CHAPTER 7

Induction Motor Slip-Power Recovery Drives

7.1 INTRODUCTION

Induction motor drives with full-power control on the stator side, as discussed in Chapters 4. 5 and 6, are widely used in industrial applications. Although either a cage-type or wound-rotor machine can be used in the drive, the former is always preferred because a wound-rotor machine is heavier, more expensive, has higher rotor inertia, a higher speed limitation, and maintenance and reliability problems due to brushes and slip rings. However, it is interesting to note that a wound-rotor machine with a mechanically varying rotor circuit rheostat is possibly the simplest and oldest method of ac motor speed control. One feature of this machine is that the slip power becomes easily available from the slip rings, which can be electronically controlled to control speed of the motor. For limited-range speed control applications, where the slip power is only a fraction of the total power rating of the machine, the converter cost reduction can be substantial. This advantage offsets the demerits of the wound-rotor machine to some extent. Slip-power recovery drives have been used in the following applications:

• Large-capacity pumps and fan drives

• Variable-speed wind energy systems

• Shipboard VSCF (variable-speed/constant-frequency) systems

• Variable-speed hydro pumps/generators

• Utility system flywheel energy storage systems

In this chapter, we will study the principles of slip-power control, particularly the popular static Kramer and static Scherbius drives. It should be noted that the nomenclature in these classes of drives is not consistent.

307

308

Chapter 7 • Induction Motor Slip-Power Recovery Drives

7.2 DOUBLY-FED MACHINE SPEED CONTROL BY ROTOR RHEOSTAT

A simple and primitive method of speed control of a wound-rotor induction motor is by mechanical variation of the rotor circuit resistance, as shown in Figure 7.1.

The torque-slip curves of the motor for varying rotor resistance Rr, as calculated by Equation (2.32), are shown in Figure 7.2. With external resistance R[ = 0, that is, with the slip rings shorted, the inherent torque-slip curve of the machine gives a speed corresponding to point A at the rated load torque. As the resistance is increased, the curve becomes flatter, giving less speed until the speed becomes zero at high resistance (> R4)' The maximum or breakdown torque (see Equation (2.35)) remains constant, but the starting torque, given by Equation (2.33), increases with higher resistance. The mechanical variation of resistance has the inherent disadvantage. In addition, this method of speed control is very inefficient because the slip energy is wasted in the rotor circuit resistance. However, several advantages of this method are: absence of in-rush starting current, availability of full-rated torque at starting, high line power factor, absence of line current harmonics, and smooth and wide range of speed control. The scheme is hardly used now-a-days.

Instead of mechanically varying the resistance, the equivalent resistance in the rotor circuit can be varied statically by using a diode bridge rectifier and chopper as shown in Figure 7.3. As usual, the stator of the machine is connected directly to the line power supply, but in the rotor circuit, the slip voltage is rectified to de by the diode rectifier. The dc voltage is converted to current source Id by connecting a large series inductor Ld. It is then fed to an IGBT shunt chopper with resistance R as shown. The chopper is pulse width modulated with duty cycle 8 = tOl/T, where lou = on-time and T = time period. When the IGBT is off, the resistance is connected in the circuit and the de link current Id flows through it. On the other hand, if the device is on, the resistance is short-circuited and the current Id is bypassed through it. It can be shown that the

3~, 60 Hz supply

Stator

Rotor

R

Figure 7.1 Doubly-fed induction motor speed control by rotor rheostat

Static Kramer Drive

309

Q)

r-

R1=()

Rated torque

-~

A.

Starting torque

0.75

0.5

0.25

o

Slip S

Figure 7.2 Torque-slip curves of motor with variable rotor resistance

3<1>,60 Hz supply

machine

L.,

= A

J R

R

__;j,. B

Bridge Chopper

rectifier

Figure 7.3 Motor speed control with rotor circuit chopper

duty cycle control of the chopper offers an equivalent resistance Ro = (I - 5)R between points A and B. Therefore, the developed torque and speed of the machine can be controlled by the variation of the duty cycle of the chopper. This electronic control of rotor resistance is definitely advantageous compared to rheostatic control, but the problem of poor drive efficiency remains the same. This scheme has been used in intermittent speed control applications in a limited speed range, where the efficiency penalty is not or great concern.

7.3 STATIC KRAMER DRIVE

Instead of wasting the slip power in the rotor circuit resistance, it can be converted to 60 Hz ac and pumped back to the line. The slip power-controlled drive that permits only a subsynchronous range of speed control through a converter cascade is known as a static Kramer

310

Chapter 7 • Induction Motor Slip-Power Recovery Drives

3~, 60 Hz supply

Power input

Power i feedback

Shaft power

motor

+

,.

Slip power

Diode rectifier

Inverter

Figure 7.4 Static Kramer drive system

drive, and the scheme is shown in Figure 7.4. It is different from the original Kramer drive, where rotating machines were used for slip energy recovery. The static Kramer drive has been very popular in large power pump and fan-type drives, where the range of speed control is limited near, but below the synchronous speed. The drive system is very efficient and the converter power rating is low, as mentioned before, because it has to handle only the slip power. In fact, the power rating becomes lower with a more restricted range of speed control. The additional advantages, which will be explained later, are that the drive system has de machine-like characteristics and the control is very simple. These advantages largely offset the disadvantages of the wound-rotor induction machine.

The machine air gap flux is established by the stator supply, and it practically remains constant if stator drops and supply voltage fluctuation are neglected. Ideally, the machine rotor current is a six-stepped wave in phase with the rotor phase voltage if the de link current (1 is considered harmonic-free, and the commutation overlap angle of the diode rectifier is neglected. The machine fundamental frequency phasor diagram referred to the stator is shown in Figure 7.5, where VI = phase voltage, 1,/ = fundamental frequency rotor current referred to the stator, IfIR = air gap flux, 1m = magnetizing current, and <p = power factor angle. With constant air gap flux, machine torque becomes directly proportional to current I,/- Since 11/ is directly proportional to de link current Id' the torque is also proportional to Id. Instead of static resistance control as discussed in the previous section, the scheme here can be considered as CEMF control, where a variable CEMF VI is being presented by a phase-controlled, line-commutated inverter to control the de link current Id. In steady-state operation, the rectified slip voltage Vd and inverter

Static Kramer Drive

311

Figure 7.5 Machine phasor diagram referred to the stator

de voltage VI will balance, ignoring the resistive drop in the inductance Ld. The voltage Vd is proportional to slip S, whereas the current Id is proportional to developed torque. At a certain speed, the inverter's firing angle can be decreased to decrease the voltage Vb which will increase Id to increase the corresponding developed torque. The simplified speed and torque expressions can be derived as follows. Neglecting the stator and rotor drops, voltage Vd is given as Equation (3.21).

(7.1 )

where S = per-unit slip, VL = stator line voltage, and nI = stator-to-rotor turns ratio of the machine. Again, the inverter de voltage VI is given as Equation (3.57).

(7.2)

where n2 = transformer line side-to-inverter ac side turns ratio and a = inverter firing angle. For inverter operation, the range of the firing angle is 1(/2 < a < 1(. Since in steady state Vd and VI must balance, Equations (7.1) - (7.2) give

S=~lcosal

n2

(7.3)

Therefore, the speed expression (Dr can be given as

to, = we (1- S)

~ We {I-~;Icosal}

= We (I-Icosal)

(7.4)

312

Chapter 7 • Induction Motor Slip-Power Recovery Drives

where 11 I In2 = 1 has been assumed. Equation (7.4) indicates that ideally, speed can be controlled between zero and synchronous speed we by controlling inverter firing angle a. At zero speed, voltage Vd is maximum, which corresponds to angle a = IT; at synchronous speed, Vd = 0 when a = rr/2. In practice, the maximum speed should be slightly less than synchronous speed so that torque (i.e., Id) can be developed with a finite resistance drop (Vd = Id Rd) of the de link inductor at VI = O.

Again, neglecting losses, the following power equations can be written:

(7.5)

Pm = (1- S)~f( = Tewm 2

= Tewe (1- S)-

P

(7.6)

where Pg = air gap power, Pill = mechanical output power, wm = mechanical speed, and P = number of poles. Substituting Equations (7.2), (7.3), and (7.5) in (2.23) gives

P P" t, =(-)-~-

2 we

= (p) Vlld 2 SWe

1.35 I I

--VL cosa Id

= (p ) _11_;:2:....__ _

2 ~Icosalwe

n2

= (p) 1.35VL ld 2 wenl

(7.7)

This equation indicates that the torque is proportional to current ld. The drive system has nearly the characteristics of a separately excited dc motor, because the air gap flux is nearly constant and the torque is proportional to current Id. With higher load torque Tv the machine tends to slow down and current ld increases so that Te = TL. In other words, for a fixed firing angle of the inverter, the voltage VI is fixed. Therefore, to balance the resistance drop of the de link inductor, Vel must slightly increase, giving speed drop characteristics like a de machine. Figure 7.6 gives torque-speed curves for different firing angles a. More accurate torque-speed relations will be developed later.

The static Kramer drive has one-quadrant speed control characteristics. The drive cannot have regenerative braking capability, and speed reversal is not possible. Regenerative braking in the subsynchronous speed range will be discussed later. For reversal of speed, a circuit breaker

Static Kramer Drive

313

Speed (cor) (pu)

o

0.5 Torque (T e) pu

1.0 (pu)

Figure 7.6 Typical torque-speed curves at different inverter firing angles

can be installed on the stator side, which should reverse the phase sequence of the line voltages. For most pump and fan drive applications, simple one-quadrant speed control is acceptable.

7.3.1 Phasor Diagram

A fundamental frequency phasor diagram can be drawn to explain the performance of the drive system. In practice, the rotor current displacement factor will slightly deviate from unity because of the commutation overlap angle shown in Figure 7.7. In fact, the overlap angle f.1 introduces a lagging angle CPr to the fundamental current, as indicated in the figure. This current increases as ld increases with the increase of slip S. Near zero slip, when the rotor voltage is very small, a large current ld may cause overlap angle u to exceed the 1[/3 angle, causing a short circuit between the upper and lower diodes.

Figure 7.8 shows the approximate phasor diagram of the drive system at the rated torque condition, where all the phasors are referred to the line or stator side. The stator draws a

Fundamental component

Irf

(0 t

Figure 7.7 Rotor phase voltage and current waves

314

Chapter 7 • Induction Motor Slip-Power Recovery Drives

;' u. = 0

S = 1.0

constant slip line

T 0.5 ~

t

I.S = 0

\

n)n1 = 1 l.cos a

Figure 7.8 Phaser diagram of static Kramer drive system at rated voltage

magnetizing current 1111, which lags 7[12 angle with respect to the stator phase voltage VI' The total stator current 1\ lags the stator voltage by angle qJs as shown. On the inverter side, although the active power is fed back to the line, it also demands lagging reactive current from the line because of phase control. This additional reactive current drawn by the inverter reduces the overall power factor of the system. Assuming continuous conduction of the inverter and ripple-free current (" the inverter output power factor is [cos qJl =Icos ai, that is, its power factor varies linearly with de voltage VI' This, of course, neglects the inverter commutation overlap effect. Consider no transformer connection for the present and n I = I. The phasor diagram shows the inverter line current h at slip S = 0.5. Phasors IT and 1,/ are nearly equal in magnitude because of the nearly identical waves of these currents. The active component h cos qJ opposes the stator active current, whereas the reactive component IT sin qJ adds to the stator magnetizing current 1/11' The total line current I L is the phasor sum of II and IT and it lags at angle qJL' which is larger than the stator power factor angle qJ.\. With constant torque, the magnitude of Ir remains constant, but as the slip varies between 0 and I, the phasor IT rotates from a = 90° to 160°, as shown in the figure. At zero speed (S = I), the machine acts as a transformer, and ignoring losses, all the active power is transferred back to the line through the inverter. The result is that both the machine and inverter consume only reactive power. The inverter margin angle (f3) of 20° for the inverter covers both commutation un and turn-off (y) angles. From the phasor diagram, it is evident that at S = 0, the system power factor is lagging at low value, which deteriorates as the slip increases. Therefore, with restricted speed range close to synchronous speed, the power factor is comparatively better. For reduced torque condition, the current IT is proportionally less. The corresponding phasor diagram modification is left as an exercise to the reader. If, for example, the

Static Kramer Drive

315

torque is reduced to 50 percent at S = 0.5, angle cp remains the same as shown and current IT is reduced to 50 percent, that is, both IT cos cp and IT sin cp are also proportionately reduced.

For a restricted speed range closer to synchronous speed, the system power factor can be further improved by using a step-down transformer. The transformer primary-to-secondary turns ratio n2 can be adjusted so that at the desired maximum slip, angle cp = n. Of course. the inverter margin angle should always be maintained. Substituting this condition in Equation (7.3) gives

(7.8)

For example, if Smax = 0.5 and n I = I, then 112 should ideally be 2. In the phasor diagram of Figure 7.8, this condition corresponds to the transformer line current I; = 0.5IT, which clearly indicates the power factor improvement of line current h. As the speed is increased to change the slip from 0.5 to 0, the phasor I; rotates anti-clockwise, as shown, until cp = n12. Equation (7.3) indicates that for constant slip, the variation of 1121111 linearly varies cos a magnitude. The constant slip line at S = 0.5 is indicated in Figure 7.8.

The step-down transformer has essentially two functions: besides improving the line power factor, it also helps to reduce the converter power ratings. Both the rectifier and inverter should be designed to handle the same current Id as dictated by the torque requirement. The rectifier should be designed for the slip voltage given by SV Lin l> whereas the inverter should be designed for the line voltage VL in the absence of the transformer. The rectifier voltage and corresponding power rating decrease with a smaller speed range, but the inverter must be designed for full power. Installation of the transformer reduces the voltage and corresponding power rating of the inverter, and the criteria for the turns ratio 112 design is the same as that of Equation (7.8). For the same example (i.e., Smux = 0.5, 11 I = I, and 112 = 2), both the rectifier and inverter have an equal power rating, which is 50 percent of the full power. It can be shown easily that the converter power rating can be reduced proportionately as Sma.\" is reduced. This is an important advantage of the slip-power recovery drive. The discussion above assumes that the machine is not started with the converters in the circuit. Otherwise, the advantage of a reduced converter rating is lost.

A typical starting method of a Kramer drive with resistance switching is shown in Figure 7.9. The motor is started with switch I on and switches 2 and 3 off. As the speed builds up, resistances RI and R2 are shorted sequentially until at the desired Smar value, switch I is opened and the drive controller is brought into action.

316

Chapter 7 • Induction Motor Slip-Power Recovery Drives

3~ ac suppIJl---1- ....,

2 3

Figure 7.9 Motor starting method

7.3.2 AC Equivalent Circuit

The drive system performance can be analyzed with the help of a de or ac equivalent circuit of the machine. We will attempt here an approximate ac equivalent circuit with respect to the rotor. Neglecting drops in the semiconductor devices, the slip-power output is partly dissipated in the de link resistance Rd of the inductor and is partly fed back to the ac line through the transformer.

The respective power components can be given as

(7.9)

(7.10)

The equivalent rotor circuit power per phase is given as

(7.11 )

Therefore, the machine air gap power per phase, which includes the rotor copper loss, is given as

(7.12)

where I,. = rotor rms current per phase, R,. = rotor resistance, and Pn/ = mechanical output power per phase. The torque and corresponding mechanical power Pm' are essentially contributed by

Static Kramer Drive

317

the fundamental component of rotor current Irjonly. The expression for rotor circuit copper loss per phase is

, 2 1 2

~'I = t, R; +-Id Rd 3

= 1/ (Rr +O.SRd)

where Irt = J6 Id has been substituted for a six-step wave. Therefore, the expression for PII/ is

• . Tr

(7.13)

grven as

, l-S

~II = (Fund.freq.slip power)--

S

+/(Rr+O.5Rd)+ 3~. L~VL liflcosalJC~SJ

where the 11 = ~ Id relation has been used to replace Id in Equation (7.10). The air gap power Pg' in Equation (7.12) can be written by substituting Equation (7.14) as follows:

(7.14)

(7.IS)

where

(7.16)

1C 1.3SVL I I

RA = tR, + O.SRd) + r; . cosrz

3...;6 172Id

= (R; + O.SRd ) + ~ Icosal

17211

where the I, = ~ 11 relation has been used in Equation (7.16) to eliminate II" Note that the air gap power P g' consists of two components: one is the fundamental frequency slip power and the other is the ripple power loss. Equation (7.1S) indicates that the rotor circuit, which absorbs the active power, can be represented by a per-phase passive ac equivalent circuit where RA is the equivalent resistance given by Equation (7.17). The resistance Rx represents an additional resis-

(7.17)

318

Chapter 7 • Induction Motor Slip-Power Recovery Drives

Stator side

.

----+ .... --

:--+ Pg' Rotor side --+>

Figure 7.10 Per-phase passive equivalent circuit of the machine (with respect to rotor

tance that consumes harmonic power. The equivalent circuit is shown in Figure 7.10, where RA is a function of Vs' Ir!' and cos a.

Note that all the stator circuit parameters and the supply voltage are multiplied by S to refer to the rotor circuit. The symbol'(prime) indicates rotor-referred parameters with turns ratio n I; in other words,

(7.18)

(7.19)

(7.20)

(7.21 )

(7.22)

It is more convenient to represent the equivalent circuit in terms of the CEMF presented by the inverter. Equation (7.15) can also be written in the form

(7.23)

Static Kramer Drive

319

where

(7.24 )

(7.25)

Figure 7.11 shows part of the rotor referred equivalent circuit with the CEMF Vc' where torque can be increased by increasing Iif, that is, by decreasing Vc (with the help of S and cos a parameters).

7.3.3 Torque Expression

The average torque developed by the machine is given by the total fundamental air gap power divided by the synchronous speed we- The expression in terms of a passive ac equivalent circuit is

(7.26)

where P d = fundamental frequency per-phase air gap power, and the expression of RA is given in Equation (7.17). The equation can be solved in terms of circuit parameters as

(7.27)

1 V

V c= S n; I cosc I

"_____'p

: 9

.

"

Figure 7.11 Part of rotor referred per-phase equivalent circuit with CEMF Vc

320

Chapter 7 • Induction Motor Slip-Power Recovery Drives

where

(7.28)

has been substituted. Equation (7.27) relates torque as a function of slip S, rotor current Irf" and inverter firing angle a. An approximate torque expression relating slip and the a angle can be derived more conveniently from the equivalent circuit of Figure 7.11. The torque in terms of fundamental air gap power P J is given from Equation (7.23) as

(7.29)

An approximate expression of Irl can be written from the equivalent circuit by neglecting the reactances and stator resistance, which are small at small values of slip. Therefore,

(7.30)

Substituting Equations (7.30) and (7.25) in (7.29) yields

(7.31 )

where RB = RrlS has been substituted, neglecting the resistance Rd' Equation (7.31) relates torque as a function of both the slip S and a angle approximately. A more accurate torque expression can be derived by a computer program using the equivalent circuit. The results are plotted in Figure 7.12. The shaded area in the figure indicates the normal zone of operation, which can be compared with the approximate curves shown in Figure 7.6.

Static Kramer Drive

321

S' -9;

(J) :::J

~

o

r-=- 1.0 t--~:---......lI~";T.~ob77.:i"*---t~----

Normal operation zone

Load torque

rr/2 < o. < rr

o

0.5

1.0

Speed (pu)

Figure 7.12 Torque-speed curves at different firing angles of inverter

7.3.4 Harmonics

The rectification of slip power causes harmonic currents in the rotor, and these harmonics are reflected to the stator by the transformer action of the machine. The harmonic currents are also injected to the ac line by the inverter. As a result, the machine losses are increased and some amount of harmonic torque is produced. The rotor current wave is ideally six-stepped, which is given by the Fourier series

(7 . .32 )

where the fundamental component contributes the useful torque, but the lower order harmonics, (such as 5th and 7th) have dominating harmful effects. Each harmonic current in the rotor will create a rotating magnetic field and its direction of rotation will depend on the order of the harmonic. The 5th harmonic, for example, at frequency Sws/' rotates opposite to the direction of the rotor, whereas the 7th harmonic, at frequency 7 ws/' rotates in the same direction. The interaction of different harmonics with the air gap flux creates pulsating torque. For example, the 5th and 7th harmonics, interacting with the fundamental lJIg wave, contribute to the 6th harmonic pulsating torque, which is discussed in Chapter 2. However, it can be shown that the harmonic torque is small compared to the average torque and can be neglected in a practical drive.

322

Chapter 7 • Induction Motor Slip-Power Recovery Drives

+ (Or-+

Figure 7.13 Speed control of static Kramer drive

7.3.5 Speed Control of a Kramer Drive

A speed control system of a Kramer drive is shown in Figure 7.13, and Figure 7.14 shows its typical performance. As explained before, the drive has the characteristics of a separately excited de motor, and therefore, the control strategy is similar to a phase-controlled rectifier de drive. With essentially constant air gap flux, the torque is proportional to de link current Id. which is controlled by an inner feedback loop. If the command speed OJ,." is increased by a step. as shown in Figure 7.14, the motor accelerates at a constant developed torque corresponding to the Idt limit set by the speed control loop. The inverter firing angle (X initially decreases with high slope to establish Id• and then gradually decreases as speed increases. As the actual speed approaches the command speed, the de link current is reduced to balance the load torque at a certain (X angle in steady state. As the speed command is decreased by a step, Id approaches zero and the machine slows down by the inherent load torque braking effect. During deceleration. the (X angle increases continuously so that the inverter voltage VI balances the rectifier voltage Vd. Then, as the speed error tends to be zero in the steady state, Iii is restored so that the developed torque balances with the load torque. The air gap flux during the whole operation remains approximately constant, as dictated by the stator voltage and frequency. As mentioned before. the maximum speed should be slightly less than the synchronous speed so that the current It! can be established with finite Yd'

7.3.6 Power Factor Improvement

As discussed above, the static Kramer drive is characterized by poor line power factor because of a phase-controlled inverter. The power factor can be improved by a scheme called

Static Kramer Drive

323

~:

Speed (ro.) 1_ (U,

DC current (Id)

·

Firing angle a n t .. · .. ·"t· ... ·· ........ · .. · .. t::~:;( .. ··

.............................. ·t .

rr/2: .:

· '.

· '.

· '.

· '.

Time

Figure 7.14 Performance characteristics of Kramer drive

commutatorless Kramer drive, which is shown in Figure 7.15. The scheme is somewhat analogous to a primitive Kramer system, where a de motor is coupled to the induction motor shaft and is fed by the rectified slip power through a diode rectifier. In such a case, the de motor absorbs the slip power and returns to its shaft coupled to the ac machine. In Figure 7.15. the de motor is replaced by a synchronous motor with a load-commutated inverter, which acts as a "commutatorless de machine." The commutatorless de motor will be explained in Chapter 9. The inverter is fired from signals of an absolute position sensor mounted on the synchronous machine shaft.

The power flow diagram in the commutatorless Kramer drive is shown in Figure 7.16. The air gap power P; flowing from the stator is split into shaft input power and slip power as in the static Kramer system. But, the slip power drives the synchronous motor and adds to the shaft

SP] +-

: l- ..

: If

Rectifier

~P(i

········111 •••

Field

Synchronous machine

Shaft

•• PS:

•• •

. :

Slip power SP9

: .

Comutatorless de motor

Figure 7.15 Commutatorless Kramer drive system

324

Chapter 7 • Induction Motor Slip-Power Recovery Drives

SUpply_ ...... power

Air gap power

Sh aft in put

Shaft output

Induction motor stator Input

Sl/ne motor stator Input

Stator copper loss

Rotor COIT'ielter

copper IIJSS

I O~3 S

Figure 7.16 Power flow diagram in commutatorless Kramer drive system

input power to constitute the total mechanical power. The synchronous motor field is supplied from the line through a controlled rectifier. The speed and torque of the drive system are controlled by the field current Ifwith the inverter tiring angle ex set to an optimum value for load commutation of the inverter. At any speed, a higher value of If will increase the machine CEMF, that is. the inverter de voltage VI, which will decrease the developed torque. As the machine speed decreases from synchronous speed, Vd increases linearly. but Irincreases vr: K(Vj/wr», reaching saturation soon at a lower speed. Again, as is characteristic to the load-com mutated inverter drive, speed control is not possible at low values because of insufficient CEMF. Besides having an improved power factor, the system will operate reliably with short-time power failure. which is not possible in a static Kramer drive. The drive, as usual, has one-quadrant characteristics.

7.4 STATIC SCHERIUS DRIVE

As explained above and indicated by the phasor diagram of Figure 7.5, the Kramer drive has only a forward motoring mode (one quadrant) of operation. For regenerative mode operation, rotor current wave should be reversed and the corresponding phasor IrF should be negative. as indicated in Figure 7.17.

This feature requires that the slip power in the rotor flow in the reverse direction. If the diode rectifier on the machine side is replaced by a thyristor bridge. as shown in Figure 7.18, the slip power can be controlled to flow in either direction. With reverse slip-power flow at subsynchronous speed. the power corresponding to shaft input mechanical power can be pumped out of the stator. It can be shown that such a drive system, with bidirectional slip-power flow. can be controlled for motoring and regenerating in both the subsynchronous and supersynchronous ranges of speed. This scheme is often defined as a static Scherbius drive system. The line commutation of the machine-side converter becomes difficult near synchronous speed (excessive commutation overlap time), when the ac voltage is very small.

Static Scherius Drive

325

(a)

tVg

(b)

Figure 7.17 (a) Waveforms for regenerative operation of Kramer drive, (b) Phasor diagram

3~, 60 Hz supply

(1-S)Pg

Figure 7.18 Static Scherbius drive system using de-link thyristor converters

The dual-bridge converter system in Figure 7.18 can be replaced by a single phasecontrolled line-commutated cycloconverter, as shown in Figure 7.19. The use of a cycloconverter means additional cost and complexity of control, but the resulting advantages are obvious. The problem of commutation near synchronous speed disappears, and the cycloconverter can easily operate as a phase-controlled rectifier, supplying de current in the rotor and permitting true synchronous machine operation. The additional advantages are near-sinusoidal current waves in the rotor, which reduce harmonic loss, and a machine over-excitation capability that permits leading power factor operation on the stator side. In fact, the cycloconverter's input lagging power factor can be cancelled by the leading machine's power factor so that the line's power factor is unity. The cycloconverter should be controlled so that its output frequency and phase track precisely with those of the rotor slip frequency voltages. Like a Kramer drive, a Scherbius drive also

326

Chapter 7 • Induction Motor Slip-Power Recovery Drives

Transformer

Induct ma

J1f'
y ~ tJ I

W
. t:~"t\
t~~n~;.. I ~' , ~, ~, ~ , f, ~, , J ',f, ~, ~
.) r.) J "'" ~)
~,,') J ~) r. ~" ~
;-- +J-
+ - + -
Cycloconverter A B C

3<jl, ac suppl

Bidirectional slip power flow,6 SPg

Figure 7.19 Static Scherbius drive using cycloconverter

requires resistive starter starting, which is not shown in Figure 7.19. Of course, like the Kramer drive, speed control is only possible in the forward direction. The Scherbius drive has recently found application in multi-MW, variable-speed pumps/generators [7] and flywheel energy storage [81 systems. The transformer shown on the line side of the cyc1oconverter is needed, as explained before, to reduce the converter power rating.

7.4.1 Modes of Operation

There are four modes of operation of a Scherbius drive, which can be explained with the help of Figure 7.20. In all the cases, it is assumed that the shaft torque is constant to the rated value and losses in the machine and cycloconverter are neglected.

Mode 1: Subsynchronous Motoring

This mode, shown in Figure 7 .20( a), is identical to that of the static Kramer system. The stator input or air gap power P g is positive and remains constant, and the slip power SP g' which is proportional to slip (which is positive), is returned back to the line through the cyc1oconverter, as shown. Therefore, the line supplies the net mechanical power output Pm = (I - S)P g consumed by the shaft. The slip frequency current in the rotor creates a rotating magnetic field in the same direction as in the stator, and the rotor speed to; corresponds to the difference (we - wsI) between these two frequencies. At true synchronous speed (S = 0), the cyc1oconverter supplies de excitation to the rotor and the machine behaves like a standard synchronous motor.

Mode 3: Subsynchronous Regeneration

In regenerative braking condition, as shown in Figure 7 .20(b), the shaft is driven by the load and the mechanical energy is converted into electrical energy and pumped out of the stator. With negative rated torque, the mechanical power input to the shaft Pm = (1 - S)P g increases with speed and this equals the electrical power fed to the line. In the subsynchronous speed

Static Scherius Drive

327

1

(1-S)Pg ~

l

I'

Figure 7.20 Modes of operation of static Scherbius drive

(cycloconverter not shown):

(a) Mode 1: Subsynchronous motoring,

(b) Mode 3: Subsynchronous regeneration, (c) Mode 2: Supersynchronous motoring,

(d) Mode 4: Supersynchronous regeneration

range, the slip S is positive and the air gap power Pg is negative; correspondingly, negative slip power SP g is fed to the rotor from the cycloconverter so that the total air gap power is constant. The rotor current has positive phase sequence as before. At synchronous speed, the cyclocon-

328

Chapter 7 • Induction Motor Slip-Power Recovery Drives

verter supplies de excitation current to the rotor and the machine behaves as a synchronous generator. The drive can have sustained operation in mode 3. The typical application in this is a variable-speed wind generation system.

Mode 2: Supersynchronous Motoring

In this mode, as shown in Figure 7.20(c), the shaft speed increases beyond the synchronous speed, the slip becomes negative, and the slip power is absorbed by the rotor. The slip power supplements the air gap power for the total mechanical power output (l + S)p z: The line therefore supplies slip power in addition to stator input power. At this condition, the phase sequence of slip frequency is reversed so that the slip current-induced rotating magnetic field is opposite to that of the stator.

Mode 4: Supersynchronous Regeneration

In this mode, indicated in Figure 7.20(d), the stator output power Pg remains constant, but the additional mechanical power input is reflected as slip-power output. The cycloconverter phase sequence is now reversed so that the rotor field rotates in the opposite direction. The variable-speed wind generation mentioned for mode 3 can also be used in this mode.

Power distribution as a function of slip in subsynchronous and supersynchronous speed ranges is summarized for all four modes in Figure 7.21, where the operating speed range of ±50 percent about the synchronous speed is indicated. The control of the Scherbius drive is somewhat complex. It will be discussed in Chapter 8 with the help of vector or field-oriented control (see Figure 8.45).

Bidirectional slip-power flow with a cycloconverter, as discussed above, is also possible if the cycloconverter is replaced by a double-sided, PWM, voltage-fed converter system, as shown in Figure 7.22. For a high power rating, IGBTs can be replaced by GTOs. The dc link voltage Vd should be sufficiently higher than the inverter line voltage to permit PWM operation in the linear or undermodulation region. Both the rectifier and inverter can be operated at a programmable input power factor so that the effective line power factor can be maintained at unity. The rectifier operates satisfactorily at variable voltage and variable slip frequency on the ac side, including the ideal de condition at synchronous speed.

7.4.2 Modified Scherbius Drive for VSCF Power Generation

A modified Scherbius drive, which has a somewhat similar topology to Figure 7.15, has been used for stand-alone shipboard VSCF power generation. The scheme, shown in Figure 7.23, has some interesting features. Induction generator output power is fed to a stand-alone, 60 Hz, constant voltage bus, which supplies the active and reactive load power as shown. The distribution of active and reactive powers in the supersynchronous and subsynchronous ranges is shown in the figure. The stator active power output Pm of the generator is equal to the turbine shaft power and the slip power fed to the rotor by the cycloconverter. The stator reactive power output QL is reflected to the rotor as SQL, which is added with the machine magnetizing power

Static Scherius Drive

329

Mechanical output power

o

_Slip

S=1

S=D

I Operating __j

r speed range

I I

S= -1 Speed (pu)-

S=1

S=D

Speed (pu) _ S= -1

o

_Slip

Slip-power output SPg

Mechanical input power

Figure 7.21 Power distribution vs. slip power in sub/supersynchronous speed ranges:

(a) Motoring at constant torque,

(b) Generation at constant torque

Slip-power input, -SP~

requirement to constitute the total reactive power QL' of the cycloconverter. The power QL' is further increased to QL" at the cycloconverter input, which is supplied by the shaft-mounted synchronous exciter. The slip frequency and its phase sequence are adjusted for varying shaft speed so that the resultant air gap flux rotates at synchronous speed, as explained before. In subsynchronous speed range, the slip power SPm is supplied to the rotor by the exciter, and therefore, the remaining output power (1 - S)Pm is supplied by the shaft. In supersynchronous speed range. the rotor output power tlows in the opposite direction and runs the excitor as a synchronous motor. Therefore, the total shaft power increases to (I + S)Pm. Rotor voltage and frequency vary linearly with deviation from synchronous speed. For example, if the shaft speed varies in the range of 800 to 1600 rpm with 1200 rpm as the synchronous speed (S = ±0.33). the corresponding range of slip frequency is 0 to 20 Hz for 60 Hz supply frequency.

330

Chapter 7 • Induction Motor Slip-Power Recovery Drives

3<1>,60 Hz supply

PII\IM rectifier

PI/\/M inverter

Figure 7.22 Static Scherbius drive using doublesided PWM voltage-fed converter

ac exciter

(1±S)Pm

3(11, 60 Hz bus at constant voltage

Turbine shaft (variable-speed)

+SP~

SUB

Rotor

L..----- .... Cycloconverter ...... -----'

excitation

+-- SUPER -SPm

Figure 7.23 Variable-speed, constant frequency generation using modified Scherbius drive

The modified Scherbius system as a VSCF generator has several advantages over the conventional Scherbius system. One principal difference is that the wrap-around, or circulating, KVA demanded by the rotor is supplied from a separate exciter instead of being supplied from the machine's stator terminals. As a result, the main machine is much smaller in size, although in this case, the power is distributed between the two machines. The VSCF bus is much cleaner with regard to harmonics, because the cycloconverter input harmonics are reflected to the exciter. The rotor excitation circuit can be designed with a higher voltage, and the necessity of an input transformer is eliminated. In the case of a supply brown-out or temporary short-circuit fault, the system has improved controllability and reliability of power supply over the standard Scherbius system. Again, the cycloconverter can be replaced by a double-sided, PWM voltage-

Summary

331

fed converter system, which will relieve the additional reactive and harmonic power loadings of the exciter.

7.5 SUMMARY

This chapter gives a broad review of different types of slip power-controlled drives. Unlike a standard cage-type induction motor, these drives require a doubly-fed wound-rotor induction motor, which is more expensive and has the disadvantages of slip rings and brushes. In the beginning, the primitive rotor rheostat-type speed control was discussed. Then, the more important slip-power recovery-type drives, such as the static Kramer and static Scherbius drives. were discussed in detail for limited speed range applications. The important advantages of these drives are reduced power rating of the converter (at the expense of a 60 Hz transformer) and a fast de machine-like transient response. The additional disadvantages of these drives are the need of a separate starting method, low line power factor, and non-reversible speed control. For Scherbius drives with cycloconverters or double-sided voltage-fed PWM converters. the power factor problem does not arise. For large power pump and compressor-type applications within a limited speed range, these drives have been widely used. Scherbius drives have also been used III variable-speed wind generation, hydro/pump storage and utility systems. and flywheel energy storage systems.

REFERENCES

1. A. Lavi and R. L. Polge, "Induction motor speed control with static inverter in the rotor", IL'EE Trans. Powc:

Appar. Syst .. vol. 85. pp. 76-84, Jan. 1966.

2. T. Hori, H. Nagase, and M. Hornbu, "Induction Motor Control Systems", Industrial Electronics Handbook . J. D.

Irwin, pp. 310-315, CRC Press, 1997.

3. T. Wakabayashi, T. Hori, K. Shimzu, and T. Yoshioka, "Cornmutatorless Kramer control system for large capacity induction motors for driving water service pumps", ILEE/IAS Annu. Meet. Con]. Rec .. pp. 822-828. I 1.J76.

4. H. W. Weiss. "Adjustable speed ac drive systems for pump and compressor applications", ILU:' Trans. Oil Ind.

Appl .. vol. 10. pp. 162-157. Jan.lFeb. 1975.

S. P. Zimmermann. "Super synchronous static converter cascade". Con/ Rec. IFAC S\'IIII) Oil Control ill P(III'('/" l.lcc, and Electrical Drives, pp. SS9-S74, 1977.

6. G. A. Smith. "Static Scherbius system of induction motor speed control", Proc. IEL, vol. 124. pp, SS7-S6S. I 1.J77.

7. S. Mori et al., "Commissioning of 400 MW adjustable speed pumped storage system for Ohkawachi hydro power plant", Pmc. Cigre Svntp., No. 520-04. 1995.

8. T. Nohara, H. Scnaha, T. Kagcyarna, and T. Tsukada. "Successful commercial operation of doubly-fed adjustablespeed flywheel generating station." Proc. ()j'c/CRt/IEE Japan Colloquium 1111 Rotating Electric Muchincrv Liic Extn .. pp. 1-6, 19I.J7.

9. R. Pcana, J. C. Clare, and G. M. Asher. "Doubly fed induction generator using back-to-buck PWM converters and its application to varriablc speed wind energy generation." I !:'E Prot. Oil Elec. Power App .. \01. 14.1. pp. 231-2--1. May 1996.

10. R. Dalla and V. T. Rangunathan, "Decou pled control of active and reactive power for a grid-connected doubly-fed wound rotor induction machine without position sensors:" II:'!:!: lAS AIIIIU. Meet. COllI Rcc .. pp. 262.\-'::630. 19l)l).