A Single-Phase Induction Machine Model for Real-

Time Digital Simulation

A. B. Dehkordi

Abstract— Most small power induction motors need to operate open phase. The power factor and efficiency degrade while the
with single-phase a.c. power supplies that are readily available at peak torque also decreases significantly [1]. Thus, except for
homes, and remote rural areas [1]. These machine are used to low powers (less than ¼ kW in general), the auxiliary winding
drive fans, pumps, air compressors, refrigeration compressors,
air conditioning fans and blowers, saws, grinders and office
is active also during running conditions to improve the
machines [2]. performance. The following are the main types of single-phase
This paper presents a real-time electromagnetic transient induction motors in use today:
model for single-phase induction machines (SPIM). The paper • Split-Phase Induction Motors
starts with the introduction and analysis of unsymmetrical 2- • Capacitor Induction Motors
phase induction machines. Although, main and auxiliary
windings are not symmetric, a technique has been introduced • Shaded-Pole Induction Motors
which facilitates the inclusion of magnetic saturation in this The split-phase motor, sometimes called resistance split-
model. The embedded phase domain approach [3] is utilized for phase, achieves its starting torque by having a higher
incorporation of this model into the network solution of the resistance and possibly lower reactance in the auxiliary circuit,
RTDSTM simulator. Various features of the model such as which is usually wound 90 electrical degrees from the main
embedded breakers are introduced in the paper. The model is winding. At a speed in the region of maximum torque, the
validated using stand-alone numerical simulations and analytical
solutions. Possible configurations of the model such as auxiliary winding is switched off. The switch may be activated
“capacitor-start induction motor” and “capacitor-start by speed (centrifugal), voltage, current, or temperature [2].
capacitor-run induction motor”, are demonstrated. Scenarios The starting torque of these motors is moderate, but the
including stalling of air-conditioning pumps during a fault in starting current is relatively high. Efficiency and power factor
distribution circuits are presented. are moderate, with efficiency ranging from about 40% at 0.1
hp to about 70% at 0.75 hp [1].
Keywords: real-time digital simulation, single-phase induction
machine, saturation, embedded phase domain approach, digital Capacitor induction motors are in the forms of “capacitor-
transient network analyzer, distribution network start motors”, “permanent-split capacitor motors” and “two
value capacitor motors”. In the “capacitor-start” motor the
I. INTRODUCTION starting torque is obtained by the use of a capacitor in series
with the auxiliary winding while starting, then switching the
S mall power motors in distribution networks normally have
to operate with single-phase a.c. power. These machines
are either series connected d.c. machines or single-phase
auxiliary winding out as the motor reaches running speed. The
capacitor causes the auxiliary winding current to lead the main
induction machines (SPIMs). Single-phase induction machines current. These machines are used for hard to start applications
are used to drive fans, pumps, air compressors, refrigeration such as pumps, compressors, etc. up to the rating of 5 hp [2].
compressors, air conditioning fans and blowers, saws, grinders A permanent-split capacitor motor is designed for applications
and office machines [2]. Air conditioning systems used in where starting torque requirements are low, but improved
households generally include a condenser unit, a compressor running performance is required. In this case, the motor is
unit and an air handler fan. Both the compressor unit and the designed to have a capacitor in series with the auxiliary
fan are run by single-phase induction motors. winding all the time [2].
To be self-starting, the induction machine needs a rotating In a “two-value capacitor motor”, a start capacitor is placed
field at zero speed. This in turn implies the presence of two in parallel with the run capacitor. This allows the motor to be
windings in the stator, while the rotor has a standard squirrel designed for optimum running efficiency without sacrificing
cage [1]. The stator first winding is called the main winding efficiency to get starting torque. Two-value capacitor motors
whereas the second winding is called the auxiliary winding have been available for years in the range of 1-10 hp [2].
[1]. SPIMs may run only on the main winding once they In “shaded-pole machines”, the auxiliary winding is usually
started on two windings. A typical case of single-phase single- a simple shorted turn of conductor around one side of each
winding IM occurs when a three-phase IM ends up with an stator pole, called a shading coil. Shaded-pole motors are
simple in construction and are therefore relatively low cost
and reliable. Starting current is relatively high, starting torque
A. B. Dehkordi is with RTDS Technologies Inc., Winnipeg, Canada (e-mail: is relatively low, running current is relatively high, and
dehkordi@rtds.com). efficiency and power factor are low [2]. They are widely used
to drive small fans (1/5 hp and below) because they are low in
Paper submitted to the International Conference on Power Systems
Transients (IPST2015) in Cavtat, Croatia June 15-18, 2015. cost and reliable.
Electric utilities have been having occurrences of delayed
voltage recovery following faults on the electrical system In this analysis, stator phase as is the main winding of the
when the load is comprised largely of loads driven by machine and stator phase bs is the auxiliary winding. As
induction machines. Under normal conditions, voltage previously mentioned, they are not identical and they do not
recovers to nominal levels in less than one second after the have the same parameters. The ratio of effective turns between
fault is cleared. In several cases, voltage recovery has been the auxiliary and main windings N bs N as is usually shown by
delayed for over 30 seconds after normal fault clearing in
the symbol a. The rotor windings are symmetric and have
some substations, especially when the air temperature and
similar parameters.
electrical system loading was high [5]. This fault induced
ibs
delayed voltage recovery or FIDVR, is being attributed to
stalling of air conditioner units. Mainly, stalling of the single- N bs vbs
phase induction motor operating the compressor unit. Delayed
θr N as
voltage recovery may lead to a system voltage collapse in the ωr bs
ias
worst case. One of the recent applications of the real-time
simulators has been on the modeling of distribution circuits vas
br
and interaction of air conditioning loads within these networks ar
[6]. as' as
This paper presents development of a single-phase ibr
vbr
induction machine model (SPIM) for real-time digital
Nr
simulation. The paper starts with the introduction and analysis ar' br'
of unsymmetrical 2-phase induction machines. Methods of Nr
transient simulation of single-phase induction machines are iar
discussed. Although, the main and auxiliary windings are not bs' var
symmetrical, a technique has been introduced which facilitates
Fig. 1. Diagram of an idealized two-phase induction machine.
the inclusion of magnetic saturation in this model. The
embedded phase domain approach [4] is utilized for B. Voltage and Flux Linkage Equations and
incorporation of this model into the network solution of the Equivalent Circuits
simulation. Various features of the model such as embedded Based on the above description of a SPIM, voltage and flux
breakers are introduced in the paper. The model is validated linkage equations of a single-phase induction machine are
using stand-alone numerical simulations and analytical developed. The following assumptions are made in this
solutions. Possible configurations of the model such as analysis:
“capacitor-start induction motor” and “capacitor-start • A multi-pole single-phase induction machine is
capacitor-run induction motor”, are demonstrated. Examples, modeled as an equivalent two pole machine. Torque
including stalling of air-conditioning pumps during a fault in and speed are monitored in p.u.
distribution circuits, are presented. • It is assumed that the machine windings produce a
sinusoidal MMF, thus space harmonics are ignored.
II. THEORY AND ANALYSIS OF A SPIM Voltage equations for a two-phase induction machine are
This section briefly describes the analysis and modeling presented in (1). Equation (1) shows the relation between
method of single-phase induction machines. Since the goal is voltage, current and flux linkage vectors of the machine.
incorporation of the model into an electromagnetic transient Stator phases a and b have different resistances due to the fact
program, the coupled electric circuit approach is used for the that stator windings are not symmetrical.
analysis of this electric machine. Some of the previous vabs  rs  
iabs Ψabs

  

contributions [7]-[9] are outlined and techniques offered in v   r        
0   0 0   ias   Ψ as 
this paper are described.  as    as        
vbs    0 rbs   0 0   ibs  d  Ψbs 
A. General Description of a Single-Phase Induction =       + (1)
v   0 0  r 0  i  dt  Ψ 
Machine  ar      r   ar   ar 
v   0 0   0 r  i  
 Ψbr 
As mentioned above, a single-phase induction machine  br    
 
 r    br 
   
consists of two stator windings known as the main and vabr  r 
 r 
 iabr Ψabr

auxiliary windings with the angular space of normally 90°. The relation between flux linkages and currents is shown in
The stator windings are usually asymmetric, i.e. the main and
(2). Here, Ψas and Ψ bs are flux linkages for stator windings a
auxiliary windings do not have the same number of turns and
parameters. The rotor has the standard squirrel cage, which in and b, and Ψar and Ψbr are flux linkages for rotor windings
rotating field theory can be represented by two d- and q-axis
respectively. [ Lss ] and [ Lrr ] are the inductance matrices for
windings. Fig. 1 shows the diagram of an idealized two-phase
induction machine. Windings as and bs represent stator stator and rotor windings, and [ Lsr ] is the matrix for the
windings and windings ar and br represent rotor windings. mutual inductances between the stator and rotor windings. The
The rotor angle is shown by θr and rotor speed is shown by ωr. values of inductances in (2) are functions of rotor position and
saturation. system components are discretized to a Norton equivalent of
 Ψabs   L  L  i  conductances and current sources which are passed to the
  =   ss 
 sr   abs  where: network solution during a communication interval (T0) every


 Ψ abr   L T L  i abr  time-step [11].
 sr  rr  
Interface vs Embedded Approach:
L + Lmas 0 
L  =  las  In electromagnetic transient programs, one approach of
 ss   0 Llbs + Lmbs  incorporating electric machines into the network solution is to
  (2) consider a machine an external sub-network to the main
L + L 0 
L  =  lr mr  system; receive the terminal voltages at every time-step and
 rr   0 Llr + Lmr  compute winding currents and inject them back to the network
 
solution. This method is called interfaced-based approach
M cos(θ ) −M sin(θ )
L  =  ar r ar r  [13], [14]. Another approach, referred to as the embedded
 sr   M sin(θ ) M cos(θ )  approach, considers the machine as a set of mutually coupled
 br r br r 
inductances and applies precise implementation of Dommel
The voltage and flux linkage equations in (1) and (2) can be
algorithm to extract the discretized Norton equivalent of the
transferred to stator frame of reference so that the inductances
machine model. This method is computationally more
no longer depend on the rotor position. With that, d- and q-
extensive than the interfaced-based approach as the time-
axis equivalent circuits can be achieved. Note that, due to the
varying inductance matrix of the machine needs to be inverted
asymmetry in stator windings, transforming the above
every time-step. Also, elements of the equivalent admittance
equations to rotor frame of reference does not result in a
matrix need to be sent to the network solution of the RTDS
position-independent inductance matrix [7]-[9].
simulator every time-step. Details of implementing this
In [8], q- and d-axis equivalent circuits are extracted with
method are discussed previously in [3], [4]. This approach
reference to main and auxiliary windings individually, thus the
provides a more numerically stable solution than the
q- and d-axis magnetizing inductances are not equal. Fig. 2
interfaced-based approach [3], [4]. This method also provides
shows the equivalent circuit of the machine using the approach
more flexibility in providing enhancements to the model such
used in [8]. It is usual to have SPIM data available in this
as inclusion of embedded breakers and arbitrary connection of
form.
1 's
the windings and nodes.
ω Ψdr 's
iqss = ias ras llas '
llar r a rar' iqr In this paper, the embedded phase-domain approach is used
for incorporating the single-phase induction machine model
into the network solution of RTDS: The trapezoidal rule of
Lmas d 's vqr' s integration is directly applied to discretize the phase-domain
vas = vqss Ψ
dt qr differential equation of the SPIM described in (1) and (2). As
SPIM inductances are changing with rotor position and
magnetic saturation, equivalent conductance matrix and
's
ω a ⋅ Ψqr history terms are calculated and passed to the network solution
s
ids = ibs rbs llbs '
llbr r rbr' idr's
of the simulator every each time-step.
The “CBuilder” utility of RSCAD software [11] is used for
s Lmbs d 's 's
vdr implementation of this model. This utility allows the user to
vbs = vds Ψ
dt dr implement a model in C language and creates a suitable
assembly code for the target processor.
Fig. 2. Equivalent circuit of an idealized two-phase induction machine Inclusion of the effects of magnetic saturation:
(rotor windings reflected to main and aux. windings). Inclusion of magnetic saturation into the transient
simulation of the SPIM model is one of the contributions of
III. IMPLEMENTATION OF THE SPIM MODEL INTO THE this paper. The rotor of a single-phase induction machine is
ENVIRONMENT OF THE REAL-TIME DIGITAL SIMULATOR
round similar to that of a 3-phase induction machine. The air-
The RTDS simulator is a combination of computer gap is also uniform. This means that similar to a 3-phase
hardware and software designed specifically for the solution induction machine, magnetic saturation can be included by
of power system electromagnetic transients in real time. Each adjusting the magnetizing inductance as a function of total
unit of RTDS is called a rack. Every rack consists of magnetizing flux. However, as a SPIM has unsymmetrical
processors dedicated to the solution of power system network, stator windings (main and auxiliary), magnetizing inductance
power system devices and control components [11]. Each seen from the q- and d-axis Lmq and Lmd are not equal.
simulation time-step is divided into computation intervals and
communication intervals. Similar to other electromagnetic Numerical problems may arise if these magnetizing
transient programs such as EMTP and EMTDC, Dommel inductances are to be adjusted independently. Furthermore, in
algorithm [12] is used for discretization of the system’s contrast to previous work [7]-[9], the implementation of this
differential equation. Subsequently, nodal solution is used for model is done in phase-domain, thus phase domain
calculation of node voltages. Differential equations of power inductances need to be evaluated each time-step as a function
of saturation.
 The key factor in considering saturation is to observe the Features and Capabilities of the Model:
rotor in its natural state as a round rotor, i.e. having equal With the above description, the single-phase induction
magnetizing inductances on both q- and d-axes. This can be machine model is implemented in the environment of RTDS
done by observing all rotor and stator windings from the port simulator. This section describes the capabilities of this model.
of one of the stator windings e.g. stator main winding (see the The basic icon for this component appears as shown in Fig. 4.
equivalent circuit in Fig. 3.). This approach [9], reflects all of The terminals for stator main and auxiliary windings and the
the windings to the main stator winding. Fig. 3 shows the neutral point are respectively shown by M, A and N. Each of
equivalent circuit of the machine using this approach. the main and auxiliary windings have embedded series
Usually, data for single-phase induction machines is breakers that can be controlled independently.
available in both formats shown in Fig. 2 and Fig. 3. A capacitor, a resistor or a combination of resistive and
Parameters can be transferred from one circuit to another capacitive elements can be inserted between nodes A and M,
using the value of Aux. / Main turns ratio ( a ) and can be resulting in a capacitor-start, capacitor-run, two-value
measured using standard no-load, locked-rotor and standstill capacitor motor or other arrangements. Twelve instances of
frequency response tests [10]. In the model developed in this this model can be simulated in real-time using one PB5
paper, the input data for the rotor is in the form of parameters processor [11] of the RTDS simulator.
in the equivalent circuit shown in Fig. 3. Therefore, only one
value for rotor resistance and leakage inductance is required.
If the available data is in the form of equivalent circuit shown
in Fig. 2, then q-axis rotor resistance and leakage inductance is
adequate for the model. D-axis rotor parameters are equal to q-
axis parameters multiplied by ( a 2 ). See [11] for more details.
's
ω Ψdr 's
iqss = ias ras llas llr' r rr' iqr

Fig. 4. The icon for RTDS single-phase induction machine model.

s Lms d 's vqr's
vas = vqs Ψ
dt qr
IV. VALIDATION OF THE MODEL AND TIME-DOMAIN
's
SIMULATION RESULTS
s
ids' s rbs' ' ω Ψqr rr' idr' s
ids = ibs llbs llr' r
This section provides validations for the SPIM model
a :1 presented in this paper. The focus of this section is to provide
s d 's validations for the capability of the real-time model in
vbs = vds 's
vds Lms Ψ vdr' s
dt dr accurately solving the differential equations of a single-phase
induction machine stated in (1) and (2). Note that, these
Fig. 3. Equivalent circuit of an idealized two-phase induction machine (all equations and the corresponding equivalent circuits are
windings reflected to the main windings). accepted forms for the SPIMs in the analysis of electric
In every time-step, d- and q-axes magnetizing currents are machinery [7]-[10].
calculated and from those the total magnetizing current The real-time simulation results are compared against the
analytical solution of the model to provide validations in
im = imd 2 + imq 2 is evaluated. Saturated magnetizing steady state performance of the model.
inductance Lms is evaluated from the saturation curve and state To validate the performance of the real-time model in
of the magnetizing inductance. Then, saturated values of transient conditions, a small off-line stand-alone
phase-domain inductances are evaluated from the saturated electromagnetic transient program including a stand-alone
magnetizing inductance. Subsequently, a new inductance SPIM model is developed. This tool allows lowering of the
matrix is used in the embedded phase domain algorithm of the simulation time-step to values much lower than 50.0 µs (e.g.
machine. 1.0 µs) in order to obtain more accurate results. Magnetic
Torque and Mechanical Swing Equations: saturation is also included in the off-line program. Note that,
The following set of equations is used for calculating the inclusion of magnetic saturation into the analysis of SPIMs
electric torques and solving mechanical swing equations. for electromagnetic transient programs is one of the
contributions of this paper. Comparing the real-time
p
Te = ⋅L ⋅ simulation results against the results of the off-line program
2 ms
  '  provides validations for the real-time model in transient
 − sin(θr ) − cos(θr ) iar 
(
ias ibs 
)    conditions.
 ' 
a ⋅ cos(θr ) −a ⋅ sin(θr ) ibr  Signals labeled “RTDS” in this section and all the signals
(3) in Fig. 10 are real-time response from the model. Parameters
Tm (pu ) − Te (pu ) = 2H ⋅ d
ω(pu ) + D ⋅ ω(pu ) for the SPIM under study are shown in the Appendix.
dt
 A. Comparison Against Analytical Solution B. Comparison Against Stand-Alone Models
Using the revolving field theory and decomposing the To validate the real-time model further, a stand-alone
rotating magnetic field into forward and backward fields, the single-phase induction machine model is developed. The
steady state equivalent circuit of a single-phase induction solution of this model is also in phase domain. In this
machine [1], [2] and [15] can be extracted as shown in Fig. 5. experiment, the start-up of a SPIM from zero speed is
Using this equivalent circuit, at standstill, the magnitude of the simulated. During the start-up a constant mechanical torque of
impedance seen from the main winding terminal is 1.0 pu is applied on the shaft.
analytically calculated and compared against the numerical The main and auxiliary windings are energized with
results from the real-time model. The results are shown in Fig.
6. 0.11 kV∡0 and 0.13 kV ∡ − 93 60 Hz voltage sources.
In another experiment, the magnitude of this impedance is During the start-up, once the machine speed reaches 0.8 pu ,
calculated while the main winding is supplied with a 60Hz the auxiliary winding breaker is opened which takes this
voltage source and the rotor speed is changed from -1.0 pu – winding out of circuit. This experiment is repeated with the
1.0 pu (slip = -2.0 – 0.0) . Analytical solution is compared real-time model and the results are compared in the following
against the simulation results in Fig. 7. As can be seen in both figures. Fig. 8a shows the stator main currents captured from
figures, simulation results and analytical solutions are the stand-alone offline simulation and the real-time model.
identical. Note that in the above analysis, the auxiliary Fig. 8b and Fig. 8c respectively show the start-up torque and
winding is open and saturation effects are ignored. speed signals from the above approaches. As can be seen,
there is a very good agreement between the results of the
iqss = ias ras j ωs ⋅ llas 1
2
j ωs ⋅ llr' stand-alone offline program and of the real-time model
presented in this paper.
1
j ωs ⋅ Lms r'
1 r
RTDS Off_Line Solution
2 0.03
2
s
j ωs ⋅ llr' Current (kA)
s 1 0.01
vas = vqs 2

1 1
rr' -0.01
j ωs ⋅ Lms 2
2 2−s
-0.03
Fig. 5. Steady-state equivalent circuit of a single-phase induction 1.5 2 Time (s) 2.5 3
machine.
RTDS Off-Line Solution
RTDS Analytical Solution 2
15
0
Torque (pu)
Impedance (Ohm)

-2
10
-4

5 -6

-8
0 1.5 2 Time (s) 2.5 3
0 20 40 60 80 100
RTDS Off-Line Solution
Frequency (Hz) 1
Fig. 6. Standstill frequency response of the SPIM from the port of main 0.8
Speed (pu)

winding.
0.6
RTDS Analytical Solution 0.4
0.2
35
Impedance (Ohm)

0
1.5 2 Time (s) 2.5 3
25
Fig. 8. Start-up of the SPIM. (a) Main winding current. (b) Electric
torque. (c) Speed
15
V. EXAMPLE OF STALLING AIR-CONDITIONING COMPRESSOR
5 UNITS AFTER A REMOTE FAULT
0 0.5 1 1.5 2
Slip (pu) This section presents an example of SPIMs of the type
Fig. 7. Impedance vs slip characteristics of the SPIM
typically used in air-conditioning units. The SPIMs stall after
a fault. A small circuit containing two single-phase induction
machines is simulated using the RTDS simulator. As shown in
Fig. 9, a permanent capacitor with the capacitance of 15.4 µF Vm1

Main Winding Voltage (V)

200
and resistance of 9.0 Ω is connected between the main and
auxiliary terminals of the SPIMs. In this example, the 100

auxiliary winding breaker is always closed. Induction machine 0

parameters are shown in the Appendix. The mechanical load is
-100
based on the typical torque-speed characteristics of a
centrifugal pump [16]. Both machines are supplied from a -200

0.11 kV, 60 Hz voltage source, representing one phase of a Imain1

Main Winding Current (A)

10
distribution system. In steady-state, a solid remote fault with
the duration of 3 s is applied to the source. Fig. 10 shows the 5

simulation results during this fault and after it is cleared. The 0

voltage drop (Fig. 10a) during the fault reduces the SPIM
-5
electromagnetic torque significantly which in turn decreases
rotor speed. Since the fault duration is long enough, speed -10

signals for both SPIMs eventually reach zero causing stalling Pmac1 Ploss1

Input & Loss Power (kW)

0.3
of the SPIMs and compressors (Fig. 10d). Once the fault is
cleared, the voltages at the terminals of the SPIMs tend to 0.2
recover. However, the SPIMs are still stalled and draw a large
amount of current (Fig. 10b) which results in a large voltage 0.1

drop across the series impedance of the source, causing

0
terminal voltages not to recover fully. At this point, the
electromagnetic torque produced by each SPIM is not large SPDOUT1 SPDOUT2
0.2
enough to surpass the minimum torque [16] required to rotate 0
the stalled pump. This causes the SPIMs and compressors to -0.2

Speed (pu)
stay stalled even after the fault is cleared (Fig. 10d). The -0.4

SPIMs stay stalled until one of them is disconnected from the -0.6

circuit. Another interesting observation in this experiment is -0.8

-1
the SPIM’s loss during the fault. As can be seen in Fig. 10c, 0 0.67 1.33 2 2.67 3.33 4
once a SPIM is stalled, most of the power absorbed by a Time (s)
machine is turned to resistive losses. This is consistent with
Fig. 10. SPIM signals during a remote fault.
the expected behavior of physical SPIMs in air-conditioning
systems. This loss causes extra heat in the SPIMs and
eventually thermal protection may isolate the machines from
the circuit. Similar results are achieved when the fault is VI. CONCLUSIONS
applied to the line. A detailed transient single-phase induction machine model
was developed in this paper which is suitable for
electromagnetic transient programs and real-time simulation.
The model has the following important properties:
• The embedded phase domain approach [3], [4] is
utilized for incorporation of this model into the
network solution of the RTDS simulator.
• Magnetic saturation is included into the model by
reflecting all the windings to the stator main
winding and adjusting the magnetizing inductance
according to the total magnetizing flux.
The model presented in this paper was validated using
analytical solutions and stand-alone off-line simulations.
Examples for applications of this model in simulating
scenarios related to the stalling of air-conditioning pumps in
distribution systems were presented. This detailed model can
be used for a variety of studies in distribution networks. It also
assists with the growing applications of electromagnetic
transient programs and real-time simulators in studies related
to distribution systems.
Fig. 9. Simulated circuit for stalling of SPIMs.
 VII. APPENDIX [9] S. D. Sudhoff, ”Multiple Reference Frame Analysis of an
Unsymmetrical Induction Machine, ” IEEE Trans. Energy Conversion,
TABLE I vol. 8, No.3, pp. 425-432, Sep. 1993.
PARAMETERS OF THE SPIM [8] [10] P. Vas, Parameter, Parameter Estimation, Condition Monitoring,
Per-Unit Base Value Diagnosis of Electrical Machines, Oxford Science. Publications, 1993.
Line-neutral rated voltage 0.11 kV [11] RTDS User’s Manual, RTDS Technologies Inc., Winnipeg, Canada,
2014.
Rated MVA 0.186 kVA [12] H. W. Dommel, “Digital computer solution of electromagnetic transients
Rated Frequency 60 Hz in single and multiphase networks" ,IEEE Trans. Power Apparatus and
Impedance base value 64.9 Ω Systems, vol.PAS-88, No.4, pp. 388-399, Apr. 1969.
[13] A.M. Gole, R.W. Menzies, D.A. Woodford and H. Turanli, “Improved
Inductance base value 172.16 mH interfacing of electrical machine models in electromagnetic transient
Aux. / main turn ratio 1.18 programs," IEEE Trans. Power Apparatus and Systems, vol. PAS-103,
Inertia Constant (H) 1.39 s no. 9, pp. 2446-2451, Sept. 1984.
[14] T.L. Maguire, ”An Efficient Saturation Algorithm for Real Time
Per- Synchronous Machine Models using Flux Linkages as State Variables,”
Parameter Value Physical
Unit Electrimacs 2002, Montreal, Canada, June 2002. IEEE Trans. Power
Main winding resistance 0.031 2.02 Ω Apparatus and Systems, vol. PAS-103, no. 9, pp. 2446-2451, Sept. 1984.
Main winding leakage reactance 0.043 2.79 Ω [15] G.R. Slemon, Magnetoelectric devices, New York, John Wiley &Sons,
Inc., 1966, pp. 39-64.
Unsaturated magnetizing reactance 1.029 66.8 Ω
[16] Sulzer Pumps, Sulzer Centrifugal Pump Handbook, Elsevier Advanced
Aux. winding resistance 0.11 7.14 Ω Technology, 1998, p. 62.
Aux. winding leakage reactance 0.05 3.22 Ω
Rotor winding resistance 0.063 4.12 Ω Ali B. Dehkordi (S’97, M’10) received the B.Sc.
and M.Sc. degrees in electrical engineering from
Rotor winding leakage reactance 0.033 2.12 Ω Sharif University of Technology, Tehran, Iran and
Saturation Data Ph.D. degree from the University of Manitoba in
Magnetizing Current (Norm) Voltage (pu) 1999, 2002 and 2010 respectively. He worked as a
research engineer for Sharif University of
0.0 0.0 Technology in power quality projects in 2002. His
0.5 0.5 research interests are on the modeling of power
0.8 0.79 system components and in particular electric
machines for Electromagnetic Transient Programs
1.0 0.947 and Real Time Digital Simulation, the area that he
1.2 1.076 has been working on for over 10 years. Dr. Dehkordi is a recipient of Dennis
Woodford Prize, for “the most outstanding graduate thesis dealing with power
1.5 1.2
system modeling and simulation”. He is currently employed by RTDS
1.8 1.3 Technologies Inc., Winnipeg, Canada.
2.2 1.39
3.2 1.58
4.2 1.74

VIII. ACKNOWLEDGMENT
The author sincerely appreciates the high valued technical
support from the colleagues at RTDS Technologies Inc.

IX. REFERENCES
[1] Ion Boldea, Syed A. Nasar, The Induction Machine Handbook, CRC
Press LLC, 2002, Chapter 23.
[2] Hamid A. Toliyat, Gerald B. Kliman, Handbook of Electric Motors,
CRC Press LLC, 2004, Chapter 2.
[3] A. B. Dehkordi, P. Neti, A.M Gole, and T.L. Maguire, “Development
and Validation of a Comprehensive Synchronous Machine Model for a
Real-Time Environment”, IEEE Trans. On Energy Conversion, vol. 25,
no.1, pp. 34-48, Mar 2010.
[4] A.B. Dehkordi, A.M Gole, and T.L. Maguire, “Permanent magnet
synchronous machine model for real-time simulation”, International
Power System Transient Conference (IPST 2005), Montreal, June, 2005.
[5] B. R. Williams, W. R. Schmus, and D. C. Dawson, “Transmission
voltage recovery delayed by stalled air conditioner compressors,” Power
Systems, IEEE Transactions on, vol. 7, no. 3, pp. 1173–1181, 1992.
[6] K. Prabakar, D. T. Rizy, F. Li, “A study on the effect of distribution
circuit loading on air conditioner motor stall using a real time
simulator,” T&D Conference and Exposition, 2014 IEEE PES, April
2014.
[7] P.C. Krause, O.Wasynczuk, S.D. Sudhoff, Analysis of electric
machinery, New York, IEEE Press, 1995.
[8] P. C. Krause, ”Simulation of Unsymmetrical 2-Phase Induction
Machines, ” IEEE Trans. Power Apparatus and Systems , vol. PAS-84,
No.11, pp. 1025-1037, Nov. 1965.