This action might not be possible to undo. Are you sure you want to continue?

2.1 Introduction

For wind power generation systems, the doubly-fed induction generator (DFIG) currently dominates with its variable wind speed tracking ability, and relatively low cost compared to full-rated converter systems, e.g. permanent magnet synchronous generator (PMSG). However, a significant disadvantage of the DFIG is its vulnerability to grid disturbances because the stator windings are connected to the grid through a transformer and switchgear with only the rotor-side buffered from the grid via a partially rated converter. Therefore, to protect the wind farm from interruptions due to onshore grid faults and wind farm faults, a crowbar protects the induction generator and associated power electronic devices. This protection system is widely used in industrial applications. A major disadvantage of crowbar protection is that the rotor-side converter (RSC) has to be disabled when the crowbar is active and therefore the generator consumes reactive power leading to further deterioration of grid voltage. In line with developing fault ride-through (FRT) requirements, an active crowbar control scheme is proposed [2.1], [2.2] to shorten the time the crowbar is in operation but this does not avoid the reactive power consumption. Researchers have developed a new fault-control strategy [2.3] and a fault-tolerant series grid-side converter (GSC) topology [2.4]. However, these make the control systems complex or increase the issues with control coordination between normal and fault operation. A series resistor can share the rotor circuit voltage and hence limit the rotor current during the fault, and is an alternative to crowbar protection. However, to the authors

34

Chapter 2

Doubly-Fed Induction Generator Fault Protection Schemes

35

knowledge, there has been no published literature on such a series resistor-based protection scheme. Therefore, the research in this chapter assesses series protection for effective turbine and converter protection during various fault conditions. The chapter is organised as follows. In Section 2.2, existing protection schemes for DFIG systems are summarised. Then, a protection scheme with series dynamic resistor (SDR) connected to the rotor winding is proposed. The faults that can occur in wind farms and the currents in the rotor windings of DFIGs are discussed in detail as the basis of the converter protection scheme design: fault rotor current expressions are given theoretically and with simulation results; and the difference between rotor current characteristics for symmetrical and asymmetrical faults is discussed which highlights the advantage of series dynamic resistors as the primary protection of the converter. In Section 2.4, a new converter protection scheme combining the series dynamic resistor and the crowbar is introduced. Analysis and discussion of PSCAD/EMTDC simulations are provided in Section 2.3 and 2.5.

2.2

2.2.1

**Converter Protection Schemes for DFIG
**

Crowbar Protection

The prevalent DFIG protection scheme is crowbar protection. A crowbar is a set of resistors that are connected in parallel with the rotor winding on occurrence of an interruption, bypassing the rotor-side converter. The active crowbar control scheme connects the crowbar resistance when necessary and disables it to resume DFIG control. For active crowbar control schemes, the control signals are activated by the rotor-side converter devices [which are usually insulated-gate bipolar transistors (IGBTs)]. These have voltage and current limits that must not be exceeded. Therefore, the rotor-side converter voltages and currents are the critical regulation references. The DC-link bus voltage can increase rapidly under these conditions, so it is also used as a monitored variable for crowbar triggering. Bi-directional thyristors [2.5], gate turn-off thyristors (GTOs) [2.2], [2.6] or IGBTs [2.7] are typically used for crowbar switching.

limiting the current reduces the charging current of the DC-link capacitor. which has been used in the stator side of generators. Being controlled by a power electronic switch. the rotor-side converter does not need to be inhibited during the fault. but also limits the high rotor current. In addition. The latter are shunt-connected and control the voltage while the series dynamic resistor has the distinct advantage of controlling the current magnitude directly. It is also used as protection for the DC-link capacitor in full-rated converter topologies. The adverse impact of temporarily losing rotor-side control of a DFIG in a small-scale wind farm can be tolerated since it only involves a small amount of reactive power consumption which is not presently the case for large-scale offshore wind farms.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 36 2. the induced overvoltage may not lead to the loss of converter control.9]. The difference between the series dynamic resistor and the crowbar or DC-link braking resistor is its topology. in normal operation. Moreover. therefore.3 Series Dynamic Resistor In a similar way to the series dynamic braking resistor [2.2. but this has no effect on the rotor current. The series topology is straightforward enough to limit the overcurrent and share overvoltage but there appears to be no literature investigating its use. a dynamic resistor is proposed to be switched in series with the rotor (series dynamic resistor) and this limits the rotor overcurrent. it not only controls the rotor overvoltage which could cause the rotor-side converter to lose control. with the series dynamic resistor. with the series dynamic resistor.2. . the switch is on and the resistor is bypassed. 2. the switch is off and the resistor is connected in series to the rotor winding. for example.2] and [2.8]. This protects the IGBTs from overvoltage and can dissipate energy. The crowbar is adequate for protection of the wind turbine system during grid faults in onshore developments. Therefore. a braking resistor (DC-chopper) is connected in parallel with the DC-link capacitor to limit the overcharge during low grid voltage. based on PMSGs [2.10]. Hence.2 DC-Chopper In [2. during fault conditions. the high voltage will be shared by the resistance because of the series topology. which helps avoid DC-link overvoltage.

4) r where vs is imposed by the grid.5) becomes d dt vr vro Rr Lr j r ir . (2.6) Then.4) is can be eliminated to obtain an expression.2) (2.e. The rotor voltage vr is controlled by the rotor-side converter and used to perform generator control.11] vs vr Rr ir s Rs is d r dt Ls is Lm is d s dt j Lm ir Lr ir r r (2. multiply both sides by e j rt ) .3) (2.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 37 To demonstrate the protection schemes and their interaction with the rotor circuit.8) The rotor voltage in (2. (2.5) Defining the leakage factor as 1 (2.3) and (2.1) (2. Ls Lr d dt j r ir . the rotor equivalent circuit is described first with the general Parks model of induction generators. From (2. eliminating L2 m Ls Lr r vr Lm d Ls dt j r s Rr Lr 1 L2 m . using a voltage source vro to represent the voltage due to the stator flux such that vro Lm d Ls dt j .7) r s (2.8) can be expressed in a rotor reference frame (i. (2. From the voltage and flux equations of induction generators in a static stator-oriented reference frame [2.

DC-link ShuntResistor ShuntResistor ir r + Rotor v r r Rr Lr r vro Bi-directional Bypass Switch + RSC DC-Chopper Crowbar Series-Resistor Series Dynamic Resistor Figure 2.1: DFIG rotor equivalent circuit with all protection schemes shown. However.13].12]. this is important for the design of DFIG protection systems. The approximate maximum stator fault current expression was also discussed from the analysis of DFIG physical response with crowbar protection [2. the rotor equivalent circuit is obtained and shown with all the above protection schemes in Figure 2. In this chapter.11] and [2.5].1. Therefore.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 38 vrr r vro Rr ir r Lr dir r . dt (2. exact expressions of stator and rotor currents during the short-circuit are derived mathematically.9) This is the relationship between rotor voltage and current. there has been no analysis of fault currents during less serious voltage dips or asymmetrical disturbances. Nonetheless. the rotor current expressions during various fault conditions will be deduced on the basis of the analysis of [2.10) This can be written as a linear differential equation for ira(t) . 2. The phase-a rotor voltage expression is vra (t ) r Re{vro } Rr ira (t ) Lr dira (t ) . In [2.3 DFIG Rotor Currents during Fault Conditions DFIG rotor currents under three-phase short-circuit faults have been thoroughly analysed. dt (2.

with the converter in operation. (2. vra(t) = Vrcos(s st + ). which is very small because of the small stator resistance of the generator. the phase-a rotor voltage angle at the instant the fault occurs. and s is the slip.11] L (1 p )Vs m se js Ls Lm Ls 1 s v r ro st j r pVs e j s t s .12) With time constants defined as Lr . vro in (2.13) Equation (2. The voltage is [2.9) can exceed the maximum voltage that the rotor converter can generate.11) and (2.14).3. therefore Lm s (1 p )e js Ls t st r vro Vs 1 s pe j rt e s .12) can be simplified by omitting 1/ s.1 Symmetrical Fault Conditions For a symmetrical voltage disturbance on the stator side. is 2. Rs r s s r r s . which causes current control to fail. if there is a three-phase step r amplitude change from Vs to (1p)Vs (p is the voltage dip ratio).14) From (2.15) . (2. (2.11) where. the final expression of ira(t) can be solved and divided into four components ira (t ) iDC where the components are ivr ivrf ivrn (2.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 39 dira (t ) dt Rr ira (t ) Lr 1 r vra (t ) Re{vro } Lr (2. Rr Ls .

the stator voltage is divided into three parts: positive-. vro in (2.16) ivr Vr Lr 1 r 2 r 2 r cos(s s t ) 2 r r 2 r 2 r 1 sin(s st ) (2.21) (2. (2. and zero-sequence components.18) 2 t r 2 ivrn 1 2 2 r cos( r t ) 1 sin( r t ) e 2 r s .2 Asymmetrical Fault Conditions For asymmetrical faults.22) where vrr1 Vs1 st .13] vs Vs1e j st Vs 2 e j st Vs 0 (2.20) r Then.17) ivrf 1 L Vs m s (1 Lr Ls Vs Lm (1 s ) p Lr Ls p) r 1 2 r s 2 s cos( s st ) 1 2 r 2 r s s s 2 s sin( s s t) (2.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 40 iDC ira (t 0 ) 1 Lr 1 r 2 r s 2 s Vr cos Vs Lm s(1 p ) Ls 1 L Vs m (1 s ) p Lr Ls 1 t 2 2 r e r (2.1 with the frequency and time constant characteristics.9) can also be expressed as r vro vrr1 vrr2 Lm js se Ls r vrn (2. using symmetrical component theory [2.3.1: Symmetrical Fault Rotor Current Components Frequency DC s s s s r Component iDC ivr ivrf ivrn Decaying time constant r s 2. negative-.19) The components are listed in Table 2. Table 2.

23) t s r vrn j r e j rt (2. The positive-. t s (2.27) From the natural flux initial value analysis in [2.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 41 vrr2 Vs 2 Lm (s 2)e Ls Lm Ls n0 e j (2 s) st (2.25) where p is the phase-a voltage dip ratio due to the fault. Vs 0 . Therefore. and 1) Single-Phase Voltage Dip: n0 depend on the type of fault. and zero-sequence components of the stator voltage are Vs1 Vs (1 p / 3) Vs 2 Vs 0 Vs ( p / 3) Vs ( p / 3) (2. Phase a suffers a voltage dip.26) vrr2 Vs ( p / 3) Lm (s 2)e Ls j(2 s) st (2. negative-.30) . Vs 2 . the aforementioned vrr0 components are Lm js se Ls vrr1 Vs (1 p / 3) st (2.29) hence vro Vs Lm s 1 p e js Ls 3 st 2 L j Vs m (1 s ) pe 3 Ls e j rt (2.28) v r rn 2 L j Vs m (1 s ) pe 3 Ls L p Vs m (s 2) e Ls 3 j ( 2 s) st e j rt .24) The components Vs1 .13] Vs (2 / 3) p n0 s t s (2.

34) ivr 2 L 1 p Vs m ( s 2) Lr Ls 3 2 Vs Lm (1 s ) p 3 Lr Ls r 1 2 2 r (2 s ) 2 2 s cos (2 s) st 2 r 1 2 r ( 2 s) s (2 s ) 2 t 2 s sin ( 2 s) st (2.32) e r 1 L p Vs m ( s 2) Lr Ls 31 Vr Lr 1 r 2 r (2 s) 2 r 2 2 s 2 (1 s ) p 3 1 ivr 2 r cos(s s t ) r 2 r 2 r 1 sin(s s t ) (2.37) where p is the phase b and c voltage dip ratio due to the fault.36) 2) Phase-to-Phase Fault: Here. the final expression of ira(t) can be solved and divided into five components ira (t ) where the components are 1 Lr 1 Lm s 1 Ls p 3 2 r 2 2 r t iDC ivr ivr1 ivr 2 ivrn (2.33) ivr1 Vs Lm s 1 Lr Ls p 3 r 1 2 r s 2 s cos( s s t ) 1 2 r 2 r s s s 2 s sin( s s t) (2. (2. the initial value of natural flux is [2. and zero-sequence components of the stator voltage are Vs1 Vs (1 p / 2) Vs 2 Vs 0 Vs ( p / 2) Vs ( p / 2) (2.30).11) and (2.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 42 From (2. phases b and c are shorted together leading to a voltage dip at the stator terminals. Then the positive-. negative-.13] .35) ivrn r 1 2 2 r cos( r t ) 1 2 2 r sin( r t ) e s .31) iDC ira (t 0 ) r 2 r s 2 s Vr cos r 2 r Vs (2. Also.

41) (2.39) e r (2 s) 2 2 s (1 s ) p 1 ivr Vr Lr 1 r 2 r 2 r cos(s st ) 2 r r 2 r 2 r 1 sin(s st ) (2. (2. The components are listed in Table 2.42) (2. but different amplitudes. The induction generator parameters are shown in Table 2.2: Asymmetrical Fault Rotor Current Components Frequency DC s s s s s Component iDC ivr ivr1 ivr2 ivrn Decaying time constant r s (2s) r The rotor currents during the fault are simulated in PSCAD/EMTDC to compare with the analysis. and the rotor-side converter is controlled using a voltage-regulating vector controller. The simulations have the rotor-side converter connected when faults occur. .Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 43 Vs p n0 s . with the same five components. The components are 1 Lr 1 2) p 21 Lm s 1 Ls 2 r 2 r iDC ira (t 0 ) L 1 Vs m ( s Lr Ls r 2 r s 2 s Vr cos Vs p 2 t r 2 r 2 (2.40) ivr1 Vs Lm s 1 Lr Ls p 2 r 1 2 r s 2 s cos( s s t) 1 2 r 2 r s s s 2 s sin( s 2 r s t) (2.38) The current expression.3.2. is similar to the single-phase fault case. in this case. as shown in Figure 2.43) ivr 2 L 1 p Vs m ( s 2) Lr Ls 2 r 1 2 2 r (2 s ) 2 2 s cos (2 s ) s t 1 2 r (2 s) s (2 s) 2 t s 2 s sin ( 2 s ) s t ivrn 1 L Vs m (1 s ) p Lr Ls r 1 2 2 r cos( rt) 1 2 2 r sin( rt) e . Table 2.2 with the frequency and time constants.

for the less serious voltage dip and asymmetrical .05 0.35 0.0 p.0055 p.2 0.4 0. 0.2(a) and Figure 2.5 (c) 0 -5 0 0..0050 p.5 10 Simulation Theoretical 5 (d) 0 -5 0 0.15 0.1 0. (c) single-phase (phase a) voltage dip of 1.3 0.105 p.35 0.3 0.5 (b) 0 -2 -4 0 10 Simulation Theoretical 5 0. 3.100 p. voltage dip.953 p. The three-phase short-circuit fault causes an abrupt change at the moment the fault with highest peak values [Figure 2.3] Value 2 MW 690 V 50 Hz 0.63 3.45 0.25 time (s) 0. 44 Parameter Rated power Pn Rated stator voltage Vsn Rated frequency fs Stator leakage inductance Lls Rotor leakage inductance Llr Magnetizing inductance Lm 6 4 2 Simulation Theoretical (a) 0 -2 -4 -6 0 6 4 2 Simulation Theoretical 0.4 0. (d) phase-to-phase (phase b to c) short circuit. Each fault displays different frequency components and characteristics. 0.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes Table 2.45 0.45 0.05 0.u.1 0.2 0.1 0.05 0.15 0.u.u.15 0. Pp Stator resistance Rs Rotor resistance Rr Value 0.2 0.5 Figure 2.15 0.u. (b) three-phase 0.u.u.2: Comparison of simulation and theoretical rotor currents during fault conditions (for 0.3 0.5 s): (a) three-phase 1.5 s 2 0.4 0. However. Parameter Ratio Ns/Nr Inertia constant H Pole pair no.35 0.u.25 0.2 0.4 0. voltage dip.3: Induction Generator Parameters [2.0 p.05 0.u.25 0.3(a)].25 0.1 0.35 0.45 0.6 p.3 0.3(a)] but with relatively short duration [see Figure 2.

5 s): (a) three-phase 1.0 p. All three-phase rotor currents are shown in Figure 2. Therefore it will contribute to the converter protection scheme design in Section 2.6 p. the high magnitude. 2. (b) three-phase 0.4. the converter has to be inhibited and then the DFIG absorbs reactive power from the grid. which adversely affects grid recovery. The comparisons show that the analysis is in accordance with theory and is valid for the study of the fault conditions. The crowbar need only switch on for a short time. voltage dip. For asymmetrical dips.2(b) (d)].0 p. The same simulation system will also be used for the protection scheme verification that follows. (a) (b) (c) Figure 2.u. the rotor currents increase abruptly both at the beginning and the end of the fault.u.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 45 faults [see Figure 2. (d) phase-to-phase (phase b to c) short circuit. (c) single-phase (phase a) 1.u. For symmetrical faults.3. To protect the system.4 Protection Scheme Based on Series Dynamic Resistor The above rotor fault current analysis and simulation highlights a major difference between symmetrical and asymmetrical fault currents. the crowbar does not solve the problem because it needs to be active throughout the duration of . voltage dip. voltage dip. high-frequency oscillation makes it is impossible to switch off the crowbar protection.3: (d) Three-phase rotor currents during different fault conditions (for 0.

Therefore. i. 2.13]. the switch off is delayed for a period of the high frequency component. A series dynamic resistor is used as the primary protection. To avoid the protections switching frequently because of the high-frequency component of rotor current. only rotor currents are monitored for series dynamic resistor and crowbar protections. The DC-chopper is used for DC-link overvoltage limitation. with the crowbar circuit used if the series dynamic resistor cannot protect because of a deteriorating situation. 2) Protection Disengaged: The protections themselves can be seen as disturbances.e. 1) Protection Engaged: The voltage change is not as abrupt as the current and can be shared by the series dynamic resistor. its change can be further reduced by the DC-chopper. The final switching strategy is shown in Figure 2.1 Switching Strategy It is observed in the previous section that asymmetrical faults are more hazardous than symmetrical faults for the DFIG because of the continuous overcurrent in the rotor.4.4. This can be explained by the difference in flux components for different faults [2. a new protection scheme based on a series dynamic resistor is proposed which also combines and coordinates the existing crowbar and DC-chopper protection. The crowbar is engaged only at the beginning or the end of the fault.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 46 the dip. t_delay = 2 /(1s) currents decrease below the threshold value. In this section. From the above overcurrent analysis a switching strategy is devised to determine when to engage the protection measures using current thresholds. s 2 / s after all the three-phase . if required. requiring the generator to be disconnected from the grid. For the DC-link voltage.

Then.15) are made: e (2. abc Ith_CB > Crowbar OFF Rotor-Side Converter ON Series Dynamic Resistor OFF AND t_delay = 2 /(1s) s Timer vDC Vth_DC < DC-Chopper ON DC-Chopper OFF Figure 2. 2.19).2 Series Dynamic Resistance Calculations Resistance values are calculated for the most serious condition (with the highest peak current value): symmetrical voltage dip up to 1.45) ivrf 0 r 2 r 2 r (2.46) sin( r t ) ivrn L 1 Vs m (1 s) Lr Ls 1 (2. the current components are expressed as a single trigonometric function as 1 Lr 1 L 1 V s m (1 s ) Lr Ls 1 t 2 2 r i DC i ra (t 0 ) r 2 r s V cos 2 r s e r (2. CB Crowbar.4: Combined converter protection switching strategy (for subscripts: th threshold values.44) ivr Vr Lr 1 r 2 r 2 r sin(s st ) (2.u. Due to the small stator resistance. abc Ith_SDR 47 > Series Dynamic Resistor ON AND Crowbar ON Rotor-Side Converter OFF ir.47) where tan 1 1 r r .4. SDR Series Dynamic Resistor). the following approximations t/ s 1. . The rotor current expressions are (2.0 p. r.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes ir.

u. r (2. for its shunt connection: RCB ir. With the Therefore. where the resistance has a lower and upper limit.max ira (t0 ) r 2 r r r 2 r 2 r 2 r .max Vr. . includes RSDR and RCB.2 s. Vr Vth_RSC.5 Simulation Results The proposed converter protection method is verified by PSCAD/EMTDC simulations.48) and (2. The current-limiting function is provided by the series dynamic resistor. the crowbar resistance is a small contribution to the total Rprotection. If the rotor fault currents still cannot be limited effectively. the boundary conditions are ira.49) can be solved.max.49) are equations where protection schemes Lr . for 0. (2. Therefore. 2) a single-phase (phase a) grounding for 0.max Ith_SDR. 2.48) 1 2 r 2 r Also. R protection r Rr (2. The generator parameters are listed in Table 2.5]. the critical resistance value Rprotection can be calculated. (2.2 s. the crowbar can be used as further protection. This is simpler than using crowbar protection alone. The total resistance is Rprotection.50) Then. hence the critical criterion of crowbar resistance is the voltage across it must be within the rotor voltage limit. The faults simulated are: 1) a three-phase voltage dip of 0.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 48 Considering the amplitude of each component at the maximum current value L 1 V m (1 s ) Lr s Ls 1 Vr Lr L 1 V m (1 s) Lr s Ls 1 ira .95 p. The minimum value of resistance is restricted by the rotor winding current limit. while the maximum is set by the voltage limit at the converter terminals [2.3.

0 p.u.6 show the system response to a 0.5 to Figure 2. the first torque peak is safely avoided.6. preceding the faults. Rotor slip is s = 0.5.65 ms. The rotor currents reach around 10. a 7. for the most serious phase.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 49 3) a two-phase short-circuit (phase b to c) for 0. voltage dip for 0. Then. in Figure 2. The individual crowbar and SDR torque performances will be compared in Section 2.2 p.2 s.u.u. This is due to the crowbar protection switching in as a further protection measure. and 4) a three-phase voltage dip of 0. while crowbar is helpful for protection during fault recovery. The simulation results show that with series dynamic resistor protection. Rprotection = 0. crowbar is switched in for five times with the series dynamic resistor connected as the rotor current increases beyond the crowbar threshold. This also restricts the DC-link voltage increase (less than 0.u.6).2 p. The threshold values for calculating RSDR and RCB are set as Ith_SDR = 1. r = 0. The rotor speed increase is effectively restrained from 1.5.22 p.6.5 .09 . The rotor current amplitude is limited within 1. . Large electrical torque fluctuations occur. for 1.1 Symmetrical Fault Condition Figures 2. series dynamic resistor is switched in ten times in total to limit the rotor current. However.0p. as required.u. The DC-chopper function is not required.05 p. DC-link voltage and rotor speed both increase until the fault is cleared. = 0.95 p.u.48) and (2.2 s with and without protection respectively.59 .207 p.3 which . The large 5.6 with the series dynamic resistor.5 p.u. Ith_CB = 1.8 p.6 p.u. torque fluctuation at the start of the fault is avoided. compared to 1. In the simulation without protection. the rotor-side converter is blocked during the fault.49).u.u. The value of DC-chopper resistance is not so critical as it is only related to the DC-link voltage. so here choose 2.5 and 2.u.u. In Figure 2..u.u. to 1. without protection.0 s. RCB = 0.0 p.. From (2. During the recovery of the fault. torque fluctuation occurs during the fault recovery phase in Figure 2. the selected resistance values are RSDR = 0. compare Figure 2.987 p.5 p.5 RDCC = 0.

).u.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 50 shows that all of the 7.)].). (g) rotor speed r (p.10 (d) and (e)]. Tm is in blue and Te is in green and that in Figure 2.).u. Tm is in green while Te is in blue. voltage dip for 0.0 p.u.) and phase-a RSC voltage vrsc.u. (e) DC-link voltage vDC (p.u.u. (c) three-phase rotor currents ir a.2 s without protection: (a) three-phase stator voltages vs a.5: Three-phase 0.u. (b) three-phase stator currents is a.u.a (p. (h) electrical torque Te (p.u.).6.c (p.b.c (p. (d) phase-a rotor voltage vra (p. Note that in Figure 2. .u.b.u.u.c [in per unit (p.5.u.) and mechanical torque Tm (p.).) and reactive power Qs (p.). (f) stator side active power Ps (p.).95 p. (a) (b) (c) (d) (e) (f) (g) (h) Figure 2. torque pulsation that occurs at fault recovery is due to the crowbar circuit [see Figure 2.b.

) and phase-a RSC voltage vRSC.).c (p.u.c (p.).)]. (b) three-phase stator currents is a.b. (j) rotor speed r (p.).) and mechanical torque Tm (p.u. voltage dip for 0. (h) DC-link voltage vDC (p.u.6: Three-phase 0. (f) DC-chopper switching signal SDCC.).u. (i) stator side active power Ps (p.a (p.c [in per unit (p. (g) phase-a rotor voltage vra (p.).2 s with converter protection: (a) three-phase stator voltages vs a.) and reactive power Qs (p. (c) three-phase rotor currents ir a.u.95 p.u.b.b. .u.u.u.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 51 (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) Figure 2.u. (e) crowbar switching signal SCB.u.u.). (d) SDR switching signal SSDR.). (k) electrical torque Te (p.

However. Instead of increasing. The rotor voltages display switching frequency components due to the pulse-width modulation of the rotor-side converter.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 52 Although there is no rotor voltage monitoring in the switching strategy.8.8 and 2. Large transients occur during the fault clearing mainly due to the impact of crowbar protection switching. the disturbances are clamped after about 0.u. Figure 2. it is still limited effectively to the value before the fault because of the voltage sharing ability of the series dynamic resistor. It should be noted that whilst the crowbar is used in this particular case.05 s. The rotor currents are also limited within 1.u. The high voltage is shared across the series resistor and the converter which results in a lower converter side voltage (vRSC. the series dynamic resistor and crowbar protection switching events are similar to the symmetrical fault conditions.a (p.a in Figure 2. the rotor speed decreases because the DFIG is still under control with active power supplied to the grid.2 Asymmetrical Fault Conditions Figures 2.1 p. there is one period of DC-chopper switching because of the gradual increase of DC-link voltage to 1.u.9 show the system responses during asymmetrical fault conditions.)] and rotor-side converter voltage vRSC. it is not necessary under all faults.5.7).) comparison (zoomed from 1 s to 1.5 p.7: The rotor voltage vra [in per unit (p. For the phase-a fault in Figure 2. An overspeed condition is avoided as the electrical torque balances the mechanical torque from the wind turbines blade system. but together with series dynamic resistor protection. 2.u.1 s). .

u.u.b.) and reactive power Qs (p. (e) crowbar switching signal SCB. (j) rotor speed r (p.u.c (p.u.) and phase-a RSC voltage vRSC. (k) electrical torque Te (p. (f) DC-chopper switching signal SDCC. voltage dip for 0.).8: Phase-a 1.).u.u.u.u.).a (p. (b) three-phase stator currents is a.u.)]. (c) three-phase rotor currents ir a.).). .b.u.b.) and mechanical torque Tm (p. (i) stator side active power Ps (p.).0 p. (g) phase-a rotor voltage vra (p.u.).2 s with converter protection: (a) three-phase stator voltages vs a.c [in per unit (p. (d) SDR switching signal SSDR.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 53 (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) Figure 2. (h) DC-link voltage vDC (p.c (p.u.

u.).9: Phase b to c short circuit for 0. (h) DC-link voltage vDC (p. (d) SDR switching signal SSDR.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 54 (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) Figure 2. . (c) three-phase rotor currents ir a. (k) electrical torque Te (p. (g) phase-a rotor voltage vra (p.2 s with converter protection: (a) three-phase stator voltages vs a.u.u.c (p.c (p. (f) DC-chopper switching signal SDCC.).) and reactive power Qs (p.).b. (b) three-phase stator currents is a.)].).). (e) crowbar switching signal SCB.) and phase-a RSC voltage vRSC.c [in per unit (p.u.u.b. (i) stator side active power Ps (p.u.b.a (p.).u.u.u. (j) rotor speed r (p.).) and mechanical torque Tm (p.u.u.

This is due to the higher resistance in the rotor winding and DFIG control system performance during faults. .9. Both of the two strategies experience reactive power and electrical torque fluctuations during the fault.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 55 The phase b to c short-circuit in Figure 2. But in terms of stator voltage. the series dynamic resistor protection has a significant advantage. There is no need for both crowbar and DC-chopper operation. Although most of the variables are under control.u. the asynchronous machine absorbs reactive power.5. The reactive power and electrical torque ripples are larger with series dynamic resistor protection compared to crowbar protection.10(b) is expanded to show the reactive power. 2.21 p. The reactive power. different at the peak prior to recovery. which needs further exploration. for crowbar protection.6 when the crowbar is engaged. from 1.3 Performance Comparison Between Crowbar and SDR The performance of the crowbar and the series dynamic resistor protection schemes are compared. electrical torque and rotor speed of the DFIG system are simulated and compared in Figure 2. no reactive power is absorbed. Figure 2. The two asymmetrical conditions result in fluctuations after stator voltage recovery. in terms of fault current. It can be seen that with the rotor-side converter connected with the series dynamic resistor protection scheme. For rotor speed changes they are about 0. However. This leads to the large torque fluctuation seen in Figure 2. to 1. up to 0.2 p.02 p. This results in gradual increase of rotor speed. is less serious than in the single-phase case. it is clear that the peak torque that occurs at crowbar turn-on and turn-off is significantly higher than that for the series dynamic resistor.u.10. these fluctuations should be studied in more detail. There are much larger power and electrical torque fluctuations during the fault. for crowbar protection.u. However. However. Therefore.u.20 p. this is more serious than for a single-phase fault. The series dynamic resistor reduces the rotor overspeed more effectively than the crowbar circuit. The series dynamic resistor is effective in this condition. but this is not serious. they are much larger. as it doest not further contribute to voltage drop in the network due to reactive power. in terms of grid voltage recovery.

5 0.). especially during switching off.5 1 1.).u.2 0.6 0.3 1.4 -0.2 1 1.)]. (p.6 0.9.5 -1 -1.22 1.10: System response comparison between crowbar and series dynamic resistor protections.1]) can cause frequent switching which affects fault recovery.5 1 1.4 0. Without crowbar switching. Improper crowbar switch-off strategy (without the coordination of controller reference setting [2.5 (e) 0 -0.1 SDR CB (b) 0 -0.24 1.) with CB protection.9 2 1.) with SDR protection.26 CB SDR (c) 1.6 p. the series dynamic resistor has a much smaller impact than the crowbar.5 1 0. This can also be seen from the comparison of voltage recovery in Figures 2. the voltage recovery for the two-phase short-circuit shows minimal fluctuation. . More importantly.7 1.2 -0.5 1 1.u.8 1.2 SDR CB (a) 0 -0.1 -0.3 1.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 56 0.u.5 1.6 1. (d) electrical torque Te (p. (c) rotor speed r electrical torque Te (p.5 1 0 (d) -1 -2 -3 -4 0.5 1 1.u. (b) zoomed reactive power Qs (p.5 2 2.8 and 2.5 2 2.5 2 2.5 time (s) 2 2. for 2 s: (a) stator-side reactive power Qs [in per unit (p.5 0.4 1.2 1.1 1.2 0.u.28 1.5 Figure 2. (e) voltage dip of 0.u.

1 Application Discussions Switch Time of the Bypass Switch In practical applications. switching times of the crowbar and series-dynamic-resistor power electronic switches are considered by disabling the interpolation in PSCAD/EMTDC. 2) to maximise the operation time of the rotor-side converter. In the above simulations.6. which is enough for the IGBTs in applications (commonly several microseconds [2. But compared to the stator side braking resistor bypass-switches [2.2]. this is far lower due to the lower power rating on the rotor side. various resistor protection schemes are reviewed. 2. so the actual switch time for IGBT is 20 s. In this chapter. However.7 Conclusion Converter protection is necessary for DFIG wind power generation systems during fault conditions.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 57 2. the switch time may be an issue.6. This will influence the protection performance. The rotor currents during various fault conditions are discussed and current expressions are given to instruct the design of the protection scheme. .14]). 2. e. specifically device ON-state losses.6 2. and 3) to reduce torque fluctuations during protection operation. Resistance calculations for the series dynamic resistor and crowbar using the expression of maximum rotor current are described. The purposes of a series dynamic resistor are: 1) to avoid the frequent use of crowbar short-circuit. the bypass switch that is closed during normal operation will produce additional losses.10]. some crowbar thyristor switches cannot interrupt the current before zero-crossing [2. This solves the conflict between immediate switching operation with simulation time step.2 Switch Normal Operation Losses The series dynamic resistor is here realised by a power electronic switch. The simulation time step is set as 20 s.. especially for serious fault protection and recovery when fast switching response is required.g.

. the resumption time can be shorter than for a system with normal active crowbar protection. In the next chapter. This is helpful for resuming normal control and provides reactive power for grid voltage support. electrical torque. the protection for another popular wind power generation system based on PMSG is investigated. During this process. inspection of the reactive power.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 58 The series dynamic resistor can operate with the rotor-side converter control functioning. and rotor speed fluctuations shows that the proposed method enhances DFIG fault ride-through capability. For the control of the grid-side converter to DC-link bus voltage.

Li. no. M. Causebrook. 652-662. Power Convers. no. no. and S. Rodríguez.. 3. 966-975. Sarasola. Erlich. Kretschmann. vol. Dresden. pp. Int. Flannery and G. 2006. Energy Convers. Xiang.. Energy Convers. Jack. 3. H. Electrical Machines and System. 3. Sep. Yang. Sep. in Proc. Wrede. A. M. Conroy and R. J. A fault tolerant doubly fed induction generator wind turbine using a parallel grid side rectifier and series grid side converter.10] [2. Morren. López. IEEE Trans. X. Sep. Dynamic behavior of DFIG-based wind turbines during grid faults. Conf.9] [2. P.1] References J. 2007. Aug. 8-11. vol. Control strategy of an active crowbar for DFIG based wind turbine under grid voltage dips. pp. 2007. vol. and H. 23. Energy Convers. vol. and C. I. pp. 2007. 1. P. F. Mueller-Engelhardt. Wrede. no. J. Korea. 11-14. Apr. 182-189. 1126-1135.8 [2. J.4] [2. J. vol. Sanchis. Power Syst. Japan. A. I. 22. 3. 435-441. Venkataramanan. 709-717. Atkinson. J. 2007. and A. Low-voltage ride-through of a full converter wind turbine with permanent magnet generator. de Haan. H. S. Erlich. vol. W. Abad. J. 22. 20. S. S. Short-circuit current of wind turbines with doubly fed induction generator. Tavner. 2007. J.. Watson..H. Zhou and Y. Marroyo. R. Morren and S. in Proc... and A. 2005. I. pp. D. 2005.11] [2. Power Electron.2] [2. D. in Proc. Tegopoulos. 1. Ridethrough of wind turbines with doubly-fed induction generator during a voltage dip.. 2.7] [2. no. Control of a doubly fed induction generator in a wind turbine during grid fault ride-through. IEEE Trans. Gilabert.6] [2.5] [2. Oct. de Haan. He. Energy Convers. IEEE Trans. Appl. no. Transient state analysis of a doubly-fed [2. Fault ride-through of large wind farms using series dynamic braking resistors (March 2007). 21.. pp. Fortmann. Power Electron. vol. Conf.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 59 2. Mar. 22. 11th Eur. and L. 909-919. 2007. IEEE Trans. 3. Aug. vol. May 2008. 174-180.W. Roboam. 3. Vicatos and J.12] . IEEE Trans. Nagoya. 22. no. P. Seoul. Crowbar control algorithms for doubly fed induction generator during voltage dips. IEEE Trans. 2007. 2007. G. pp. Conf. Jun. Germany. IEEE Trans..8] [2. pp. pp. 2-5. Power Syst.. Feltes. P. no. G.3] [2. and S. IET Renew. Power Gener. Modeling of wind turbines based on doubly-fed induction generators for power system stability studies. Dynamic behavior of the doubly fed induction generator during three-phase voltage dips.

321-330. 1991. vol. Energy Convers. 2006.. 62-68. Analysis and comparison of a fast turn-on series IGBT stack and high-voltage-rated commercial IGBTs. no. 23. vol. 5. vol. pp. E. Curry. and L. Energy Convers.13] J. Loree. Castagno. 6. S. Roboam. Marroyo. 1. López. no. 1. D. Oct. 2008. X. P. [2. Mar. IEEE Trans. Wind turbines based on doubly fed induction generator under asymmetrical voltage dips. Gubía. [2. R. 34. Mar.. pp 1692-1696. Sanchis.Chapter 2 Doubly-Fed Induction Generator Fault Protection Schemes 60 induction generator under three phase short circuit. and E. pp. IEEE Trans.14] . Plasma Science. IEEE Trans. no.

- SF6 vs VCB - MV
- Transients in Wind Power Plants - IEEE
- GTP-GREENKO-RAJASTHAN.pdf
- User Charges
- Sub 15572
- M.R
- EE General Formulas
- Illumination 1
- Arc-Flash
- Final Specifications TN IT 1
- 33138
- 746-hasan
- 746-hasan.pdf
- CTPT Sets Vendors
- 00796224
- HVDC Ground Electrode
- is.1363.2.2002
- WRA-HB
- 400kV GIS Gurgoan
- 2
- Wind Farm Electrical System
- 122977049-Edsa-Paladin
- 29 Clamps Connectors 220 400kv Ss
- 400kV Layout Gujarat
- TechSpecshTAeriaBunchCable-21feb

Doubly Fed Induction Generator

Doubly Fed Induction Generator

- 06170989by tissatom
- Equivalent Model for Calculating Short Circuit Current of Doubly Fed Wind Generator Under Uniterrupted Excitationby Madhusudhan Srinivasan
- 1MRG001910_en_Rotor_earth-fault_protection_with_injection_unit_RXTTE4_and_REG670by Alfonso Núñez Schorr
- An Analytical Method for Analysis of Large Disturbance Stability of Induction Generatorsby yourou1000

- DFIG modelling
- NIT Rourkela Thesis
- DFIG
- Modelling of DFIG
- DFIG DOC
- dfig
- XR-EE-ES_2010_004
- Doubly fed Induction Generators in wind energy
- Wind Farm (DFIG Average Model)
- Dfig Thesis
- RSO.pdf
- 1MRG001910 B en Rotor Earth-fault Protection With Injection Unit RXTTE4 and REG670
- 06170989
- Equivalent Model for Calculating Short Circuit Current of Doubly Fed Wind Generator Under Uniterrupted Excitation
- 1MRG001910_en_Rotor_earth-fault_protection_with_injection_unit_RXTTE4_and_REG670
- An Analytical Method for Analysis of Large Disturbance Stability of Induction Generators
- Short-Circuit Modeling of a Wind Power Plant
- Generator Protection
- R10
- Synchronous Mach
- Wound Rotor Motor Testing
- Fault Analysis of an Unbalanced Distribution System With Distributed Generation
- Sliding Mode Control for a Wound Rotor
- Generador Stamford HCI534F
- A Single Phase Induction Generator as Wind Generator
- HCI4F
- LCR meter
- Comparison of Fixed Speed and Doubly-fed Induction Wind Turbines During Power System Disturbances
- auto volt regulate
- Simulation of Symmetrical Induction Machinery
- DFIG

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd