Module 6: A.C. Electrical Machines For Hybrid and Electric Vehicles Lecture 21: Field Oriented Control of Induction Motor
Module 6: A.C. Electrical Machines For Hybrid and Electric Vehicles Lecture 21: Field Oriented Control of Induction Motor
Module 6: A.C. Electrical Machines For Hybrid and Electric Vehicles Lecture 21: Field Oriented Control of Induction Motor
Current
vector
s
iqdr qdr
s
r θ
B
θ
r q axis
d axis
Figure 1a: A coil in a magnetic field Figure 1b: Orientation of magnetic field in IM
From equation 1 it is evident that the torque is maximised when the current vector is
perpendicular to the magnetic field. The same conclusion can be applied to an IM. In
Figure 1b orientations of magnetic fields and currents in an IM are shown. The rotor
current and flux linkage vectors are shown in Figure 1 at some instant of time. Hence,
the torque produced by the motor (refer to Lecture 19) is given by
Te
3P
22
qr idr dr iqr
(2)
The equation 2 can be re-written as
3P
Te iqdr
qdr sin
22
(3)
The equation 3 is analogous to equation 1. Hence, for a given magnitude of flux
linkage, torque is maximised when the flux linkage and current vectors are
perpendicular. Therefore, it is desirable to keep the rotor flux linkage perpendicular to
rotor current vector.
In the analysis of FOC the following convention will be used:
The parameters with a superscript “ s ” are in stator frame of reference.
The parameters with a superscript “ e ” are in synchronous frame of reference.
The parameters with subscript “ r ” indicate rotor parameters.
The parameters with subscript “ s ” indicate stator parameters.
All rotor quantities are referred to stator using the turns ratio of the windings
(Lecture 17) and hence “ ' ” is dropped.
In case of singly excited IMs (in singly excited IM, the rotor winding is not fed by any
external voltage source. In case of wound rotor machines, they are short circuited
using slip rings. For cage IMs, the rotor bars are short circuited at the terminals), the
rotor flux linkage vector and rotor current vector are always perpendicular. The
voltage equations for the IM (refer to Lecture 19) in synchronous frame of reference
are
v qse rs iqse e dse pqse
v dse rs idse e qse p dse
vose rs iose pose
v qre rr iqre e r dre pqre
v dre rr idre e r qre pdre
vore rr iore pore
where
e is the rotational speed of synchronous frame of reference
(1) In case of singly excited IM, the rotor voltages are zero, that is ve 0 , qr
1
0 rr iqre e r dre pqre iqre e r dre pqre
rr
1
0 rr idre e r qre pdre idre e r qre pdre
rr
pore
0 rr iore pore iore
rr
(2) Since steady state operation of IM is considered, the time derivative term of flux
linkage in equation 2 will vanish. Hence, the rotor currents are:
1
iqre e r dre
rr
1
idre e r qre
rr
ior 0
e
(3)
The dot product of the rotor flux linkage and rotor current vectors may be expressed
as
qdr
e e
.iqdr qre .iqre dre .idre
(4)
Substituting the values of iqre and idre from equation 3 into equation 4 gives
qre dre
.i
e e
qdr qdr e r dre e r qre 0
rr rr
(5)
Form equation 5 it can be seen that the dot product between the rotor flux and rotor
current vectors is zero in case of singly excited IM. Hence, it can be concluded that
the rotor flux and rotor current vectors are perpendicular to each other in steady state
operation. The defining feature of FOC is that this characteristic (that the rotor flux
and rotor current vectors are perpendicular to each other) is maintained during
transient conditions as well.
In both direct and indirect FOC, the 90o shift between the rotor flux and rotor current
vector can be achieved in two steps:
The first step is to ensure that
qre 0
(6)
The second step is to ensure that
idre 0
(7)
By suitable choice of s on an instantaneous basis, equation 6 can be achieved.
Satisfying equation 7 can be accomplished by forcing d -axis stator current to remain
constant. To see this, consider the d -axis rotor voltage equation
0 rr idre e r qre pdre
(8)
Since qre 0 , equation 8 can be written as
0 rr idre pdre
(9)
The d -axis rotor flux linkage is given by (refer Lecture 19):
dre Llr idre Lm idse idR
e
(10)
Substituting the value of dre from equation 10 into equation 9 gives:
rr e Lm e
pidre idr pids
Llr Llr
(11)
If idss is held constant, then pidse 0 and the solution of equation 11 becomes
r
r t
Llr
i Ce
e
dr
where
C is a constant of integration
(12)
It is evident from equation 12 that the rotor current idre will decay to zero and stay at
zero regardless of other transients that may be taking place. Hence, the torque (from
equation 2) is given by
3P e e
Te dr iqr
22
(13)
The q -axis rotor flux is given by (refer Lecture 19)
qre Llr iqre Lm iqse iqre
(14)
Since, qre 0 , the q -axis rotor current is given by
Lm e
iqre iqs
Llr
(15)
Combining equations 13 and 15 gives
3 P Lm e e
Te dr iqs
2 2 Llr
(16)
The d -axis rotor flux is given by (refer Lecture 19)
dre Llr idre Lm idse idre
(17)
The equation 7 gives idre 0 , hence equation 17 can be written as
dre Lmidse
(18)
Together, equation 19 and equation 21 suggest the generic rotor flux oriented control
shown in Figure 2.
DC
Inverter IM
Source
iˆabcs
Current
control
Te e
iqs
4 Lrr , est Te
3PLm.est dr
'e
dr
e
e
ids
e 1
dr
Lm,est
e
Figure 2: Generic Field Oriented Control (FOC) [1]
In Figure 2 the variables of the form x , x and x̂ denote command, measured and
estimated values respectively. In case of parameters that are estimated, a subscript
”est” is used. The working of the controller is as follows:
i. Based on the torque command ( Te ), the assumed values of the parameters and
the estimated value of d -axis rotor flux ˆdrs is used to formulate a q -axis stator
current command iqss .
ii. The d -axis stator current command idss is calculated such as to achieve a rotor
flux command drs (using equation 12).s
iii. The q -axis and d -axis stator current command is then achieved using a
current source control.
The above description of rotor flux oriented FOC is incomplete with determination of
ˆdrs and s . The difference between direct and indirect FOC is in how these two
variables are determined.
Direct Rotor Oriented FOC
In direct FOC, the position of the synchronous reference frame ( e ) is determined
based on the values of q -axis and d -axis rotor flux linkages in the stationary
reference frame. The relation of flux linkages in synchronous reference frame and
stationary reference frame is
qre cos e sin e qrs
e sin cos s
dr e e dr
where
qrs is the rotor q-axis flux linkage in stationary frame of reference
drs is the rotor d -axis flux linkage in stationary frame of reference
(19)
In order to achieve qre 0 , it is sufficient to define the position of the synchronous
reference frame as
qrs
e tan 1
s
dr 2
(20)
The difficulty with this approach is that qrs and drs are not directly measurable
quantities. However, they can be estimated using direct measurement of air gap flux.
To measure the air gap flux, hall-effect sensors are placed in the air gap and used to
measure the air-gap flux in q -axis and d -axis. Since the hall-effect sensors are
stationary, the flux measured by them is in stationary reference frame. The flux
measured by the sensors is the net flux in the air gap (combination of stator and rotor
flux). The net flux in the air gap is given by:
qm
s
Lm iqss iqrs
where
Lm is the magnetization inductance
(21)
From equation 21, the rotor q -axis current is obtained as
qm
s
Lm iqss
iqrs
Lm
(22)
The q -axis rotor flux linkage is given by
qrs Llr iqrs Lm iqss iqrs
(23)
Substituting the rotor q -axis current from equation 22 into equation 23 gives
Llr s
qrs qm Llr iqss
Lm
(24)
An identical derivation for d -axis gives
Llr s
drs dm Llr idss
Lm
(25)
ˆqr
e
ˆ 2
1
iˆds
s
L'lr ,est rfc s 1 angle ˆ
qr jdr
e
ˆdr
e
L'rr , est
ˆdm
s
Lm, est
ˆqdm
s
DC
Inverter IM
Source
iˆabcs
Current Rotor flux
K ss
control calculator
e
iqs
e
4 Lrr , est Te
Te
3PLm.est dr
'e
e
ˆdr
e
e
ids
dr 1
Lm, est
(32)
e
rr iqs e
0r i e
qr pdre
Llr idse
r dr
(33)
Substituting the value of d -axis rotor flux from equations 17 into equation 33 gives
qre Lm iqse* e rr iqse
e lr dr
0 rr iqr L i e* Lmidse* pqre
L L i
lr lr ds
(34)
e
Llr idse
r dr
(35)
If the d -axis rotor current is held constant, then pidre* 0 and rearranging equations 34
and 35 gives
rr e i e*
pqre qr rr qse* idre
Llr ids
(36)
e*
rr e r i
pidre dr r qse* dre
Llr Lllr ids
(37)
The solution of equations 37 and 38 will decay to zero (same argument as used for
equation 12), hence qre and iqre will eventually become zero. In Figure 4 the
implementation of indirect FOC is shown and it is much simpler than the direct FOC.
DC
r
Inverter IM
Source
T1 T 6
Current iabcs
control
e
iqs
4 Lrr , est Te
Te
3PLm.est dr 'e
Reset
e N e
e
dr 1 ids
N rr,est s
Lm, est D Lrr ,est +
D
Figure 4: Indirect FOC [1]
References:
[1] P. C. Krause, O. Wasynczuk, S. D. Sudhoff, Analysis of electric machinery, IEEE
Press, 1995
Suggested Reading
[1] R. Krishnan, Electric motor drives: modeling, analysis, and control, Prentice Hall,
2001