IEEE Power Electronics Magazine - June 2021 - 48

represent the synchronous servo motor and the position
encoder, respectively. The actual control operation is carried
out in the rotor-fixed coordinate system. Therefore, the
control loop can be set-up like the one used in dc machines.
The variable iwq
current and torque during operation. The variable iwd
is defined as the reference variable for the
is
set to zero during the operation. Other values allow the
motor to be operated in the field weakening range. The
motor current is detected and transformed into the rotorfixed
coordinate system with Clarke and Park transformation
aid. The variables isd
and isq
function as controlled
variables for the PI-controller. The controller's manipulated
variables are fed through a decoupling network and
transformed into the stator-fixed coordinate system. The
decoupling network ensures clean separation of ud
thereby facilitating the operation of the controllers.
and
u ,q
The stator-fixed variables can be used to generate control
signals for the IGBTs.
Figure 5 shows the experimental set-up for the case
studies. After getting the results for three-phase currents
Table 2. Technical data of synchronous servo motor.
Parameter
Armature resistance
Armature inductance
Moment of Inertia
Permanent-magnet flux
Number of poles
Rated torque
Rated power
Standstill torque
Standstill current
Reference Speed
Symbol
Rs
Ls
J
zm
P
Mn
Pn
To
Io
~r
Value
4.74Ω
8.6 e-3H
0.33 Kg.cm2
0.089Wb
8
0.97 NM
0.61 kW
1.2 Nm
2.99 A
1000 rpm
Novel Model Predictive Control of Fault-Tolerant
Synchronous Servo Motor
In this case study, a model is build applying NMPC to determine
the performance of the system under single phase open
circuit fault in an inverter. This investigation displays a shortcoming
examination of machine drives. The deficiency leeway
time is a couple of milliseconds, where the machine
activity stays constant, notwithstanding the flaw, and shows
the presentation in a lot more limited time than the existing
control techniques. NMPC is an advanced control technique
that handles difficult multivariable control problems. It controls
the multi-input multi-output process by satisfying
inequality constraints that impact the input and output variables.
If a feasible and accurate dynamic model of the process
is available, then the future output values can be easily
predicted using a model and current measurements. Both
predictions and measurements can also calculate the appropriate
changes in the input variables [11]. In NMPC, the
phase currents are predicted in abc reference frame instead
of d-q [12].
via simulation, the model has been deployed to the hardware,
and the implementation results can be seen using
Lucas Nulle's scope in few minutes, shown in video [9].
Figure 6(a) shows the simulation results, and Figure 6(b)
shows the implementation results for three-phase currents.
The Y axis on the experimental graph indicates the value
of current in milliamperes. Though the simulation graph
shows a slightly lower value of the standstill current, the
implementation graph shows the current's accurate value,
which is 2.99 Amperes.
Figure 7 shows the system's dynamic response when
speed changes from 500 rpm to 1000 rpm at 0.03 seconds. It
can be seen that the speed tracking time is very short.
Case Study Two
iwd
+
_
iwq
PI
Control
Decoupling
+
_
isd
isq
φel
isd
d, q
isq
Rotor-Fixed
FIG 4 Block diagram of the closed-loop current control.
48 IEEE POWER ELECTRONICS MAGAZINE z June 2021
φel
E
iα
α, β
α, β
iβ
a, b, c
ia
ib
ic
Stator-Fixed
PI
Control
Usq∗
Usq
α, β
uβ
a, b, c
Usd∗
Usd
d, q
uα
α, β
Three
Phase
VSI
M
IEEE Power Electronics Magazine - June 2021

Table of Contents for the Digital Edition of IEEE Power Electronics Magazine - June 2021
