punjab, India

Inderpreet Singh

Member, IEEE: E.I.E.D.

Thapar University, Patiala, Punjab, India

AbstractIn the recent times, due to environmental concerns, there is rapid growth of wind power in the electric power systems. Power system planners and operators are facing many difficulties while integrating wind power because of its inherent characteristics. To solve these difficulties there is need for various studies and models of wind turbines. Doubly Fed Induction Generator (DFIG) based Wind Turbine (WT) is one of the most popular configurations being adapted. In this paper a dynamic model of DFIG based WT is developed in EMTDC/PSCAD and the results of various dynamic studies have been presented.

Keywords Doubly fed induction generator, Grid Side Converter (GSC), RSC (Rotor Side Converter), Crowbar protection.

output in synchronous frame. Apart from this crowbar protection circuit is also added so as to control over-current phenomenon in Rotor Side Converter which may arise in fault conditions.

  1. COMPONENTS AND OPERATING MODES OF DFIG

    The main components of DFIG are wound rotor induction generator, aerodynamic system with gear box, Rotor side converter, Grid side converter, coupling transformer, filter & protection system. The rotor shaft is connected to drive train system of WT and stator terminal is connected to grid.

    1. INTRODUCTION

      Due to environmental concerns caused by excessive exploitation of conventional resources, now the focus is diverted to non renewable resources especially solar & wind as these are environmentally clean and eco-friendly. With the modern technology incorporated in the wind turbines, wind power generation limits have been uplifted to considerable level in the grid. Hence penetration level of wind power has become more significant and is leading to more complex, sophisticated and reliable interconnection requirements.

      According to the requirements, dynamic behavior of grid should not get affected by operation of wind farm. But when grid is attributed to fault and voltage dips, the disconnection of the wind farm creates shedding of loads resulting in unreliable power supply. Therefore according to the magnitude of voltage at point of interconnection, the fault ride through capability is specified to withstand voltage dips without load shedding. Some more areas in which constraints need to be incorporated in study are power quality problems, protection of hardware equipments and ancillary services [1]. To study these issues, dynamic model of WT has been developed. For the present study DFIG WT is considered. This WT is connected to grid through step up transformer. The Grid Side Converter and Rotor Side Converter are connected back to back to control generator output parameters in both normal & abnormal conditions. The Rotor Side Converter is current controlled & Grid Side Converter is voltage controlled. Both control schemes are based on per unit system. In these schemes, stationary frame is converted to synchronous frame and vice-versa. In synchronous frame, steady state or DC values are compared to desired references to achieve required

      Fig. 1 Components of DFIG-based WECS [2]

      The rotor terminals are connected to back to back dc link connected converter system. It is capable to transmit 25-30% of rated power supplied to grid. Protection system provides safe operation of converters. The converter system is bidirectional so DFIG can be operated in sub-synchronous and super-synchronous mode.

  2. MODELLING AND CONTROL SCHEMES OF RSC & GSC

    1. Wind park modelling:

      The terms which relate wind speed to developed power are as follows:

      1. Power in air flow:

        Pair=0.5A3 also CP = P wind turbine / Pair , on substituting we get P wind turbine = 0.5A3 CP

      2. Tip speed ratio: =r/

      3. Beta Limit:

        The maximum power extraction limit of wind turbine is 59.3% of wind power. Here the value of CP is 0.28 in model. [3]

        Fig. 2 Model of Wind Park

    2. Current Reference PWM control RSC

      V* =V = R I + L dI /dt

      qr qr r qr r qr

      Fig. 3 Block diagram of rotor side converter control scheme [6]

      dr

      qr

      In the model, the current reference PWM control scheme is applied. Therefore rotor current is split into two orthogonal

      The DFIM is controlled in synchronous dq reference

      components d and q. The voltage V*

      (or Vdr

      ) and (V*

      ) or

      frame with d-axis aligned to rotor flux vector position [4]. The id component refers to magnitude of rotor flux in air gap aligned in direction to stator flux vector; whereas iq component produces quadrature flux vector. The developed torque is vectoring cross product of id & iq vectors, and hence contributes to developed power. The id component controls magnitude of reactive power entering DFIM. The id and iq can be controlled so as to control real and reactive power flow [5]. In stator-flux orientation, the torque and the dq axis voltages, currents and fluxes are related as: [6]

      Vqr of block diagram are replaced & modeled as Ird and Irq. The reactive power controls the voltage magnitude and power factor at the grid within limits, the reactive powers are taken to be input to comparator. They both are compared and after PI controller defines the accurate instantaneous value of Ird., which controls reactive power Qg.

      V* = R I

      + L dI

      /dt –

      L I

      dr r dr

      r dr

      slip r qr

      V* = R I + L dI

      /dt +

      (L I + L I )

      qr r qr

      r qr

      slip

      m ms

      r dr

      Te = -3p Lm ImsIqr /2 slip = e – r

      The slipLrIqr component is neglected from expression of Vdr. By this approximation, rotor excitation current Idr is controlled using Vdr. The d-axis reference is Reactive power reference Qg* in model instead of Idr*. Assuming that all reactive power to the machine is supplied by the stator, Reactive power reference Qg* or Idr* is set to zero. The approximated expression for d- axis reduces to

      Fig. 4 Model of generation of reference currents in RSC

      Here q component of the current is used to regulate the torque. The torque controller in the loop modifies magnitude of electromagnetic torque in accordance to the variation to wind speed and produce reference active power operating

      V* =V

      = R I

      + L dI

      /dt

      point.

      dr dr

      r dr

      r dr

      Similarly the slip(Lm Ims + LrIdr) is neglected from expression of Vqr. By this approximation, the torque is proportional to Iqr and Iqr can be regulated using Vqr.

      The reference q-axis rotor current can be obtained either from outer speed-control loop or from a reference torque imposed on the machine.

      These two options may be termed a speed-control mode or torque-control mode for the generator. In this model, first speed control mode is applied for 0 to 0.5 seconds and then torque control mode is enabled.

      The q-axis reference Wref is considered in model instead of

      Iqr. The approximated expression for d- axis comes to be:

      Fig. 5 Determination of location of stator flux

      The reference torque calculation includes the rotor speed r magnitude and is inferred from MPPT WT characteristics. Further q-axis reference current Irq is decided by manipulating torqu magnitude [3].

      The present location of stator flux (s) or phis is determined in order to convert dq variables from synchronous frame to stationary. The phis s can be determined by transforming stationary stator voltages to rotating axis components or – components using Clarke transformation. The and components are orthogonal and denoted by and , the stator flux angle s is expressed as.

      g = tan- ( Vbeta / Valpha)

      | |

      2 2 ,

      / )

      1

      s

      tan (

      In this model, r is measured by the induction generator itself. The other parameters measured by it are Tm Te and Wp.u.. Now consider rotor is rotating and having instant location at an angle r and reference frame is with respect to rotor, then stator magnetic field vector location is expressed as s- r, which we refer to the slip angle slip .

      Fig. 6 Model of determination of slip angle

      Once the reference currents are determined, the actual current are compared with them and the current-reference pulse width modulation output obtained are used as firing pulses to RSC in order to get desired phase current waveforms.

      Fig. 7 Model of Rotor Side Converter PWM

      In this way the desired Ira, Irb and Irc are synthesized. More the precise waveform of Ira, Irb, Irc are obtained, better will be controlling of reactive power and rotor speed and hence determines the status of stability.

    3. Voltage oriented control GSC control

      Here VOC is implemented at grid side converter system and oriented along synchronous reference frame, all variables are in steady state i.e. DC magnitudes. For realization of VOC, the grid voltage needs to be measured and grid angle g is to be detected. The grid angle g is used for grid voltage transformation from stationary 3 phase to dq synchronous frame and vice versa. Let us consider Va ,Vb ,Vc are 3-phase balanced sinusoidal voltages and Valpha & Vbeta are rotating voltages. Valpha & Vbeta are determined through Clarkes transformation obtained from 3-phase balanced voltages; then grid angle g can be expressed as

      Fig. 8 Block diagram of Grid Side Converter control scheme [6]

      Fig. 9 Determination of stator phase angle in GSC

      The VOC scheme has three feedback control loops; two of them are inner control loops for controlling of dq current components i1d and i1q and third feedback loop is DC voltage loop and this is to control capacitor /DC link voltage Ecap to maintain constant magnitude. In this scheme, 3-phase line currents i1a, i1b, and i1c of stationary frame are transformed to dq synchronous frame and these dq components i1d and i1q are responsible for active and reactive components of 3-phase line currents. These i1d and i1q components are controlled independently so as to control active and reactive power respectively and provide better control of DFIM system.

      In order to achieve VOC control scheme, d-axis of synchronous frame is oriented to grid voltage vector Vdg, or Vdg = Vg and remaining q-axis grid voltage Vqg is equal to zero as inferred by the following expression

      Fig. 10 Determination of dq stator current

      Pg = 3/2 [Vdg * i1d – Vqg * i1q]; Pg = 3/2 [Vdg * i1d]

      Qg = 3/2 [Vqg * i1d – Vdg * i1q] ; Qg= 3/2 [- Vdg * i1q] for Vqg= 0

      Similarly the q-axis reference current iqref* can be expressed as

      iqref* = Qg* / ( -1.5* Vdg )

      Here Qg* stands for reference reactive power, its magnitude is zero due to unity power factor operation. The DC link voltage of capacitor Ecap is kept constant as its reference voltage Ecapref is maintained at constant magnitude. Therefore magnitude of d-axis current generated idg* is in accordance to the operating conditions and it is generated through proper tuning of PI controller. If the losses in the inverter operation is neglected, then DC power of dc link will be equal to the magnitude of active power produced on AC side mathematically, this condition can be expressed as

      Pg = 3/2 [Vdg * i1d] = Ecap * i1d

      Fig. 11 Model of generation of reference dq voltages

      In effect, the Grid side converter is supplying the real power demands of the rotor side converter. It is possible to control the d axis current by controlling the d-component of the SPWM output waveform and the q axis current via the q component. However, this leads to a poor control system response, because attempting to change id also causes iq to change transiently. Hence, modifications have to be made to decoupled response. The PI controller makes the decoupled control design more convenient and makes the system output more accurate, precise and stabilized [7]. If these reference voltages Vdref1 and Vqref1 are applied at the secondary of the transformer, the desired currents idref and iqref will flow in the circuit.

      Fig. 12 Generation of voltage reference

      The remaining parts of the controls are standard PWM controls in which each of the phase voltages is compared with a high frequency triangle wave to determine the firing pulse patterns.

      Fig. 13 Model of generation of firing pulses by PWM

    4. Crowbar Protection

    For the hardware protection in converters, the crowbar protection is used frequently. This is installed between RSC and rotor terminals. These are specially used for overvoltage condition created due to malfunctioning or damage of converters. In this case heavy current is passed through RSC to rotor terminals. The crowbar circuit provides alternate path and reduce the magnitude of over current spikes and bring it to normal magnitudes to provide stable operation in abnormal conditions. The crowbar can also provide LVRT capability

    i.e. eliminating short circuit conditions without disconnecting turbine from grid. [4]

    Fig. 14 Crowbar Protection for RSC

    Ir is rotor current which is compared with threshold value. When it becomes greater than that value, S1 operates & crowbar becomes active.

  3. RESULTS AND DISCUSSIONS

    The single line diagram in the model is as follows

    Fig. 15 Single line diagram of grid interconnection

    It consists of induction generator, Wind Park, grid side & rotor side converter, step up transformer, transmission line and grid. The base MVA is 1 p.u. and the generated voltage is set to 0.69 KV, which is fed to grid via step up transformer. In the model, wind park, converter and induction generator are modeled is based on p.u. values. At the secondary side of transformer, symmetrical fault is modeled.

    The Induction Machine is made to operate on speed control mode from 0 < t < 0.5 second and then mode is changed to torque control mode. Until this time (t=0.5 sec), the machine will rotate at W = 0.55 p.u. At t = 3 second fault occurs, the dynamic variation of active power, reactive power, Mechanical torque and rotor speed is analyzed for stability aspects. Also the induction machine is subjected to the step change in the wind speed at t = 8 second. The induction machine is made to operate in super synchronous mode. (Wref = 1.054 p.u.)

    Fig. 16 Voltage magnitude (in p.u. )

    When Fault occurs, voltage magnitude dips to 0.38 p.u. from 0.69 p.u.

    During grid connectivity, rotor speed dips up to 1.05 p.u. from 1.054 p.u. In case of fault, rotor speed rises and oscillates.

    Fig. 17 Actual & Reference rotor speed (in p.u. )

    The wind speed changes to 10.5 m/sec from 11.5 m/sec. at t = 8 second. The rotor speed Wp.u. decreases as Wref decrease.

    Fig. 18 Actual & Reference rotor speed (in p.u.)

    The stator side voltage and current at the time of prefault, fault & postfault are shown in figure. At prefault condition, the voltage and current waveform maintains outphase relation (180 degree) approximately [8]. So the active power supplied is almost constant. But in fault duration, voltage dips and current spikes, hence active power dips and occurs oscillations in reactive power.

    Fig. 19 Stator voltage & stator current

    At last the powergets decrease due to drop in wind speed.

    Fig. 20 Active power supplied to grid (in p.u.)

    Similarly the reactive power is supplied at the time of fault or grid interconnection.

    Fig. 21 Reactive power supplied to grid (in p.u. )

    After the text edit has been completed, the paper is ready for the template. Duplicate the template file by using the Save As command, and use the naming convention prescribed by your conference for the name of your paper. In this newly created file, highlight all of the contents and import your prepared text file. You are now ready to style your paper; use the scroll down window on the left of the MS Word Formatting toolbar.

    1. Rotor side Converter (Current Reference PWM):

      Fig. 22 Generation of firing pulses (RSC)

      The T3 and T6 are fired in consecutive in order to control the tracking of Irb_ref. Similarly Ira and Irc are also synthesized.

      Fig. 23 Stator current chasing reference current of b-phase

    2. Grid side Converter (Voltage oriented Control)

      As the model is run at super synchronous mode the direction of active power will be RSC to GSC. The magnitude of active power at normal conditions

      Pg= Ecap * i1d =0.8*(-0.772) = -0.62 p.u

      The negative sign indicates that the power is generated not absorbed.

      Fig. 24 Actual & Reference Capacitor voltage of DC link

      Qg = 3/2 [- Vdg * i1q], i1q

      Oscillates are at zero magnitude in graph, so reactive power is just oscillating as that of i1q.

      Fig. 25 q axis current in GSC Scheme

    3. Crowbar Protection

    The rotor current Ir has constant magnitude in normal conditions. At the condition of fault, it spikes to 0.7 p.u. but it is below the threshold value. So the switch S1 remains at off state i.e. at 0.

    Fig. 26 Rotor current & Status of crowbar switch

  4. CONCLUSION

The PSCAD simulation results verify that the control schemes used in the RSC and GSC increases the fault ride through capability of DFIG as compared to conventional protection schemes. The transient characteristics of voltage at faulty condition recovered in a very short time due to adequate supply of reactive power to grid. This ensures better power quality status at the abnormal conditions. Also variation in rotor speed is controlled effectively to ensure stability aspects. The active & reactive power flow is made independent by the decoupling method resulting in better control & performance of DFIG. Hence the operation of hardware protection is not frequent.

REFERENCES

  1. P. Bousseau, F. Fesquet, R. Belhomme, S. Nguefeu, T. C. Thai, Solutions for The Grid Integration of Wind Farms a Survey Wind Energy, vol- 9, pp 13-25, 2006

  2. S. Muller, M. Deicke & Rik W. De Donker, Doubly fed Induction Generator for wind systems, IEEE Industry Applications Magazine, pp 26-33, May-June 2002

  3. O. Aanaya-lara, N. Jenkins, J. Ekanayake, P. Cartwright, M. Hughes Wind Energy Generation Modelling and Control, John Wiley, 2009, pp 4-5,84-85

  4. G. Abad, .J. Lo´pez, M. A. Rodr´guez, L. Marroy, Doubly Fed Induction Machine, Modeling and Control for Wind Energy Generation., John Wiley, 2011, pp 304, 481-483

  5. A. Gole, Vector Controlled Doubly Fed Induction Generator for Wind Applications, Department of Electrical and Computer Engineering,

    University of Manitoba, Canada and Nayak Corporation, Princeton, New Jersey. pp. 1-2, 2-3.

  6. R. Pena, J. C. Clare, G.M. Asher, Doubly fed induction generator using back-to-back PWM conveners and its application to variable speed wind-energy generation, IEEE Proceedings Electric Power Applications, vol. 143, no. 3, pp.31-38. 1996.

  7. Wu, Y. Lang, N. Zargari, S. Kouro, Power Conversion and Control of Wind Energy System, John Wiley 2011, pp17, 144-148.

  8. R. Wang, F. Lin, R. Hao, X. You, T. Q. Zhang,, VSCF Doubly-fed induction generator control strategies and simulation research, Presented at 3rd IEEE Conference on Industrial Electronics and Applications, 2008

Leave a Reply