FLUX ® 2D Application Scalar Command of An Induction Machine Technical Paper

CAD Package for Electromagnetic and Thermal
Analysis using Finite Elements
FLUX
2D Application
Scalar command of an induction
machine
technical paper
Copyright - September 2004
FLUX is a registered trademark.
FLUX software
FLUX2D technical papers
:
:
COPYRIGHT CEDRAT/INPG/CNRS/EDF
COPYRIGHT CEDRAT
FLUX2D's Quality Assessment

(Electricit de France, registered number AQMIL013)
This technical paper was edited on 24 September 2004.

Ref.: K205-R-810-EN-09/04
CEDRAT
15 Chemin de Malacher - Inovalle
38246 Meylan Cedex
FRANCE
Tlphone: +33 (0)4 76 90 50 45
Tlcopie: +33 (0)4 56 38 08 30
E-mail: cedrat@cedrat.com
Web: http://www.cedrat.com
- REMARK The files corresponding to different cases studied in this technical paper
are available in the folder:
\DocExamples\Examples2D\DriveMotorWithSimulink\
FluxFiles\
The corresponding applications are ready to be solved. This allows you
to adapt this technical paper to your needs.
Introduction
This technical paper shows an example of the simulation of the drive of a rotating machine
through the link FLUX to Simulink Technology.
Precisely we will present the scalar command of an induction machine.
One will first define the FLUX model of the induction machine. After computing a simplified
model of the complete system with Simulink, the computation will be made using FLUX to
Simulink Technology. Results will then be compared.
FLUX
TABLE OF CONTENTS
TABLE OF CONTENTS
PART A: FLUX MODEL
1. Geometry .................................................................................................................... 3
1.1
Overview of the geometry ............................................................................................... 3
1.2
Stator geometry............................................................................................................... 4
1.2.1
1.2.2
1.2.3
1.2.4
1.2.5
1.3
Geometrical parameters ....................................................................................................4

Coordinate systems ...........................................................................................................4
Points and lines for the upper half of the stator slot ..........................................................5
Geometric transformations ................................................................................................6
Completing the stator geometry ........................................................................................6
Rotor geometry ............................................................................................................... 9

1.3.1
1.3.2
1.3.3
1.3.4
1.3.5
1.3.6
Geometrical parameters ....................................................................................................9

Coordinate systems ...........................................................................................................9
Points and lines for the rotor bar .....................................................................................10
Geometric transformations ..............................................................................................10
Completing the rotor geometry ........................................................................................11
Closing the air-gap...........................................................................................................13
1.4
Add and assign regions for the faces............................................................................ 14
1.5
Mesh ............................................................................................................................. 16
1.5.1
1.5.2
1.5.3
Change to the Mesh context............................................................................................16

Applying the mesh points to the geometry ......................................................................16
Generate, verify and save the mesh................................................................................17
2. Materials.................................................................................................................... 19
3. Definition of the electrical circuit ................................................................................ 21
4. Physical properties .................................................................................................... 23
4.1
General information....................................................................................................... 23
4.2
Assign materials to the regions ..................................................................................... 23
4.3
Electrical circuit ............................................................................................................. 24
SCALAR COMMAND OF AN INDUCTION MACHINE
PAGE A
FLUX
TABLE OF CONTENTS
PART B: SIMULINK MODEL
25
5. Scalar Control............................................................................................................ 27
5.1
Principle of the scalar control.........................................................................................27

5.1.1
5.1.2
5.1.3
5.2
Introduction ......................................................................................................................27
Scalar control modeling ...................................................................................................27
Relations used for the scalar control ...............................................................................29
Structure of the Scalar control .......................................................................................30
6. Simulink model .......................................................................................................... 33

6.1
Definition of the Simulink model ....................................................................................33
6.2
Definition of the blocks...................................................................................................34

6.2.1
6.2.2
6.2.3
6.2.4
6.2.5
6.2.6
6.2.7
The command ..................................................................................................................34

Subsystem of the Instruction block ..................................................................................34
Speed controller...............................................................................................................35
Subsystem of the Scalar control ......................................................................................36
The induction machines block:........................................................................................37
Outputs ............................................................................................................................41
Other blocks.....................................................................................................................41
7. Solve ......................................................................................................................... 43
PAGE B
FLUX
TABLE OF CONTENTS
PART C: FLUX TO SIMULINK MODEL
45
8. Definition of the Simulink model ................................................................................ 47

8.1
Description of the Simulink model................................................................................. 47
8.2
Definition of the coupling block ..................................................................................... 48
9. Solve ......................................................................................................................... 49
PART D: RESULTS
51
10. Simulink Results........................................................................................................ 53

10.1 No load torque............................................................................................................... 53
10.1.1 150 rad per second..........................................................................................................53
10.1.2 100 rad per second..........................................................................................................55
10.1.3 30 rad per second............................................................................................................56
10.2 With load torque ............................................................................................................ 58

10.2.1 Mechanical quantities ......................................................................................................58
10.2.2 Comments .......................................................................................................................58
10.2.3 Electrical quantities..........................................................................................................59
11. FLUX Results ............................................................................................................ 61

11.1 No load torque............................................................................................................... 61
11.1.1 150 rad per second..........................................................................................................61
11.1.2 100 Rad per second ........................................................................................................63
11.1.3 30 rad per second............................................................................................................65
11.2 With load torque ............................................................................................................ 68

11.2.1 Mechanical quantities at 150 rad per second ..................................................................68
11.2.2 Electrical quantities at 1500 rpm .....................................................................................69
12. Conclusion ................................................................................................................ 71
PAGE C
TABLE OF CONTENTS
PAGE D
FLUX
FLUX
PART A: FLUX MODEL
PART A: FLUX MODEL
PAGE 1
PART A: FLUX MODEL
PAGE 2
FLUX
FLUX
PART A: FLUX MODEL

Geometry
1.
1.1
Geometry
Overview of the geometry
Our sample problem consists of a 4-pole, 3-phase, 36-slot, 28-bar induction motor. Because of
the motors periodicity, we will model only of it (1 pole). Our model consists of 9 stator slots
and 7 rotor bars. The air-gap is set to 0.25 mm. The figure below is a diagram of our model.
Figure 1.1: Diagram of of the motor
PAGE 3
PART A: FLUX MODEL

Geometry
1.2
FLUX
Stator geometry
1.2.1
Geometrical parameters
The geometrical parameters used in the geometry part are presented in Table 1.1.
Stator Parameter Name
Comment
Value (mm)
AIRGAP
Air-gap width
0.25
SOD
stator outer diameter
170
SID
stator inner diameter
117
SSHEIGHT
Stator slot height
13
SSOPEN
Stator slot opening
3.8
SSBR
stator slot bottom radius
3.6
Table 1.1: Data to define Stator Parameters.
Note:
To create parameters, coordinate systems, points, lines and geometrical transformations:
double click on their name in the data tree.
1.2.2
Coordinate systems
The geometry of the motor is described using the coordinate system presented in Table 1.2.
Name
Type
Coordinate
of system
Comment
system
coordinat
of definition
e
STATMAIN
main stator
2D GLOBAL
coordinate system
STATWORK
STATLOC
working system
local system
LOCAL
LOCAL
STATMAIN
STATWORK
Type of
coordinates
X
Y
coor coor Z rot.
d
d
CARTESIAN 2D
CARTESIAN 2D
CARTESIAN 2D
0
sid/2
0
0
0
0
Table 1.2: Stator coordinate systems.
PAGE 4
FLUX
PART A: FLUX MODEL

Geometry
1.2.3
Points and lines for the upper half of the stator slot
The description of the stator starts from the geometry of half of a stator slot whose points defined
in STATLOC coordinate system are presented in Table 1.3.
Point
X coordinate
Y coordinate
P1
P2
SSHEIGHT
P3
SSOPEN/2
P4
SSHEIGHT-SSBR
SSBR
Table 1.3: Coordinates of points for upper half of the stator slot.
After the creation of these points we will create lines by connecting them as presented in Table
1.4:
Line
Type of line
Starting point End point Arc radius
L1
Segment defined by 2 P3
points
P4
L2
Arc defined by its radius, P2

starting and ending point
P4
SSBR
Table 1.4: Lines of stator base geometry.
You should see the screen below:
Figure 1.2: The upper half of the stator slot
PAGE 5
PART A: FLUX MODEL

Geometry
1.2.4
FLUX
Geometric transformations
The geometrical transformations in Table 1.5 and Table 1.6 are needed to complete the geometry
of the stator.
Geometric
transformation
SMIRROR
Comment
Type of geometric
transformation
Mirror image of
half stator slot
Affine transformation
with respect to a line
defined by 2 points
First
point
P1
Second
point
Scaling factor
P2
-1
(line symmetry)
Table 1.5: Mirror transformation.
Geometric
transformation
Comment
Type of
geometric
transformation
Coordinate
system
R
Theta
comp. comp.
Rot.
Z
SODUPLI
Slot duplication
Rotation defined
by angles and
pivot point
coordinates
STATWORK
90
SDUPLI
Stator side
duplication
Rotation defined
by angles and
pivot point
coordinates
STATWORK
10
Table 1.6: Rotational transformations.
1.2.5
Completing the stator geometry
We will use the first transformation, SMIRROR, to duplicate half the stator slot, thus
producing the first stator slot. Using the menu: Choose Actions, Propagate, Propagate Lines.
We will propagate lines L1 and L2 using SMIRROR.
After that, we will create a line to close the outline of the stator slot by connecting points P5
and P3. This line is a small arc based on the inner radius of the stator as it is indicated in Table
1.7.
Choose: Data, Add, Line from the menu:
Line
L5
Type of line
Arc defined by its P5
radius, starting and
ending point
Starting
point
End
point
P3
Arc
radius
SID/2
Table 1.7: Line to close the outline.
Now that the slot is closed, that is, the points have been properly connected, the face of the slot
can be generated.
PAGE 6
FLUX
PART A: FLUX MODEL

Geometry
Note:
Remember that faces are automatically generated in PREFLUX:
in the toolbar.
Click the Build Faces icon
You should see the face of the first stator slot as shown below:
Figure 1.3: The face base geometry
Next, we need to modify the local coordinate system to make sure the stator slots will be
properly aligned. Using the data tree menu to modify STATWORK coordinate System. Then
we choose a Cartesian coordinate system whose origin is [0, 0] and a rotation angle of 5
degrees.
Next we will apply the SDUPLI transformation 8 times to duplicate the first stator slot and
place them in the proper positions along the inner outline of the stator.
The proper icon to propagate the faces is
.
The building option Add faces and associated linked mesh generator should be selected.
Then, the stator geometry must be closed. A line must connect the upper left point of the first
slot and the bottom left point of the second slot (for the type and the arc radius of the line see
Table 1.7).
PAGE 7
PART A: FLUX MODEL

Geometry
FLUX
This line will be propagated 7 times by SDUPLI transformation, and we obtain the following
display:
Figure 1.4: The slots after duplication
Then the outer stator edges must be created.

We create first the bottom edge as the line connecting the two points (created in the coordinate
system STATMAIN) of Table 1.8.
Point
X coordinate
Y coordinate
P47
Sid/2
P48
Sod/2
Table 1.8: Extremities of the bottom edge.
This line will be propagated by SODUPLI. Then, the inner arc of the stators outline must be
completed connecting the bottom and top slots to the straight outer edges we have just created.
So we will connect points P47 and P5 (arc radius = sid/2) at the bottom and points
P42 and P49 (arc radius = sid/2) at the top. The menu to create the arcs is: Data,
Add, Line.
Note:
Remember that arcs must be created in an anticlockwise direction, so be careful to choose
the points in the order shown below (P47 and then P5; P42 and P49).
Finally to create the outer arc of the stator, we will connect points P48 and P50 by an arc
(arc radius = sod/2).
PAGE 8
FLUX
PART A: FLUX MODEL

Geometry
1.3
Rotor geometry
In the same way as for the stator we present here the data needed for the rotor geometry.
1.3.1
Geometrical parameters
Rotor parameter
name
Comment
Value (in millimeters)
RBHEIGHT
Rotor bar height
18
RBTOPR
Rotor bar top radius
2.75
RBBOTR
Rotor bar bottom radius
1.15
ROD
Rotor outer diameter
116.5
RBTOP
Rotor bar top location
110.26
RID
Rotor inner diameter
38
Table 1.9: Rotor parameters.
1.3.2
Coordinate systems
Name
Type
Comment
of system
ROTMAIN
Main
rotor 2D
coordinate
GLOBAL
system
ROTWORK
Working system LOCAL
ROTLOC
Local system
LOCAL
Coordinate
Type of
system of
coordinates
definition
X
coord
Y
Z
coor
rot.
d
CARTESIAN 2D 0
ROTMAIN
CARTESIAN 2D 0
ROTWORK
CARTESIAN 2D RBTOP/2
Table 1.10: Rotor coordinate systems
PAGE 9
PART A: FLUX MODEL

Geometry
1.3.3
FLUX
Points and lines for the rotor bar
The points are defined in the coordinates system ROTLOC.

Point
X coordinate
Y coordinate
P51
RBTOPR
P52
RBTOPR
P53
RBTOPR+RBBOTR-RBHEIGHT
RBBOTR
P54
RBTOPR-RBHEIGHT
Table 1.11: Rotor points.
Line
Type of line
Starting point End point Arc radius
L59
Segment defined by 2points
P53
P52
L60

P52
RBTOPR
L61

P54
RBBOTR
Table 1.12: Lines of stator base geometry.
1.3.4
Geometric transformations
Geometric
transformation
RMIRROR
Comment
Type of geometric
transformation
Mirror
transformation
for rotor bar
First
point
Affine transformation with respect P51

to a line defined by 2 points
Second
point
Ratio
P54
-1
Table 1.13: Mirror transformation.
Geometric
transformation
Comment
Type of
geometric
transformation
Coordinate
system
R
Theta
comp. comp.
Rot.
Z
ROTSIDE
Rotor side
duplication
Rotation defined
by angles and
pivot point
coordinates
ROTMAIN
90
RDUPLI
Bar duplication
Rotation defined
by angles and
pivot point
coordinates
ROTWORK
90/7
Table 1.14: Rotational transformations.
PAGE 10
FLUX
1.3.5
PART A: FLUX MODEL

Geometry
Completing the rotor geometry
After the creation of parameters, coordinate systems, points, lines and transformations we
complete the geometry as follows:
Use the first transformation, RMIRROR, to duplicate the half stator slot, thus producing the
first stator slot. We will propagate lines L60, L59 and L61 only once.
Generate the face of the rotor bar.
Then, your screen should display the rotor bar face and the stator as shown below:
Figure 1.5: The entire stator and a rotor bar
Modify the orientation of the rotor bar. Using the data tree menu to modify ROTWORK
coordinate System. Then we choose a Cartesian coordinate system which the origin is [0, 0]
and a rotation angle of 90/(7*2)] degrees.
Apply the RDUPLI transformation 6 times to create duplicates of the first rotor bar.
The building option Add faces and associated linked mesh generator should be selected.
PAGE 11
PART A: FLUX MODEL

Geometry
FLUX
Then, you should see the seven rotor bars in place as shown below:
Figure 1.6: Bars after duplication
Define points for rotor outlines. Define first points P93 and P94 which coordinates in the
coordinate system ROTMAIN are (rod/2, 0) and (rid/2, 0). These two points will be
connected by a segment (L101), which is going to be transformed by ROTSIDE.
Define the inner and outer outlines of the rotor, which will be two arcs connecting points P94
and P96, and points P93 and P95. The arc radius is respectively rid/2 and rod/2.
PAGE 12
FLUX
1.3.6
PART A: FLUX MODEL

Geometry
Closing the air-gap
To close the air-gap we will create two connecting segments, one between points P93 and P47
and the second between points P49 and P95.
Then we build the faces for the rotor and the air-gap:
If you click the Build Faces icon
in the toolbar, you should see the complete geometry with
the 19 faces constructed, as we show below:
Figure 1.7: Complete geometry
PAGE 13
PART A: FLUX MODEL

Geometry
1.4
FLUX
Add and assign regions for the faces
First, regions are created by entering names, comments (reflecting the material or source
properties, in this case) and colors. Using the menu : Choose Data, Add, Region Face.
Table 1.15 below indicates the name, comment and color to be entered for each region face of
our model.
Region Face
1
2
3
4
5
6
7
8
9
10
11
12
13
Region Face
Name
RB_1
RB_2
RB_3
RB_4
RB_5
RB_6
RB_7
Rotor
Airgap
SSA
SSB
SSC
Stator
Comment
Color
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Iron
Moving airgap
Plus A
Plus B
Minus C
Iron
Turquoise
Turquoise
Turquoise
Turquoise
Turquoise
Turquoise
Turquoise
Cyan
Yellow
Red
Magenta
Turquoise
Cyan
Table 1.16: Regions of faces.
Then, regions must be assigned for the faces. The figure below shows which features of the
geometry are assigned to each named region face.
Figure 1.8: correspondence between faces and regions
PAGE 14
FLUX
PART A: FLUX MODEL

Geometry
To assign faces to the regions we have created, from the Geometry menu:
Open the Assign Regions dialog with the icon
in the toolbar.
Finally, you should see the solid colored surfaces as shown in our figure below.
Figure 1.9: complete colored geometry
PAGE 15
PART A: FLUX MODEL

Geometry
1.5
1.5.1
FLUX
Mesh
Change to the Mesh context
The Mesh commands are available only in the Mesh context.

Add the Mesh points
In our example we will create additional mesh points to default mesh points that exist in
PREFLUX. This will give us better control over the mesh density across the geometry.
The following Table contains information to define the six mesh points:
Name of mesh point
Comment
Value (millimeters)
MSSBOT
Stator slot bottom
MRBBOT
Rotor bar bottom
2.5
MRBTOP
Rotor bar top
0.8
MAIRGAP
Moving airgap
0.5
MSOD
Stator outer diameter
9.5
MRID
Rotor inner diameter
Table 1.17: Mesh points.
Using the menu: choose Data, Add, Mesh point.
1.5.2
Applying the mesh points to the geometry
The mesh points are assigned as presented in Table 1.17:

Name of
mesh point
Points
MSSBOT
P4, P2, P6
MRBBOT
P53, P54, P55
MRBTOP
P52, P56
MAIRGAP
P51 and
all the points in RELATION with the Surfacic region AIRGAP
MSOD
P50, P48
MRID
P94, P96
Table 1.18: Assignation of mesh points.
PAGE 16
FLUX
1.5.3
PART A: FLUX MODEL

Geometry
Generate, verify and save the mesh
Mesh all lines of the geometry. Using the icon in the toolbar or choose Actions, Mesh Lines
and Faces, Mesh the Lines
Generate the surface elements of the mesh. By using the same command choose Mesh Faces.
Then save your project with a name of representing your FLUX2D problem; for example we
enter INDUCTION_PHYSIQUE.
The window seen will be as shown below:
Figure 1.10: complete colored mesh geometry
PAGE 17
PART A: FLUX MODEL

Geometry
PAGE 18
FLUX
FLUX
PART A: FLUX MODEL

Materials
2.
Materials
The B(H) dependence of stator and rotor magnetic cores is tabulated in Table 2.1.
B (T)
H (A/m)
0.50
1.10
1.60
129.50 243.25 1850.00
1.70
3700.00
1.85
9900.00
2.00
2.10
22100.00 43000.00
Table 2.1: Points for B(H) curve
The FLUX2D scalar-spline model is represented in Figure 2.1. Scalar spline model allows us to
define B(H) curve starting from experimental values of B and H. This curve represents the
interpolation of the values presented in Table 2.1 for the saturation value Js = 2.07 T.
Figure 2.1: B(H) curve of magnetic cores
Note:
The curve will be approximate by 2 values representing the saturation value Js = 2 T and
the relative slope a = 1100. This takes less time to calculate than the original curve.
PAGE 19
PART A: FLUX MODEL

Materials
FLUX
The properties of the materials used in this paper are summarized in Table 2.2.
Material
name
Iron
Comment
Nonlinear steel
Property
Isotropic
Model
B_scalar_a_sat
Permeability Iso_MU
Value
Js = 2.0 T
a=1100
Copper
Linear copper
Isotropic
Iso_RHO
resistivity
Scalar_cst
0.172.10-7m
Aluminu
m
Linear
aluminum
Isotropic
Iso_RHO
resistivity
Scalar_cst
0.278.10-7m
Table 2.2: Materials properties
PAGE 20
FLUX
PART A: FLUX MODEL

Definition of the electrical circuit
3.
The machine in our example is delta connected; its external circuit is shown below (see Figure
3.1), with the data corresponding to different components (see Table 3.1). The voltage sources
are defined as constant as they will be fully controlled from Simulink. Kirchoffs law will
deduce immediately the voltage of Phase C. Thus, there is no need to connect phase C with an
external source.
Figure 3.1: Electrical circuit
PAGE 21
PART A: FLUX MODEL

FLUX
Component
Type
name
VAC
Voltage
source
Model
Data
Constant
134.35 V rms
VBA
Voltage
source
Constant
134.35 V rms
PA
Coil
Total value
Number
of
turns:
Resistance: 0.46557 Ohm
132
PB
Coil
Total value
Number
of
turns:
132
MC
Coil
Total value
Number
of
turns:
132
Resis4
Resistor
Constant
0.5575 Ohm
Resis5
Resistor
Constant
0.5575 Ohm
Resis6
Resistor
Constant
0.5575 Ohm
Induc7
Inductance
Constant
0.0021 H
Induc8
Inductance
Constant
0.0021 H
Induc9
Inductance
Constant
0.0021 H
Constant
7 rotor bars
Ring resistance: 2.5.10-6
Ring inductance: 4.0.10-9H
Looping
type
of
the
displayed part: -1(anticyclic)
Squirrel_cage Squirrel cage
Table 3.1: Electrical components
Note:
The sources are defined as constant. Indeed, as it will be explained in chapter 2, the
machine will be controlled by the magnitude of the input voltages. There is no need then to
define the frequency or the phase of the sources
PAGE 22
FLUX
PART A: FLUX MODEL

Physical properties
4.
4.1
Physical properties
General information
The case has a constant cross section (plane problem) with a depth of 145 mm. It is solved with
the magneto-transient application.
4.2
Assign materials to the regions
There are three materials that should be assigned to regions as follows:

Name of
Material
the region
SSA
VACUUM
Property
Source
External circuit
SSB
VACUUM
Source
External circuit
SSC
VACUUM
Source
External circuit
RB1
ALUMINUM
Source
External circuit
RB2
ALUMINUM
Source
External circuit
RB3
ALUMINUM
Source
External circuit
RB4
ALUMINUM
Source
External circuit
RB5
ALUMINUM
Source
External circuit
RB6
ALUMINUM
Source
External circuit
RB7
ALUMINUM
Source
External circuit
AIRGAP
Y Rotational Mechani Constant

air gap
c values
ROTOR
and
STATOR
IRON
Source
Model
Data
J = 0.02 Kg/m
F = 0.001 N.m
Number of pole pairs: 2
No source
Table 4.1: Materials and correspondent region
PAGE 23
PART A: FLUX MODEL

Physical properties
FLUX
You will see the boundary conditions that are applied automatically:
Figure 4.1: boundary conditions (automatically assigned)
4.3
Electrical circuit
The different electrical components are described in the second chapter.
PAGE 24
FLUX

Physical properties
PAGE 25

Scalar Control
PAGE 26
FLUX
FLUX

Scalar Control
5.
5.1
5.1.1
Scalar Control
Principle of the scalar control

Introduction
The principle of this type of control is to adjust the motor speed Vs by varying the output
frequency fs such as the magnetic state of the machine is about fixed, through the preservation of
s in a constant value, and such as the torque follows the wished law of variation according to
the speed. This type of command does not allow to control the electric and magnetic transients,
thus it must be operated only for laws of control with slow dynamics for which we can consider
that the machine keeps on electric and magnetic steady state.
The simplicity of this type of control favors its implementation in many industrial variators
conceived for these types of application. Indeed, it is characterized by a simplified structure, but
requiring a speed sensor and a speed controller for position servo-control.
5.1.2
Scalar control modeling
We will use the vector expressions of the asynchronous machine in a coordinate system, which is
bound to stator:
dS
VS = RS I S + dt
(1)
V = 0 = R I + d R j
R R
R
R
dt
elm = p p Lm Imag[ I S I R* ]
in an electrical sinusoidal steady state we obtain the following model :
VS = RS I S + j S S
VR = 0 = RR I R + j S R j R
(2)
PAGE 27

Scalar Control
FLUX
and with the relations :
S = R ,
S = LS I S + Lm I R
R = Lm I S + LR I R
(3)
the model can be written also as :

Lm
1
VS = S ( S L R ) + j S S
0 = Lm 1 + j + ( 1 j )
S
S
R
R
LS R
R
elm = p p
(4)
Lm
Imag[ S R* ]
LS LR
to simplify the expressions we will introduce the following arguments:
= Arctg ( R R ) , so
sin =
= Arctg ( R R ) , so
sin =
R R
1 + ( R R )
and cos =
R R
and cos =
1 + ( R R ) 2
1
1 + ( R R ) 2
1
1 + ( R R ) 2
(5)
and some relations can be deduced from the first model:

L
j R R
IR = m
I S or
I R = j R R ,
LR 1 + j R R
RR
Lm
R =
e j I S
2
1 + ( R R )
IR =
Lm
R R
e j ( + / 2 ) I S
2
LR 1 + ( R R )
(6)
also from the second model we can deduce:

R =
Lm
1
e j S
2
LS 1 + ( R R )
elm = p p
1 Lm 2
R R
( )
2S
2
LR LS 1 + ( R R )
delm
1
which is zero for the value R =
d R
R
1 Lm 2 1 2
(elm ) max = p p
( )
S
so we can deduce the maximal torque:
LR LS 2
the last relation allows to calculate
PAGE 28
(7)
(8)
(9)
FLUX
5.1.3

Scalar Control
Relations used for the scalar control
The scalar control may allow to obtain, for a given flux s, a given torque in a given speed .
From the precedent expressions (8) we can establish the equation:
1 L
[( R ) 2 elm ] R2 [ p ( m ) 2 2S R ] R + elm = 0
(10)
LR LS
the solution of this equation is :
1 Lm 2 2
pp
( ) S R
2 2
LR LS
[( R ) elm ]
R =
1 1 4
(11)
1 Lm 2 2 2
2[( R ) 2 elm ]
[ pp
( ) S R ]
LR LS
which gives the necessary pulsation for this functioning :
S = R + p
(12)
and the current absorbed to obtain the torque and the flux desired :
IS =
1
LS
1 + ( R R ) 2
1 + ( R R )
e j ( ) S
(13)
so we can deduce the voltage to apply :

VS = RS I S + j S S
These relations show that:
We can determine, for a given flux s, the necessary voltage, in order to obtain a wished
torque in a given frequency f S =
S
in the electrical steady state.
2
Neglecting the voltage drops RS I S we have

VS j S S
so
VS S S = 2f S S
and consequently:
VS
= 2 S = constant
fS
(14)
(15)
This explains the name given generally to this type of control.
PAGE 29

Scalar Control
5.2
FLUX
Structure of the Scalar control
The scalar control assures, in permanent speed, the flux module control. It is combine with an
automatic drive where:
ws = wr + wm (w is the angular electric speed in rad/sec)
ws is the stator pulsation
wr is the rotor pulsation
wm is the mechanical speed
The basic structure is represented on the following figure:
Bridge
Filter
Inverter
MAS
PWM
VS
s
m
ref
Figure 5.1: Structure of control with variable frequency in open-loop
One of the main difficulties is the determination of the values of the Vs(fs) table for low
frequencies in order to take into account correctly the influence of the resistive term because this
one varies rapidly with:
Is , that is with the desired torque in low speeds, and especially during start-up.
Rs ,that is with the thermal state of the machine.
With some approximations we can have:
VS S S (
RS 2
) +1
LS S
(16)
This simple structure is satisfying only with slow dynamics because the law of variation of Vs is
established in the steady state. To overcome this problem we can complete the precedent
structure with a speed loop (see Figure 5.2).
PAGE 30
FLUX

Scalar Control
We complete the precedent structure, with a speed loop.

The structure is represented on the following figure:
Bridge
Inverter
Filter
MAS
PWM
VS
S
m
r
Speed
controller
Ref
Figure 5.2: Structure of control in close-loop
Note:
The scalar control allows, in steady state, to minimize the input-current for a constant
torque.
PAGE 31

Scalar Control
PAGE 32
FLUX
FLUX

Simulink model
6.
6.1
Simulink model
Definition of the Simulink model
In the following is presented the whole Simulink model. The coupling with FLUX is detailed in
chapter 3.
Figure 6.1: Whole Simulink model
The model includes:

The command, which is a scalar control. (medium grey blocks)
The induction machine. For this block we can use a model deduced from the equations of the
machine, use a model with S-functions or use the machine block of Simulink library. In our
example we will use the model based on S-functions. (dark grey block) in the Concordia
frame.
A filter to soften the reference values.
A PI regulator for machines servo-control.
The outputs to be displayed.
Some accessories to have specific measures (light grey blocks).
PAGE 33

Simulink model
6.2
6.2.1
FLUX
Definition of the blocks

The command
This part controls the machine. It will control the value of the 3-phase voltage source of the
machine (i.e. the amplitude and the pulsation).
Figure 6.2: Command part of Simulink model
6.2.2
Subsystem of the Instruction block
Figure 6.3: Subsystem : Instruction
6.2.2.1
Generation of the pulsation
The blocks parameters of the pulsation is defined as follows:
The final value ws of the step block must be the value

of the reference pulsation. For example, ws=50rad/sec
so Ws=477rpm.
Figure 6.4: step block
PAGE 34
FLUX

Simulink model
6.2.2.2
Filter
A filter is required to soften the reference value of the speed with the transfer function equal to:
F=
1
;
s
s
+ +1
400 10
Figure 6.5: Filter block
6.2.3
Speed controller
In this part, the use of PI controller is necessary for the asynchronous machine servo-control. In a
empirical way and with some computation tries, we can determine some PI regulators
coefficients. In the structure shown above, we can see the detail of speed controllers that ensure
the servo-control from the reference control variables.
With: P=25 and I=0.02
Figure 6.6: PI Regulator block
PAGE 35

Simulink model
6.2.4
FLUX
Subsystem of the Scalar control
Inside the subsystem of scalar control, we have the following model, which use some defined
functions:
Figure 6.7: Subsystem scalar control
The stator pulsation of the induction machine is the result of the addition of the reference
pulsation (or the rotor pulsation) and the mechanic pulsation.
In consequence, this command enslaves the stator pulsation to the motor angular velocity by
controlling the rotor pulsation wr. By the way of a gain representing the statoric flux we have a
proportional voltage amplitude (see Paragraph 1.1.3).
The value of the statoric flux is given by the program containing the parameters of the machine.
The defined functions used in the scalar control are in the following path in the Simulink library
browser:
Figure 6.8:the Simulink function Fcn
PAGE 36
FLUX
6.2.5

Simulink model
The induction machines block:
It is a subsystem who has voltage sources as

inputs, and currents, flux, speed and
electromagnetic torque as outputs. The machine is
also connected to a load torque input.
Figure 6.9: The induction machines subsystem
In our example we use the model based on S-functions. Inside the inductions machine
subsystem we find the following blocks:
Figure 6.10: The interior of the induction machines subsystem
PAGE 37

Simulink model
FLUX
We can see the mechanical system represented by the transfer function, and the S-function,
which use a Matlab program. The name of the S-function program and the machines parameters
used in this program must be mentioned in the S-functions block parameters as follows:
Figure 6.11: The S-function block parameters
PAGE 38
FLUX

Simulink model
The Matlab program used masyn.m is the following is the same as in chapter B
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
/////////////////////////////////////////////////////////////////////////////////
/// Asynchronous machine in the Concordia frame ///
////////////////////////////////////////////////////////////////////////////////
Inputs :
u(1)=Vsalpha
u(2)=Vsbeta
u(3)=w=pOmega
States :
x(1)=Fsalpha
x(2)=Fsbeta
x(3)=Fralpha
x(4)=Frbeta
Outputs :
y=[Fsalpha,Fsbeta,Fralpha,Frbeta,Isalpha,Isbeta,Iralpha,Irbeta,Couple]
Parameters :
Rs,Ls,Rr,Lr,Lm,p
function [sys,x0,str,ts] = masyn(t,x,u,flag,Rs,Ls,Rr,Lr,Lm,p)
% some coefficients used afterward

a2=Rs*Lm/(Ls*Lr*sig);
a1=-Rs/Ls-Lm/Ls*a2;
b1=-Rr/Lr/sig;
b2=-b1*Lm/Ls;
switch flag,
% Initialization %
case 0,
[sys,x0,str,ts]=mdlInitializeSizes;
% Derivatives %
case 1,
sys=mdlDerivatives(t,x,u,a1,a2,b1,b2);
% Outputs %
case 3,
sys=mdlOutputs(t,x,u,Ls,Lr,Lm,sig,p);
case { 2, 4, 5, 9 }
sys = [];
PAGE 39

Simulink model
FLUX
%=============================================================================
% mdlInitializeSizes
% Return the sizes, initial conditions, and sample times for the S-function.
%=============================================================================
function [sys,x0,str,ts]=mdlInitializeSizes
sizes = simsizes;
sizes.NumContStates = 4;
sizes.NumDiscStates = 0;
sizes.NumOutputs = 9;
sizes.NumInputs = 3;
sizes.DirFeedthrough = 0; %0 because the input u doesn't intervene directly in the outputs computation
sizes.NumSampleTimes = 1; % at least one sample time is needed
sys = simsizes(sizes);
% initialize the initial conditions
x0 = [0 0 0 0];
% str is always an empty matrix
str = [];
% initialize the array of sample times
ts = [0 0]; % continuous time
% end mdlInitializeSizes
%
%=============================================================================
% mdlDerivatives
% Return the derivatives for the continuous states.
%=============================================================================
function sys=mdlDerivatives(t,x,u,a1,a2,b1,b2)
x5=x(1);
x6=x(2);
x7=x(3);
x8=x(4);
dx5=a1*x5+a2*x7+u(1);
dx6=a1*x6+a2*x8+u(2);
dx7=b1*x7+b2*x5-u(3)*x8;
dx8=b1*x8+b2*x6+u(3)*x7;
sys = [dx5 dx6 dx7 dx8];
% end mdlDerivatives
%
%=============================================================================
PAGE 40
FLUX
6.2.6

Simulink model
Outputs
Figure 6.12: Outputs
We can visualize all outputs that we need by scopes or workspaces. The Scope allows a
simultaneous visualization, whereas the workspace allows the transfer of results from Simulink
to Matlab where they can be easily postprocessed. In Figure 6.12, we have an illustration of the
use of the two methods.
6.2.7
Other blocks
Figure 6.13: Selector block
Selector: Select or re-order the specified elements of an input vector or matrix. It is available in
the Simulink library browser.
Figure 6.14: Concordia subsystem block
Concordias transformation: performs the (a,b,c) to (,) transformation on a set of threephase signals.
PAGE 41

Simulink model
PAGE 42
FLUX
FLUX

Solve
7.
Solve
To initialise the Simulink model, we need a Matlab program which supplies the machines
parameters to the Simulink model. The parameters are computed in the Flux case.
The program is as follows:
disp(data loaded) ;
PAGE 43

Solve
FLUX
Before starting the solving, the simulation parameters should be defined:
Figure 7.1: Simulation parameters
PAGE 44
FLUX
PAGE 45
PAGE 46
FLUX
FLUX

8.
8.1
Description of the Simulink model
The whole Simulink model looks as follows:
Figure 8.1: Simulink model of the coupling
The model includes:

A coupling with FLUX (2D) (V8) block: it refers to FLUX during the computation.
The command: it is the same command used in the Simulink model in part B. It commands
the value of the voltage sources needed by the FLUX case. It will supply sinusoidal voltages.
It need only to supply the two first phases a and b because, as it is explained in paragraph
3 of Part A, the electrical circuit of the Flux case needs only two voltage sources, the third
phase will automatically have the correspondent value. Connect the third phase with a
terminator in order to avoid warning messages concerning unconnected output ports.
Moreover, a gain of is added to take into account the fact that only a quarter of the
machine is simulated.
A PI regulator for machines servo-control, whose value are P=2 and I=0.142857.
The inputs may also include a load torque of 20 N.m (will be used in the second part of the
results).
The outputs to be displayed (see Part B).
A gain to convert the speed from rpm to rad/sec is also added (its value is pi/30).
PAGE 47

8.2
FLUX
Definition of the coupling block
This block enables a direct co-simulation with both FLUX (2D) and Matlab-Simulink. It is
available in the Simulink Library Browser, in the folder flux_link.
It is defined by:
the TRA file name: it is the TRA file that will be solved, we must give its name without the
extension .TRA. In our example: Motor (or your problems name).
FLUX (2D) inputs: the voltage sources VBA and VAC should be defined as inputs to FLUX
(2D). The syntax to use is described in the users guide.
[VOLTAGE:VBA; VOLTAGE:VAC]
Note:
The components names correspond to the name given to components in Table 3.1 of Part
A. Do not forget to check that it corresponds to your circuit.
FLUX (2D) outputs: the mechanical values are displayed (torque and angular velocity) as well
as the electrical values (current in phase 1, 2 and 3 and in the rotor bars 40 and 41).
[CURRENT:B1;CURRENT:B2;CURRENT:B3;TORQUE;OMEGA;CURRENT:BAR40;
CURRENT:BAR41]
the time step: 1e-3
the initial conditions: there is no initial conditions to set
the initialized by a static computation case must be ticked off
Figure 8.2: Coupling with FLUX2D block
Note:
A time step equal to 1ms is sufficient for 20steps per period because the computation
frequency is 50Hz.
PAGE 48
FLUX

Solve
9.
Solve
Note:
There is no need to open FLUX (2D) to solve the case. The simulation can be handled
directly in Simulink. But the .TRA file must be in the same folder as the Simulink model and
data programs (.mdl and .m files).
The computation time step for FLUX (2D) must be the same has been defined in the Coupling
with FLUX (2D) block.
Before starting the solving, the computation range should be defined.
0.5
Figure 9.1: Simulation parameters
PAGE 49

Solve
PAGE 50
FLUX
FLUX
PART D: RESULTS
PART D: RESULTS
PAGE 51
PART D: RESULTS
PAGE 52
FLUX
FLUX
PART D: RESULTS
Simulink Results
10. Simulink Results
Several simulations have been computed for various speed instructions; for 150rad/sec,
100rad/sec and 30rad/sec and for a given flux equal to 1.71wb in a given frequency, in order to
V
obtain S = 2 S = constant , in the electrical steady state. The problem was computed with a PI
fS
regulator with the following tuning: P=25; I=0.02
10.1
No load torque
10.1.1
150 rad per second
10.1.1.1
Mechanical quantities
Figure 10.1: Angular velocity and Electromagnetic torque
PAGE 53
PART D: RESULTS
Simulink Results
10.1.1.2
FLUX
Electrical quantities
Figure 10.2: Stator voltages
Figure 10.3: Stator currents
10.1.1.3
Comments
The first graph (Figure 10.1) shows the machine's speed reaching 150 rad per second and the
electromagnetic torque of the machine.
Because the stator is fed directly by the voltage sources, a transient torque is observed.
However, this noise is not visible in the speed because it is filtered out by the machine's inertia,
but it can also be seen in the stator currents, which are observed above (Figure 10.3).
The second graph (Figure 10.2) shows that the range of the voltage sources evolves in the
same way than the speed transient s (t). If we superimpose them on a same scale, we obtain
an identical style, namely the ratio Vs/ s is constant.
Note:
As expressed in chapter B, we can see that the scalar control allows, in steady state,
minimising the input-current for a constant torque.
PAGE 54
FLUX
PART D: RESULTS
Simulink Results
10.1.2
100 rad per second
10.1.2.1
10.1.2.2
PAGE 55
PART D: RESULTS
Simulink Results
FLUX
10.1.3
30 rad per second
10.1.3.1
PAGE 56
FLUX
10.1.3.2
PART D: RESULTS
Simulink Results
PAGE 57
PART D: RESULTS
Simulink Results
10.2
FLUX
With load torque
A simulation was computed with a load torque T=20N.m at 0.7sec. The results have been
simulated only for 100rad per second. The torque value is defined on simulink interface.
10.2.1
10.2.2
Comments
We can see the torque oscillations in reply to the applied load torque at t = 0.7seconds.
However, this oscillations were not very visible in the speed because it is filtered out by the
machine's inertia, but it can also be seen in the stator currents, which are plotted next (Figure
10.12).
PAGE 58
FLUX
10.2.3
PART D: RESULTS
Simulink Results
PAGE 59
PART D: RESULTS
Simulink Results
PAGE 60
FLUX
FLUX
PART D: RESULTS
FLUX Results
11. FLUX Results
With Simulink, only the values defined as outputs will be displayed, whereas with FLUX, the
computation gives far more results. Indeed, all the quantities usually reachable with FLUX can
be displayed and computed, for example; the Spectrum of the flux density, the equiflux lines for
each time step, the position, etc.
Note:
Theses simulations were computed with PI regulators adjustment P=2 & I=0.142857.
11.1
No load torque
11.1.1
150 rad per second
11.1.1.1
Magnetic quantities
Figure 11.1: Flux density distribution at time step 49msec
PAGE 61
PART D: RESULTS
FLUX Results
11.1.1.2
FLUX
Figure 11.2: Angular velocity (=1432rpm) & Electromagnetic torque
11.1.1.3
Figure 11.3: Stator voltage (VAC)
PAGE 62
FLUX
PART D: RESULTS
FLUX Results
11.1.2
100 Rad per second
11.1.2.1
Figure 11.5: Angular velocity(954rpm) & Electromagnetic torque
PAGE 63
PART D: RESULTS
FLUX Results
FLUX
Figure 11.6: Position
11.1.2.2
Figure 11.7: Stator voltage VAC
PAGE 64
FLUX
PART D: RESULTS
FLUX Results
11.1.3
30 rad per second
11.1.3.1
Magnetic quantities
Figure 11.9: Equiflux lines for time step 0.1sec
PAGE 65
PART D: RESULTS
FLUX Results
11.1.3.2
FLUX
PAGE 66
FLUX
11.1.3.3
PART D: RESULTS
FLUX Results
PAGE 67
PART D: RESULTS
FLUX Results
11.2
FLUX
With load torque
A load torque equal to 20N.m is applied at t=1.2second.
11.2.1
Mechanical quantities at 150 rad per second
PAGE 68
FLUX
PART D: RESULTS
FLUX Results
Figure 11.14: Friction torque
11.2.2
Electrical quantities at 1500 rpm
PAGE 69
PART D: RESULTS
FLUX Results
FLUX
11.2.2.1
Comments
In first time, we notice that the range of curves in Simulink and FLUX are not very the same and
that because of the regulators tuning. Because FLUX take saturation and no-linear phenomena
into account we have made different regulators adjustment which allows a good enslavement.
Some fluctuations of speed are observed in transient speed for low reference.
The lower the speed, the higher the torque oscillations. Then, it is more difficult to allows a good
control of the torque for slow dynamic speed.
The coupling with FLUX enables to see that a proper servo-control of the torque is more difficult
for lower speed while a simple SIMULINK model does not account properly for this difficulty.
PAGE 70
FLUX
PART D: RESULTS
Conclusion
12. Conclusion
The scalar control of induction machines which gives some good results in steady state, is less
efficient during transient period accompanying variation speed.
This fact is also noticed at the starting period, when the motor is in low frequency speed and the
drop in voltage in the stators resistors are not negligible any more compared with fem.
Theses phenomena shows the interest of a direct co-simulation which permit to take into account
precisely overall of the magnetic effect (non-linear phenomena, Eddy currents)
PAGE 71
PART D: RESULTS
Conclusion
PAGE 72
FLUX

FLUX ® 2D Application Scalar Command of An Induction Machine Technical Paper

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

FLUX ® 2D Application Scalar Command of An Induction Machine Technical Paper

Hochgeladen von

Copyright:

Verfügbare Formate

CAD Package for Electromagnetic and Thermal

Analysis using Finite Elements

Copyright - September 2004

FLUX is a registered trademark.

FLUX2D's Quality Assessment

This technical paper was edited on 24 September 2004.

PART A: FLUX MODEL

Overview of the geometry ............................................................................................... 3

Geometrical parameters ....................................................................................................4

Rotor geometry ............................................................................................................... 9

Geometrical parameters ....................................................................................................9

Add and assign regions for the faces............................................................................ 14

Change to the Mesh context............................................................................................16

Assign materials to the regions ..................................................................................... 23

Electrical circuit ............................................................................................................. 24

SCALAR COMMAND OF AN INDUCTION MACHINE

PART B: SIMULINK MODEL

Principle of the scalar control.........................................................................................27

Structure of the Scalar control .......................................................................................30

6. Simulink model .......................................................................................................... 33

Definition of the Simulink model ....................................................................................33

Definition of the blocks...................................................................................................34

The command ..................................................................................................................34

SCALAR COMMAND OF AN INDUCTION MACHINE

PART C: FLUX TO SIMULINK MODEL

8. Definition of the Simulink model ................................................................................ 47

Description of the Simulink model................................................................................. 47

Definition of the coupling block ..................................................................................... 48

10. Simulink Results........................................................................................................ 53

10.2 With load torque ............................................................................................................ 58

11. FLUX Results ............................................................................................................ 61

11.2 With load torque ............................................................................................................ 68

12. Conclusion ................................................................................................................ 71

SCALAR COMMAND OF AN INDUCTION MACHINE

SCALAR COMMAND OF AN INDUCTION MACHINE

PART A: FLUX MODEL

PART A: FLUX MODEL

SCALAR COMMAND OF AN INDUCTION MACHINE

PART A: FLUX MODEL

SCALAR COMMAND OF AN INDUCTION MACHINE

PART A: FLUX MODEL

Overview of the geometry

Figure 1.1: Diagram of of the motor

SCALAR COMMAND OF AN INDUCTION MACHINE

PART A: FLUX MODEL

stator outer diameter

stator inner diameter

Stator slot height

Stator slot opening

stator slot bottom radius

Table 1.1: Data to define Stator Parameters.

Table 1.2: Stator coordinate systems.

SCALAR COMMAND OF AN INDUCTION MACHINE

PART A: FLUX MODEL

Starting point End point Arc radius

Arc defined by its radius, P2

Table 1.4: Lines of stator base geometry.

You should see the screen below:

Figure 1.2: The upper half of the stator slot

SCALAR COMMAND OF AN INDUCTION MACHINE

PART A: FLUX MODEL

Table 1.5: Mirror transformation.

Table 1.6: Rotational transformations.