ACI FE Flux estimator of the 3-ph induction motor

Description This software module implements the flux estimator with the rotor flux angle
for the 3-ph induction motor based upon the integral of back emf’s (voltage
model) approach. To reduce the errors due to pure integrator and stator
resistance measurement, the compensated voltages produced by PI
compensators are introduced. Therefore, this flux estimator can be operating
over a wide range of speed, even at very low speed.

UDsS
pu PsiDrS
UQsS pu
pu ACIFE PsiQrS
IDsS MACRO pu
pu ThetaFlux
IQsS pu
pu

Availability This IQ module is available in one interface format:

1) The C interface version

Module Properties Type: Target Independent, Application Dependent

Target Devices: 28x Fixed and Floating Point devices

C Version File Names: aci_fe.h

IQmath library files for C: IQmathLib.h, IQmath.lib

Digital Control Systems (DCS) Group 1

Texas Instruments
C Interface

C Interface

Object Definition
The structure of ACIFE object is defined by following structure definition

typedef struct { _iq ThetaFlux; // Output: Rotor flux angle

_iq IQsS; // Input: Stationary q-axis stator current
_iq IDsS; // Input: Stationary d-axis stator current
_iq K1; // Parameter: Constant using in current model
_iq FluxDrE; // Variable: Rotating d-axis rotor flux (current model)
_iq K2; // Parameter: Constant using in current model
_iq FluxQrS; // Variable: Stationary q-axis rotor flux (current model)
_iq FluxDrS; // Variable: Stationary d-axis rotor flux (current model)
_iq K3; // Parameter: Constant using in stator flux computation
_iq K4; // Parameter: Constant using in stator flux computation
_iq FluxDsS; // Variable: Stationary d-axis stator flux (current model)
_iq FluxQsS; // Variable: Stationary q-axis stator flux (current model)
_iq PsiQsS; // Variable: Stationary d-axis stator flux (voltage model)
_iq Kp; // Parameter: PI proportional gain
_iq UiDsS; // Variable: Stationary d-axis integral term
_iq UCompDsS; // Variable: Stationary d-axis compensated voltage
_iq Ki; // Parameter: PI integral gain
_iq PsiQsS; // Variable: Stationary q-axis stator flux (voltage model)
_iq UiQsS; // Variable: Stationary q-axis integral term
_iq UCompQsS; // Variable: Stationary q-axis compensated voltage
_iq EmfDsS; // Variable: Stationary d-axis back emf
_iq UDsS; // Input: Stationary d-axis stator voltage
_iq K5; // Parameter: Constant using in back emf computation
_iq K6; // Parameter: Constant using in back emf computation
_iq EmfQsS; // Variable: Stationary q-axis back emf
_iq UQsS; // Input: Stationary q-axis stator voltage
_iq K8; // Parameter: Constant using in rotor flux computation
_iq K7; // Parameter: Constant using in rotor flux computation
_iq PsiDrS; // Output: Stationary d-axis estimated rotor flux
_iq PsiQrS; // Output: Stationary q-axis estimated rotor flux
} ACIFE;

typedef ACIFE *ACIFE_handle;

Digital Control Systems (DCS) Group 2

Texas Instruments
 C Interface

Module Terminal Variables

Item Name Description Format* Range(Hex)

Inputs UDsS stationary d-axis stator voltage GLOBAL_Q 80000000-7FFFFFFF
UQsS stationary q-axis stator voltage GLOBAL_Q 80000000-7FFFFFFF
IDsS stationary d-axis stator current GLOBAL_Q 80000000-7FFFFFFF
IQsS stationary q-axis stator current GLOBAL_Q 80000000-7FFFFFFF
Outputs PsiDrS stationary d-axis rotor flux linkage GLOBAL_Q 80000000-7FFFFFFF
PsiQrS stationary q-axis rotor flux linkage GLOBAL_Q 80000000-7FFFFFFF
ThetaFlux rotor flux linkage angle GLOBAL_Q 00000000-7FFFFFFF
(0 – 360 degree)
ACIFE K1 K1 = Tr/(Tr+T) GLOBAL_Q 80000000-7FFFFFFF
parameter K2 K2 = T/(Tr+T) GLOBAL_Q 80000000-7FFFFFFF
K3 K3 = Lm/Lr GLOBAL_Q 80000000-7FFFFFFF
K4 K4 = (Ls*Lr-Lm*Lm)/(Lr*Lm) GLOBAL_Q 80000000-7FFFFFFF
K5 K5 = Rs*Ib/Vb GLOBAL_Q 80000000-7FFFFFFF
K6 K6 = T*Vb/(Lm*Ib) GLOBAL_Q 80000000-7FFFFFFF
K7 K7 = Lr/Lm GLOBAL_Q 80000000-7FFFFFFF
K8 K8 = (Ls*Lr-Lm*Lm)/(Lm*Lm) GLOBAL_Q 80000000-7FFFFFFF
Internal FluxDrE stationary d-axis rotor flux GLOBAL_Q 80000000-7FFFFFFF
FluxQrE stationary q-axis rotor flux GLOBAL_Q 80000000-7FFFFFFF
FluxDsS stationary d-axis stator flux GLOBAL_Q 80000000-7FFFFFFF
FluxQsS stationary q-axis stator flux GLOBAL_Q 80000000-7FFFFFFF
PsiQsS stationary d-axis stator flux GLOBAL_Q 80000000-7FFFFFFF
PsiQsS stationary q-axis stator flux GLOBAL_Q 80000000-7FFFFFFF
UiDsS stationary d-axis integral term GLOBAL_Q 80000000-7FFFFFFF
UiQsS stationary q-axis integral term GLOBAL_Q 80000000-7FFFFFFF
EmfDsS stationary d-axis back emf GLOBAL_Q 80000000-7FFFFFFF
EmfQsS stationary q-axis back emf GLOBAL_Q 80000000-7FFFFFFF
UCompDsS stationary d-axis comp. voltage GLOBAL_Q 80000000-7FFFFFFF
UCompQsS stationary q-axis comp. voltage GLOBAL_Q 80000000-7FFFFFFF
*
GLOBAL_Q valued between 1 and 30 is defined in the IQmathLib.h header file.

Special Constants and Data types

ACIFE
The module definition is created as a data type. This makes it convenient to instance an
interface to the flux estimator of Induction Motor module. To create multiple instances of the
module simply declare variables of type ACIFE.

ACIFE_handle
User defined Data type of pointer to ACIFE module

ACIFE_DEFAULTS
Structure symbolic constant to initialize ACIFE module. This provides the initial values to the
terminal variables as well as method pointers.

Digital Control Systems (DCS) Group 3

Texas Instruments
 C Interface

Methods

ACIFE_MACRO(ACIFE_handle);

This definition implements one method viz., the flux estimator of Induction Motor computation
macro. The input argument to this macro is the module handle.

Module Usage

Instantiation
The following example instances two ACIFE objects
ACIFE fe1, fe2;

Initialization
To Instance pre-initialized objects
ACIFE fe1 = ACIFE_DEFAULTS;
ACIFE fe2 = ACIFE_DEFAULTS;

Invoking the computation macro

ACIFE_MACRO(fe1);
ACIFE_MACRO(fe2);

Example
The following pseudo code provides the information about the module usage.

main()
{
fe1.K1 = parem1_1; // Pass parameters to fe1
. .
. .
. .
fe1.K8 = parem1_8; // Pass parameters to fe1

fe2.K1 = parem2_1; // Pass parameters to fe2

. .
. .
. .
fe2.K10_fe = parem2_8; // Pass parameters to fe2

void interrupt periodic_interrupt_isr()

{
fe1.UDsS= voltage_dq1.d; // Pass inputs to fe1
fe1.UQsS= voltage_dq1.q; // Pass inputs to fe1
fe1.IQsS=current_dq1.d; // Pass inputs to fe1
fe1.IDsS=current_dq1.q; // Pass inputs to fe1

fe2.UDsS= voltage_dq2.d; // Pass inputs to fe2

fe2.UQsS= voltage_dq2.q; // Pass inputs to fe2
fe2.IQsS=current_dq2.d; // Pass inputs to fe2
fe2.IDsS=current_dq2.q; // Pass inputs to fe2

Digital Control Systems (DCS) Group 4

Texas Instruments
 C Interface

ACIFE_MACRO(fe1); // Call compute macro for fe1

ACIFE_MACRO(fe2); // Call compute macro for fe2

flux1.d = fe1.PsiDrS; // Access the outputs of fe1

flux1.q = fe1.PsiQrS; // Access the outputs of fe1
angle1 = fe1.ThetaFlux; // Access the outputs of fe1

flux2.d = fe2.PsiDrS; // Access the outputs of fe2

flux2.q = fe2.PsiQrS; // Access the outputs of fe2
angle2 = fe2.ThetaFlux; // Access the outputs of fe2

Digital Control Systems (DCS) Group 5

Texas Instruments
Constant Computation Macro

Constant Computation Macro

Since the flux estimator of Induction motor module requires eight constants (K1,…, K8) to be
input basing on the machine parameters, base quantities, mechanical parameters, and
sampling period. These eight constants can be internally computed by the macro
(aci_fe_const.h). The followings show how to use the C constant computation macro.

Object Definition
The structure of ACIFE_CONST object is defined by following structure definition

typedef struct { float32 Rs; // Input: Stator resistance (ohm)

float32 Rr; // Input: Rotor resistance (ohm)
float32 Ls; // Input: Stator inductance (H)
float32 Lr; // Input: Rotor inductance (H)
float32 Lm; // Input: Magnetizing inductance (H)
float32 Ib; // Input: Base phase current (amp)
float32 Vb; // Input: Base phase voltage (volt)
float32 Ts; // Input: Sampling period in sec
float32 K1; // Output: constant using in rotor flux calculation
float32 K2; // Output: constant using in rotor flux calculation
float32 K3; // Output: constant using in rotor flux calculation
float32 K4; // Output: constant using in stator current calculation
float32 K5; // Output: constant using in stator current calculation
float32 K6; // Output: constant using in stator current calculation
float32 K7; // Output: constant using in stator current calculation
float32 K8; // Output: constant using in torque calculation
} ACIFE_CONST;

typedef ACIFE_CONST *ACIFE_CONST_handle;

Digital Control Systems (DCS) Group 6

Texas Instruments
 Constant Computation Macro

Module Terminal Variables

Item Name Description Format Range(Hex)

Inputs Rs Stator resistance (ohm) Floating N/A
Rr Rotor resistance (ohm) Floating N/A
Ls Stator inductance (H) Floating N/A
Lr Rotor inductance (H) Floating N/A
Lm Magnetizing inductance (H) Floating N/A
Ib Base phase current (amp) Floating N/A
Vb Base phase voltage (volt) Floating N/A
Ts Sampling period (sec) Floating N/A
Outputs K1 constant using in rotor flux calculation Floating N/A
K2 constant using in rotor flux calculation Floating N/A
K3 constant using in rotor flux calculation Floating N/A
K4 constant using in stator current cal. Floating N/A
K5 constant using in stator current cal. Floating N/A
K6 constant using in stator current cal. Floating N/A
K7 constant using in stator current cal. Floating N/A
K8 constant using in torque calculation Floating N/A

Special Constants and Data types

ACIFE_CONST
The module definition is created as a data type. This makes it convenient to instance an
interface to the flux estimation of Induction Motor constant computation module. To create
multiple instances of the module simply declare variables of type ACIFE_CONST.

ACIFE_CONST_handle
User defined Data type of pointer to ACIFE_CONST module

ACIFE_CONST_DEFAULTS
Structure symbolic constant to initialize ACIFE_CONST module. This provides the initial
values to the terminal variables as well as method pointers.

Methods

ACIFE_CONST_MACRO (ACIFE_CONST_handle);

This definition implements one method viz., the flux estimator of Induction Motor constant
computation macro. The input argument to this macro is the module handle.

Module Usage

Instantiation
The following example instances two ACIFE_CONST objects
ACIFE_CONST fe1_const, fe2_const;

Digital Control Systems (DCS) Group 7

Texas Instruments
 Constant Computation Macro

Initialization
To Instance pre-initialized objects
ACIFE_CONST fe1_const = ACIFE_CONST_DEFAULTS;
ACIFE_CONST fe2_const = ACIFE_CONST_DEFAULTS;

Invoking the computation macro

ACIFE_CONST_MACRO(fe1_const);
ACIFE_CONST_MACRO(fe2_const);

Example
The following pseudo code provides the information about the module usage.

main()
{

fe1_const.Rs = Rs1; // Pass floating-point inputs to fe1_const

fe1_const.Rr = Rr1; // Pass floating-point inputs to fe1_const
fe1_const.Ls = Ls1; // Pass floating-point inputs to fe1_const
fe1_const.Lr = Lr1; // Pass floating-point inputs to fe1_const
fe1_const.Lm = Lm1; // Pass floating-point inputs to fe1_const
fe1_const.Ib = Ib1; // Pass floating-point inputs to fe1_const
fe1_const.Vb = Vb1; // Pass floating-point inputs to fe1_const
fe1_const.Ts = Ts1; // Pass floating-point inputs to fe1_const

fe2_const.Rs = Rs2; // Pass floating-point inputs to fe2_const

fe2_const.Rr = Rr2; // Pass floating-point inputs to fe2_const
fe2_const.Ls = Ls2; // Pass floating-point inputs to fe2_const
fe2_const.Lr = Lr2; // Pass floating-point inputs to fe2_const
fe2_const.Lm = Lm2; // Pass floating-point inputs to fe2_const
fe2_const.Ib = Ib2; // Pass floating-point inputs to fe2_const
fe2_const.Vb = Vb2; // Pass floating-point inputs to fe2_const
fe2_const.Ts = Ts2; // Pass floating-point inputs to fe2_const

ACIFE_CONST_MACRO(fe1_const); // Call compute macro for fe1_const

ACIFE_CONST_MACRO (fe2_const); // Call compute macro for fe2_const

fe1.K1 = _IQ(fe1_const.K1); // Access the floating-point outputs of fe1_const

. .
. .
. .
fe1.K8 = _IQ(fe1_const.K8); // Access the floating-point outputs of fe1_const

fe2.K1 = _IQ(fe2_const.K1); // Access the floating-point outputs of fe2_const

. .
. .
. .
fe2.K8 = _IQ(fe2_const.K8); // Access the floating-point outputs of fe2_const

Digital Control Systems (DCS) Group 8

Texas Instruments
Technical Background

Technical Background

The overall of the flux estimator [1] can be shown in Figure 1. The rotor flux linkages in the
stationary reference frame are mainly computed by means of the integral of back emf’s in the
voltage model. By introducing the compensated voltages generated by PI compensators, the
errors associated with pure integrator and stator resistance measurement can be taken care.
The equations derived for this flux estimator are summarized as follows:

Continuous time:

Firstly, the rotor flux linkage dynamics in synchronously rotating reference frame
 
  e   can be shown as below:
r

d edr,i L m e 1
 i ds   edr,i   e   r  eqr,i (1)
dt r r
d eqr,i
Lm e 1
i qs   qre,i  e   r  dre,i
 (2)
dt r r
L
where Lm is the magnetizing inductance (H),  r  r is the rotor time constant (sec), and r
Rr
is the electrically angular velocity of rotor (rad/sec).

In the current model, total rotor flux linkage is aligned into the d-axis component, which is
modeled by the stator currents, thus
 er ,i   edr,i and  eqr,i  0 (3)
Substituting  qr  0 into (1)-(2), yields the oriented rotor flux dynamics are
e ,i

d edr,i L m e 1
 i ds   edr,i (4)
dt r r
 eqr,i  0 (5)
Note that (4) and (5) are the classical rotor flux vector control equations. Then, the rotor flux
linkages in (4)-(5) are transformed into the stationary reference frame performed by inverse
park transformation.
     
 sdr,i   edr,i cos   r   eqr,i sin   r   edr,i cos   r (6)

     
 sqr,i   edr,i sin   r   qre ,i cos   r   edr,i sin   r (7)
where   r is the rotor flux angle (rad).

Then, the stator flux linkages in stationary reference frame are computed from the rotor flux
linkages in (6)-(7)

Digital Control Systems (DCS) Group 9

Texas Instruments
 Technical Background

- + u sds
Rs

Rs
u sqs
s s - +
i qs i ds
Eqs. (4)-(5) Eqs. (6)-(7)
u comp , ds +
e e ,i Eqs. (10)-(11)
i  s ,i
 dr
s ,i
 ds -
s,v
 ds
s
i ds ds dr -
s
i qs
Park
Trans.
i e
qs  qr
e ,i Inverse park
Trans.
s ,i
 qr
s ,i
 s  r , is s ,i s
 s ,i
 qs + - 2 PI's u comp , qs + 
s ,v
 qs
Eqs. (8)-(9)
 ds
s ,v s,v
 qs + Eqs. (12)-(13) - Voltage Model
Current Model
r r
s,v s,v
 qs  ds
s ,v
 dr s,v
  sqr,v   dr
r tan 1  s ,v  
s ,v

s ,v
 s ,v
,i s

  qr r s s
s,v
 dr   qr
Eqs. (14)-(15)
s s
i qs i ds

Figure 1: Overall system of flux estimator

Digital Control Systems (DCS) Group 10

Texas Instruments
 Technical Background

 L s L r  L2m  s L m s ,i
  L i  L i  
s ,i
ds
s
s ds
s
m dr
i ds 
  dr (8)
 Lr  Lr
 L L  L2m  s L m s ,i
 sqs,i  L s i sqs  L m i sqr   s r i qs 
  qr (9)
 Lr  Lr
where Ls and Lr are the stator and rotor self inductance (H), respectively.

Next, the stator flux linkages in the voltage model is computed by means of back emf’s
integration with compensated voltages.

 sds, v   u sds  i sds R s  u comp,ds dt  (10)

 sqs, v   u s
qs  i sqs R s  u comp,qs dt (11)
s s
where Rs is the stator resistance (), u ds , u qs are stationary dq-axis stator voltages, and
the compensated voltages are computed by the PI control law as follows:
Kp

u comp,ds  K p  sds, v   sds,i 
TI
  
s ,v
ds 
  sds,i dt (12)


u comp,qs  K p  sqs, v   sqs,i   P
K
TI  
s,v
qs 
  sqs,i dt (13)

The proportional gain KP and the reset time TI are chosen such that the flux linkages
computed by current model is dominant at low speed because the back emf’s computed
by the voltage model are extremely low at this speed range (even zero back emf’s at zero
speed). While at high speed range, the flux linkages computed by voltage model is
dominant.

Once the stator flux linkages in (10)-(11) are calculated, the rotor flux linkages based on
the voltage model are further computed, by rearranging (8)-(9), as
 L L  L2m s L
 sdr, v   s r i ds  r  sds, v
 (14)
 Lm  Lm
 L L  L2m s L
 sqr, v   s r i qs  r  sqs, v
 (15)
 Lm  Lm
Then, the rotor flux angle based on the voltage model is finally computed as
  sqr, v 
  r  tan 1  s , v  (16)
 
 dr 
Discrete time:

The oriented rotor flux dynamics in (4) is discretized by using backward approximation as
follows:
 dre,i (k )   dre ,i (k  1) L m e 1
 i ds (k )   edr,i (k ) (17)
T r r
where T is the sampling period (sec). Rearranging (17), then it gives
 r  e,i  LmT  e
 e,i
dr (k)     dr (k-1)    i ds (k) (18)
 r  T   r  T 
Digital Control Systems (DCS) Group 11
Texas Instruments
 Technical Background

Next, the stator flux linkages in (10)-(11) are discretized by using trapezoidal (or tustin)
approximation as

 sds, v (k )   sds, v (k  1) 
T s
2

e ds (k )  e sds (k  1)  (19)

T

 sqs, v (k )   sqs, v (k  1)  e sqs (k )  e sqs (k  1)
2
 (20)

where the back emf’s are computed as

e sds (k )  u sds (k )  i sds (k )R s  u comp,ds (k ) (21)
e sqs (k )  u sqs (k )  i sqs (k )R s  u comp,qs (k ) (22)
Similarly, the PI control laws in (12)-(13) are also discretized by using trapezoidal
approximation as

 
u comp,ds (k )  K p  sds, v (k )   sds,i (k )  u comp,ds ,i (k  1) (23)
u comp,qs (k )  K p  s,v
qs (k )   sqs,i (k )   u comp ,qs ,i (k  1) (24)
where the accumulating integral terms are as

u comp,ds ,i (k )  u comp,ds ,i (k  1) 
K P T s,v
TI

 ds (k )   sds,i (k ) 
(25)

 u comp,ds ,i (k  1)  K P K I  (k )   (k ) s,v
ds
s ,i
ds 
u comp,qs ,i (k )  u comp,qs ,i (k  1) 
K T
T
 P s,v
qs (k )   sqs,i (k ) 
I (26)

 u comp,qs ,i (k  1)  K P K I  (k )   (k ) s,v
qs
s ,i
qs 
T
where KI  .
TI

Discrete time and Per-unit:

Now all equations are normalized into the per-unit by the specified base quantities.
Firstly, the rotor flux linkage in current model (18) is normalized by dividing the base flux
linkage as
   e,i  T  e
 dr,pu
e,i
(k)   r   dr,pu (k  1)    ids,pu (k) pu (27)
 r  T   r  T 
where  b  L m I b is the base flux linkage (volt.sec) and Ib is the base current (amp).

Next, the stator flux linkages in the current model (8)-(9) are similarly normalized by
dividing the base flux linkage as
 L L  L2m s L
 sds,i,pu (k )   s r i ds ,pu (k )  m  sdr,i,pu (k )
 pu (28)
 LrLm  Lr

Digital Control Systems (DCS) Group 12

Texas Instruments
 Technical Background

 L s L r  L2m s L
 s ,i
qs , pu (k )   i qs ,pu (k )  m  sqr,i,pu (k )
 pu (29)
 LrLm  Lr
Then, the back emf’s in (21)-(22) are normalized by dividing the base phase voltage Vb
IbR s s
e sds ,pu (k )  u sds ,pu (k )  i ds ,pu (k )  u comp,ds ,pu (k ) pu (30)
Vb
I R
e sqs ,pu (k )  u sqs ,pu (k )  b s i sqs ,pu (k )  u comp,qs ,pu (k ) pu (31)
Vb
Next, the stator flux linkages in the voltage model (19)-(20) are divided by the base flux
linkage.
VT  e sds ,pu (k )  e sds ,pu (k  1) 
 s,v
(k )   s,v
(k  1)  b   pu (32)
ds , pu ds , pu

LmIb 2 
 
V T  e qs ,pu (k )  e qs ,pu (k  1) 
s s

 sqs, v,pu (k )   sqs, v,pu (k  1)  b  pu (33)

L m I b  2 

Similar to (28)-(29), the normalized rotor flux linkages in voltage model are
 L L  L2m s L r s,v
 sdr, v,pu (k )   s r i ds ,pu (k )   ds ,pu (k ) pu (34)

 LmLm  Lm
 L s L r  L2m s L r s,v
 qr ,pu (k )  
s,v
i qs ,pu (k )   qs ,pu (k ) pu (35)

 LmLm  Lm
In conclusion, the discrete-time, per-unit equations are rewritten in terms of constants.

Current model – rotor flux linkage in synchronously rotating reference frame     r

 e,i
dr,pu (k)  K1 e,i
dr,pu (k  1)  K i e
2 ds,pu (k) pu (36)
r T
where K 1  , and K 2  .
r  T r  T
Current model – rotor flux linkages in the stationary reference frame   0 
 sds,i,pu (k )  K 4 i sds ,pu (k )  K 3  sdr,i,pu (k ) pu (37)
 sqs,i,pu (k )  K 4 i sqs ,pu (k )  K 3  sqr,i,pu (k ) pu (38)

Lm L L  L2m
where K 3  , and K 4  s r .
Lr LrLm
Voltage model – back emf’s in the stationary reference frame   0 
e sds ,pu (k )  u sds ,pu (k )  K 5 i sds ,pu (k )  u comp,ds ,pu (k ) pu (39)
e s
qs , pu (k )  u s
qs , pu (k )  K i s
5 qs , pu (k )  u comp,qs ,pu (k ) pu (40)
IbR s
where K 5  .
Vb
Voltage model – stator flux linkages in the stationary reference frame   0 

Digital Control Systems (DCS) Group 13

Texas Instruments
 Technical Background

 e sds ,pu (k )  e sds ,pu (k  1) 

 sds, v,pu (k )   sds, v,pu (k  1)  K 6   pu (41)
 2 
 
 e qs ,pu (k )  e qs ,pu (k  1) 
s s

 sqs, v,pu (k )   sqs, v,pu (k  1)  K 6   pu (42)

 2 
 
VT
where K 6  b .
LmIb
Voltage model – rotor flux linkages in the stationary reference frame   0 
 sdr, v,pu (k )  K 8 i sds ,pu (k )  K 7  sds, v,pu (k ) pu (43)

 sqr, v,pu (k )  K 8 i sqs ,pu (k )  K 7  sqs, v,pu (k ) pu (44)

Lr L s L r  L2m
where K 7  , and K 8  .
Lm LmLm
Voltage model – rotor flux angle
1   sqr, v,pu (k ) 
  r ,pu (k )  tan 1   pu (45)
2   s, v (k ) 
 dr , pu 
Notice that the rotor flux angle is computed by a look-up table of 0o-45o with 256 entries.

In fact, equations (36)-(44) are mainly employed to compute the estimated flux linkages
in per-unit. The required parameters for this module are summarized as follows:

The machine parameters:

- stator resistance (Rs)
- rotor resistance (Rr)
- stator leakage inductance (Lsl)
- rotor leakage inductance (Lrl)
- magnetizing inductance (Lm)

The based quantities:

- base current (Ib)
- base phase voltage (Vb)

The sampling period:

- sampling period (T)

Notice that the stator self inductance is L s  L sl  L m (H) and the rotor self inductance
is L r  L rl  L m (H).

Next, Table 1 shows the correspondence of notations between variables used here and
variables used in the program (i.e., aci_fe.c, aci_fe.h). The software module requires that
both input and output variables are in per unit values.

Digital Control Systems (DCS) Group 14

Texas Instruments
 Technical Background

Equation Variables Program Variables

u s UDsS
ds

Inputs u sqs UQsS

i sds IDsS

i sqs IQsS

 sdr, v PsiDrS
Outputs
 sqr, v PsiQrS

r ThetaFlux

 edr,i FluxDrE

 sdr,i FluxDrS

 sqr,i FluxQrS

 sds,i FluxDsS

 sqs,i FluxQsS
Others
 sds, v PsiQsS

 sqs, v PsiQsS

e sds EmfDsS

e sqs EmfQsS

u comp,ds UCompDsS

u comp,qs UCompQsS

Table 1: Correspondence of notations

References:

[1] C. Lascu, I. Boldea, and F. Blaabjerg, “A modified direct torque control for
induction motor sensorless drive”, IEEE Trans. Ind. Appl., vol. 36, no. 1, pp. 122-
130, January/February 2000.

Digital Control Systems (DCS) Group 15

Texas Instruments