33 529 2 PB

Applied and Computational Mechanics 3 (2009) 331338
Mathematical Modeling and Dynamic Simulation of a Class of

Drive Systems with Permanent Magnet Synchronous Motors
M. Mikhov
a,
a
Faculty of Automatics, Technical University of Soa, 8 Kliment Ohridski Blvd., 1757 Soa, Bulgaria
Received 30 August 2009; received in revised form 6 November 2009
Abstract
The performance of a two-coordinate drive system with permanent magnet synchronous motors is analyzed and
discussed in this paper. Both motors have been controlled in brushless DC motor mode in accordance with the
rotor positions. Detailed study has been carried out by means of mathematical modeling and computer simulation
for the respective transient and steady-state regimes at various load and work conditions. The research carried out
as well as the results obtained can be used in the design, optimization and tuning of such types of drive systems.
They could be also applied in the teaching process.
c 2009 University of West Bohemia. All rights reserved.
Keywords: two-coordinate drive, brushless DC motor control mode, dynamic simulation
1. Introduction
A number of applications in the mechanical industry require two-coordinate drive systems with
good static, dynamic and energetic characteristics. In recent years, permanent magnet syn-
chronous motors (PMSM) have been used in such cases, because they have some advantages,
including: compact form; low power loss and high efciency; high power/mass ratio; good heat
dissipation characteristics; low rotor inertia and good dynamics; high speed capabilities.
Applying some control methods, drive systems with such motors can combine the advan-
tages of both DC and AC motor systems. For example, the PMSM electric drives performance
improves signicantly if their control is executed according to the rotor position. Thus com-
mutation exibility is provided, avoiding the risk of synchronization failure in case of overload
[1, 2, 7, 8].
The complexity of electromechanical systems makes them difcult for description and study
in some cases. For this reason, mathematical modeling and computer simulation are widely
applied for the purpose of their analysis and synthesis. Such an approach provides for very good
conditions to investigate electric drives behavior in various transient and steady-state working
regimes, which is not always convenient or possible in industrial and laboratory environments
[3, 4, 5, 6].
This paper considers a two-coordinate electromechanical system with permanent magnet
synchronous motors, which are controlled in brushless DC motor mode. Detailed study has
been carried out by means of modeling and computer simulation for the respective dynamic
and static regimes at different operation modes. Behavior analysis has been made aiming at
improvement of the drive system performance.
Corresponding author. Tel.: +359 2 965 29 46, e-mail: mikhov@tu-soa.bg.

331
M. Mikhov / Applied and Computational Mechanics 3 (2009) 331338
2. Mathematical modeling of the drive system
2.1. Features of the drive system
A simplied block diagram of the drive system under consideration is represented in g. 1,
where the notations are as follows: PC position controller; SC1, SC2 speed controllers;
RC1, RC2 current reference blocks; CC1, CC2 three-phase current controllers; IC1, IC2
inverter control blocks; VI1, VI2 voltage source inverters; UR uncontrollable rectier; C
lter capacitor; PS1, PS2 position sensors; PF1, PF2 position feedback blocks; SF1, SF2
speed feedback blocks; M1, M2 motors; L1, L2 loads at the respective coordinate axes;
V
sr
1
, V
sr
2
speed reference signals; V
cr
1
, V
cr
2
current reference signals; V
pf
1
, V
pf
2
position
feedback signals; V
sf
1
, V
sf
2
speed feedback signals; V
cf
1
, V
cf
2
current feedback signals;
V
d
DC link voltage;
1
,
2
motor speeds;
1
,
2
angular positions; T
l
1
, T
l
2
load torques
applied to the respective motor shafts.
Fig. 1. Block diagram of the drive system under consideration
The motors used are characterized by having permanent magnet-produced elds on the ro-
tors and armature windings on the stators. Both subsystems have identical cascade structures
with subordinate regulation. The motors have been controlled in brushless DC motor mode in
accordance with the respective rotor positions. Control loops optimization and tuning have been
done sequentially, starting from the innermost ones.
2.2. Mathematical description
In order to obtain a suitable simulation model some assumptions have been made, such as:
motors are unsaturated;
eddy-currents and hysteresis effects have negligible inuence on the stator currents;
motors are symmetrical three-phase machines;
self and mutual inductances are constant and independent of the rotor position;
devices in the power-electronic circuits are ideal.
332
Fig. 2 shows the equivalent circuit diagram of PMSM and the voltage source inverter for
one coordinate axis. The corresponding notations are as follows: S1S6 power electronic
switches; D1D6 freewheeling diodes; v
a
, v
b
, v
c
voltages applied on stator phases a, b and
c respectively; i
a
, i
b
, i
c
phase currents; e
a
, e
b
, e
c
back EMF voltages; R
a
, R
b
, R
c
stator
phase resistances; L
a
, L
b
, L
c
phase self inductances; L
ba
, L
ca
, L
cb
mutual inductances.
Fig. 2. Equivalent circuit diagram of PMSM and the voltage source inverter
The vector-matrix model of such a set is described as follows:
v
a
v
b
v
c
R
s
0 0
0 R
s
0
0 0 R
s
i
a
i
b
i
c
L
a
L
ba
L
ca
L
ba
L
b
L
cb
L
ca
L
cb
L
c
d
dt
i
a
i
b
i
c
e
a
e
b
e
c
, (1)
where R
s
= R
a
= R
b
= R
c
is the stator phase resistance.
The back EMF voltage waveforms are expressed by the equation:
e
a
e
b
e
c
=
d
d
, (2)
where
a
,
b
,
c
are the stator magnetic uxes of the motor phases a, b and c respectively.
Since the motor windings are star connected (g. 2), the next relation is valid:
i
a
+ i
b
+ i
c
= 0. (3)
Assuming there is no change in the rotor reluctance with angular position, then:
L
a
= L
b
= L
c
= L; (4)
L
ba
= L
ca
= L
cb
= M. (5)
Taking into consideration Eqs. (4) and (5), the vector-matrix Eq. (1) is arranged as follows:
v
a
v
b
v
c
= R
s
i
a
i
b
i
c
+ L
s
d
dt
i
a
i
b
i
c
e
a
e
b
e
c
, (6)
where L
s
= L M is the stator phase inductance.
333
The motor electromagnetic torque can be expressed as:
T =
e
a
i
a
+ e
b
i
b
+ e
c
i
c
, (7)
and the mechanical dynamics equations are as follows:
J
d
dt
= T T
l
D; (8)
d
dt
= , (9)
where J is the total inertia referred to the respective motor shaft; D the viscous damping
coefcient.
3. Computer simulation and performance analysis
Every specic working regime of the two-coordinate drive system requires an appropriate con-
trol algorithm. For this reason, using the MATLAB/SIMULINK software package some com-
puter simulation models have been developed to analyze the system under consideration.
Detailed investigation has been carried out by means of mathematical modeling and com-
puter simulation for the respective transient and steady-state regimes at various working condi-
tions.
3.1. Phase currents formation
The current controllers have a programmable hysteresis band which determines the respective
modulation frequency.
Fig. 3 illustrates the principle of current pulses formation. The used notations are as follows:
i
r
reference phase current waveform; 2i
r
= i
max
i
min
reference hysteresis band; i
max
and i
min
maximum and minimum current values, respectively; T
m
= t
on
+ t
off
modulation
period.
Fig. 3. Principle of current pulses formation
a) b)
Fig. 4. Inuence of the hysteresis band
The modulation frequency is expressed by:
f
m
=
1
T
m
. (10)
The hysteresis band inuence of the rst current controller is shown in g. 4. It represents
the phase current waveforms for two different hysteresis bands (2i
r
1
and 2i
r
2
respectively).
As evident, narrowing the zone, i.e. increasing the chopping frequency will result in reduction
of the current pulsations.
The analysis carried out shows that this frequency is not of a constant value and depends on
the following factors:
334
hysteresis band;
reference motor speed;
power circuit electromagnetic time-constant;
initial current value.
This analysis allows selecting of the optimal hysteresis band in accordance with the respec-
tive power switches. In the research carried out the maximum modulation frequency has been
limited to a value of f
mmax
< 8 000 Hz.
Fig. 5 illustrates the dynamic maintenance of zero phase current in the respective interval
by means of pulse width modulation.
Fig. 5. Dynamic maintenance of zero phase current
Fig. 6 shows the phase current waveform i
a
1
and the respective trapezoidal back EMF volt-
age e
a
1
. In this case the simulation results have been obtained for a quasi-steady-state regime at
rated load applied to the motor shaft.
The three-phase current waveforms i
a
1
, i
b
1
, and i
c
1
for the same load torque are represented
in g. 7.
Fig. 6. Phase current and back EMF voltage Fig. 7. Three-phase current waveforms
335
3.2. Breaking regime
Various breaking modes have been investigated aiming at fast and efcient stopping and re-
versing. Some time-diagrams obtained at reverse speed control with electrical braking are
shown in g. 8, where T
b1
is the braking torque. The reference speed values in this case are
1r
= 157 rad/s. As evident, the braking mode applied ensures good dynamics with maxi-
mum reverse time less than 0.1 s.
3.3. Current limitation and speed stabilization
The phase currents limitation is provided through the current reference signals, formed by the
respective speed controllers. These controllers have been optimized in such a way, that the static
errors possibly caused by disturbances are eliminated.
Fig. 9 illustrates motor speed stabilization when the load torque changes. The respective
transient and steady-state regimes are represented, as well as the drive system reaction to a
disturbance expressed as load change.
Fig. 8. Reversing with electrical breaking Fig. 9. Current limitation and speed stabilization
In this case the load torque applied is equal to the rated value T
l1rat
and the respective
disturbance is T
l
1
= 0.5T
l
1
rat
. The load change is applied during the time interval of t =
(0.2 s 0.3 s). As evident, at an appropriate tuning of the speed controller, static speed error
does not appear when T
l
1
changes. The reference speed is
1r
= 157 rad/s and the starting
current is limited to the maximum admissible value of I
1 max
, which provides for a maximum
starting motor torque.
336
3.4. Two-coordinate position control
Two algorithms for trajectory formation have been investigated, namely simultaneous and con-
secutive movements along both coordinate axes.
Fig. 10 represents some simulation results obtained for two-coordinate position control with
consecutive motion along the coordinate axes. The reference distances are indicated with
1r
and
2r
respectively.
a) b)
Fig. 10. Time-diagrams for consecutive position control along both coordinate axes
Fig. 10a illustrates the realization of a reference motion along the rst coordinate axis. The
respective motor works at maximum speed of
1
=
1rat
, which ensures fast operation of the
drive system.
Fig. 10b shows the subsequent positioning along the second coordinate axis. In this case the
movement is carried out with a lower speed of
2
<
2rat
.
Fig. 11. Trajectory of simultaneous movement along both coordinate axes
337
A simultaneous movement along both coordinate axes is illustrated in g. 11, when the two
motor speeds are equal.
The motors used are identical, both with rated power P
rat
= 0.6 kW and rated angular speed
rat
= 157 rad/s.
4. Conclusion
The performance of a two-coordinate drive system with permanent magnet synchronous motors
has been discussed in this paper. Both motors have been controlled in brushless DC motor mode
in accordance with the respective rotor positions.
Using the MATLAB/SIMULINK software package a number of computer simulation mod-
els have been developed to analyze this class of drive systems. Detailed study has been carried
out for the dynamic and static regimes at various loads, disturbances and work conditions.
The analysis shows that the control approach applied ensures good performance, which
makes it suitable for a variety of applications in the mechanical industry.
The research carried out as well as the results obtained can be used in the design, optimiza-
tion and tuning of such types of drive systems. They could be also applied in the teaching
process.
Acknowledgements
This work has been supported by the Technical University of Soa under the research project
No. 091NI142-08/2009.
References
[1] Boldea, I., Nasar, S. A., Electric drives, CRC, Boca Raton, 1999.
[2] Chen, C., Cheng, M., A new cost effective sensorless commutation method for brushless DC
motors without phase shift circuit and neutral voltage, IEEE Trans. on Power Electronics, 2 (22)
(2007) 644653.
[3] Mikhov, M. R., Investigation of a permanent magnet synchronous motor control system, Technical
Ideas, 3/4 (38) (2001) 2334.
[4] Mikhov, M. R., Analysis and simulation of a servo drive system with hysteresis current con-
trol, Proceedings of the International Conference on Automatics and Informatics, Soa, Bulgaria,
2002, pp. 197200.
[5] Ong, C., Dynamic simulation of electric machinery using MATLAB/SIMULINK, Prentice Hall,
New Jersey, 1998.
[6] Sa, S. K., Acarnley, P. P., Jack, A. G., Analysis and simulation of the high-speed torque per-
formance of brushless DC motor drives, IEE Proceedings Electric Power Applications, 3 (142)
(1995) 191200.
[7] Sozer, Y., Torrey, D. A., Adaptive torque ripple control of permanent magnet brushless DC motors,
Proceedings on the Applied Power Electronics Conference, Anaheim, USA, 1998, pp. 8692.
[8] Yoon, Y., Kim, D., Lee, T., Choe, Y., Won, C., A low cost speed control system of PM brushless
DC motor using 2 Hall-Ics, Proceedings of the International Conference on Mechatronics and
Information Technology, Jecheon, Korea, 2003, pp. 150155.
338

33 529 2 PB

Uploaded by

Copyright:

Available Formats

33 529 2 PB

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

33 529 2 PB

Uploaded by

Copyright:

Available Formats

Applied and Computational Mechanics 3 (2009) 331338

Mathematical Modeling and Dynamic Simulation of a Class of

Corresponding author. Tel.: +359 2 965 29 46, e-mail: mikhov@tu-soa.bg.

You might also like