CHAPTER 2 -

CONTROL AND ESTIMATION OF INDUCTION MOTOR DRIVES

2.1. Introduction

The control and estimation of induction motor drives constitute a vast

subject, and the technology has further advanced in recent years. Induction

motor drives with cage-type machines which have been the workhorses in

industry for variable-speed applications in a wide power range that covers from

fractional horsepower to multi-megawatts. These applications include pumps

and fans, paper and textile mills, subway and locomotive propulsions, electric

and hybrid vehicles, machine tools and robotics, home appliances, heat pumps

and air conditioners, rolling mills, wind generation systems, etc. In addition to

process control, the energy-saving aspect of variable-frequency drives is

getting a lot of attention now-a-days.

The control and estimation of AC drives in general are considerably

more complex than those of DC drives, and this complexity increases

substantially if high performances are demanded. The main reasons for this

complexity are the need of variable-frequency, harmonically optimum

converter power supplies, the complex dynamics of AC machines, machine

parameter variations, and the difficulties of processing feedback signals in the

presence of harmonics.

In this chapter, different control techniques of induction motor drives,

including scalar control, vector or field-oriented control are discussed. The

estimation of feedback signals, particularly speed estimation in sensorless

vector controls was discussed. Variable-frequency power supplies will be

considered with voltage-fed inverters and current-fed inverters. However,

emphasis will be given to voltage-fed converters because of their popularity in

industrial drives.

2.2. Induction Motor Control with Small Signal Model

It consists of a converter-machine system with hierarchy of control

loops added to it. The converter-machine unit is with voltage (V,') and

frequency (w:) as control inputs. The outputs are shown as speed (a,),

developed torque (T,), stator current (I,), and rotor flux (y,). Instead of voltage

control, the converter may be current-controlled with direct or indirect voltage

control in the inner loop. Besides, there are coupling effects between the input

and output variables.

-
w,
TB
"-
-4- ""

. 'or
i

Converter- .
'74

-
Controller machine
system Is
w;
t
-P
J
. + Wl

t
Additional feedback
Wr
Signals

Fig. 2.1: General speed control block diagram of induction motor drive

20
 For example, both the torque and flux of a machine are functions of

voltage and hquency. Machine parameters may vary with saturation,

temperature, and skin effect, adding further nonlinearity to the machine model.

The converter can be described by a simplified model, which consists of an

amplifier gain and a dead-time lag due to the PWM sampling delay. The

system becomes discrete-time because of the converter and digital control

sampling effects.

All the control and feedback signals can be considered as DC and

proportional to actual variables. The speed control is shown with an inner

torque control loop, which may be optional. Adding a high-gain inner loop

control provides the advantages of linearization, improved bandwidth, and the

ability to control the signals within safe limits. Like a DC machine, the flux of

an AC machine is normally assumed to be constant at the rated value because it

gives fast response and high developed torque per ampere of current. In fact,

the flux under consideration may be stator flux (w,), rotor flux (w,), or air gap

flux (yr, or yp). However, the rotor flux control is considered in the present

case. The inner control loops have faster response (i.e., higher bandwidth) than

the outer loop.

Since a , AC drive system is multivariable, nonlinear with internal

coupling effect, and discrete-time in nature, its stability analysis is very

complex. Computer simulation study becomes very useful for investigating the

performance of the drive, particularly when a new control strategy is

developed. Once the control structure and parameters of the control system are
determined by the simulation study for acceptable performance, a prototype

system can be designed and tested with futher iteration of the controller

parameters.

2.3. Scalar Control

Scalar control, as the name indicates, is due to magnitude variation of

the control variables only, and disregards the coupling effect in the machine.

For example, the voltage of a machine can be controlled to control the flux, and

frequency or slip can be controlled to control the torque. However, flux and

torque are also functions of frequency and voltage, respectively, Scalar control

is in contrast to vector or field-oriented control is discussed in proceeding

sections, where both the magnitude and phase alignment of vector variables are

controlled. Scalar-controlled drives give somewhat inferior performance, but

they are easy to implement. Scalar-controlled drives have been widely used in

industry. However, their importance has diminished recently because of the

superior performance of vector-controlled drives, which is demanded in many

applications. In this section, a few selected scalar control techniques with

voltage-fed and current-fed inverters will be discussed.

2.4. Voltage-Fed Inverter Control

2.4.1. Open Loop Volts/& Control

The open loop volts/Hz control of an induction motor is by far the most

popular method of speed control because of its simplicity, and these types of

motors are widely used in industry. Traditionally, induction motors have been

used with open loop 60 Hz power supplies for constant speed applications. For
adjustable speed applications, frequency control is natural. However, voltage is

required to be proportional to frequency so that the flux ( V , =% ) remains

We

constant neglecting the stator resistance R, drop. Figure 2.2 shows the block

diagram of the volts/Hz speed control method. The power circuit consists of a

diode rectifier with a single- or thee-phase AC supply, LC filter, and PWM

voltage-fed inverter.

Ideally, no feedback signals are needed for the control. The fkequency

wemisthe primary control variable because it is approximately equal to speed

o,,neglecting the small slip frequency o,,of the machine. The phase voltage

v,' command is directly generated from the frequency command by the gain
factor G, as shown in fig 2.1, so that the flux yr, remains constant. If the stator

resistance and leakage inductance of the machine are neglected, the flux will

also correspond to the air gap flux yr,, or rotor flux yr,. As the frequency

decreases at low speed, the stator resistance tends to absorb the major amount

of the stator voltage, thus weakening flux.

Figure 2.3 shows the drive's steady-state performance on a torque-speed

). AS the frequency is
plane with a fan or pump-type load (TL = K urZ

gradually increased, the speed also increases almost proportionally, as

indicated by points 1, 2, 3, 4, etc. The operation can be continued smoothly in

the field-weakening region where the supply voltage V, saturates.

Now consider the load torque and line voltage variation effects. If the

initial operating point is 3 and the load torque is increased to Tt for the same

frequency command, the speed will drop from @, to o ,'. This droop is small,

23
particularly with a high-etlticiency (i.e., low slip) machine, and is easily

tolerated for a pump or fan-type drive where precision speed control is not

necessary. Assume now that the operation is at a point in another torque-speed

curve.

rectifier

Diode w
Vo Oe

v;
--+
vi = $VS sin 6,
G v;'/u, T)V; lnvedel
v; = Qvs sin (4- 2n -.--+
d 1

4. & V; = ~ Z V ,sin (0, + -1211 5

3
Speed 4
command

Motor

Figure 2.2: Open loop volts/Hz speed control with voltage-fed inverter
 If the AC line voltage decreases, the machine terminal voltage also

decreases. Their by speed will also droop corresponding to point b, as shown in

the figure 2.3 speed dmop correction in an open loop control can be achieved

by adding an estimated slip signal with the frequency command.

i-"
V

0)
3
P

- -
w;\ Speed (or)
Wr

Figure 2.3: Torque-speed curves showing eflect offrequency variation, load torque,

and supply voltage changes

If the frequency command we* is changed abruptly by a small

increment, the slip will change to change the developed torque (within the safe

limit), but the speed will tend to remain constant because of machine inertia.

However, if it is desired to change the speed by a large step of the frequency

command (positive or negative), the drive will easily become unstable. The sat-

isfactory acceleration/deceleration characteristics of the machine are explained

in Figure 2.4.
 (a) Deceleration

7
I
6
+- Deceleration --&
Acceleration 4

-0-
(b) Time
Figure 2.4: Acceleration/deceleration characteristics with volts/Hz control

(a) On torque-speed curves @) As function of time

 Assume a pure inertia-type load for simplicity and that the machine is

initially operating at point 1. The command frequency is now increased

abruptly by a small step so that with an increment of slip frequency, the

operating point shifts to point 2, which corresponds to the rated, developed

torque. The drive now goes through constant acceleration with the ramping of

the frequency command when the speed tracks the frequency within the limit of

slip frequency so that the stability and safe stator current limit are maintained.

At operating point 3, the frequency command can be decremented to attain the

steady operating point, which is 4. The machine torque and the speed are

related by the following equation:

m r = j ( K - T, Id!
where J = Moment of inertia,

T, = Developed torque, and

TL= Load torque (zero in this case).

With the rated T,, the slope of acceleration 2 as indicated in Figure 2.4

(b), is dictated by parameter J, that is, a higher J will permit slow acceleration,

and vice versa. If it is possible to estimate J on-line for a variable inertia load,

the acceleration of the drive can be predetermined. The deceleration

performance is similar, and it is also explained in Figure 2.4. With a diode

rectifier on the line side, the inverter will need a dynamic brake, as indicated in

Figure 2.2. By decrementing the frequency command in a step, the operating

point will shift from point 1 to 5, due to a negative developed torque. It will
then decelerate with constant slope given by Equation 2.1 until the steady-state

operating point, 7, can be reached again.

2.4.2. Speed Control with Slip Regulation

An improvement of open loop volts/Hz control is close loop speed

control by slip regulation as shown in Figure 2.5. Here, the speed loop error

generates the slip command a,,' through a proportional-integral (PI) controller

and limiter,

Vd
-,I
I
+

Wr
K1 t-

P-l
" I

Controller
* 1

or

'-(;//O Speed

Encoder
Motor

Figure 2.5: Close loop speed control wilh Valts/Hz control and slip regulation
 The dip is added to the feedback speed signal to generate the frequency

~ommandas shown in fig 2.5. The frequency command o,'also generates the

voltage command through a volts& function generator, which incorporates

the low-frequency stator drop compensation. Since the slip is proportional to

the developed torque at constant flux, the scheme can be considered as an open

loop torque control within a speed control loop. The feedback current signal is

not used anywhere in the loop. With a step-up speed command, the machine

accelerates freely with a slip limit that corresponds to line stator current or

torque limit, and then settles down to the slip value at steady state as dictated

by the load torque. If the command speed w: is reduced by a step, the drive

goes into regenerative or dynamic braking mode and decelerates with constant

negative slip d l ,as indicated in the figure 2.5.

Fig. 2.6: (a) Efect of h a d Torque Variation(b) Efect of Supply Voltage Variation
 The effects of load torque and line voltage variation are explained in

Figure 2.6. If the initial operating point is 1 and the load torque is increased

from TLto TL'.the speed will tend to drop corresponding to point 2. However,

the speed control loop will increase the frequency until the original speed is

restored at point 3. Since there is no close loop flux control, the line voltage

variation will cause some flux drift. Again if the initial point is 1 on curve a of

Figure.2.6 (b), the decrease of line voltage will reduce the flux, tending to shift

the operating point to 2. The resulting speed drop will act on the speed loop and

raise the frequency to restore the original speed at point 1 on curve c. The

system works well in the field-weakening mode also.

2.4.3. Speed Control with Torque and Flux Control

As discussed above, the volts/Hz control has the disadvantage that the

flux may drift, and as a result, the torque sensitivity with slip will vary. In

addition, line voltage variation, incorrect volts1Hz ratio, stator drop variation by

line current, and machine parameter variation may cause weaker flux or the

flux may saturate. In Figure 2.5, if the flux becomes weak, the developed

torque will decrease with the slip limit and the machine's acceleration1

deceleration capability will decrease.

A speed control system with closed loop torque and flux control is

shown in Figure 2.7. Additional feedback loops mean complexity of additional

feedback signal and potential stability problems.

 Figure 2.7: Close loop speed control wirh torque andfltus control

A torque loop within the speed loop improves the speed loop's response.

The flux control loop controls the voltage V,* as shown in fig. 2.7. Both the

torque and flux feedback signals can be estimated from the machine terminal

voltages and currents, as indicated. The feedback signal estimation is discussed

".
in preceding chapters. With constant Y r command, as the speed increases, the

voltage increases proportionally until square-wave mode is reached and field-

weakening mode starts. However, if PWM operation is desired in field-

weakening mode, the flux command must be decreased to vary inversely with

the speed signal so that the PWM controller does not saturate. The flux control

loop is usually slower than the torque control loop. The drive can operate in

regenerative (or dynamic) braking mode, but the reversal of speed requires a

reversal of the phase sequence of the inverter.

 With scalar control, as the frequency command w,' increased by the

torque loop. The flutemporarily decreases until it is compensated by the

sluggish flux control loop. This inherent coupling effect slows down the torque

response.

2.4.4. Current-Controlled Voltage-Fed Inverter Drive

Instead of controlling inverter voltage by the flux loop, the stator

current can also be controlled. Closed loop current control is beneficial to

power semiconductors because of the inherent protection from over current.

Besides, the torque and the flux of the machine are directly sensitive to

currents. A voltage-fed inverter drive with outer loop torque and flux control

and hysteresis-band current control in the inner loop is shown in Figure 2.8.

Instead of constant rated flux, the flux can be programmed with torque as

shown for a light-load efficiency improvement.