1332 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 35, NO.

6, NOVEMBER/DECEMBER 1999

Complex Vector Model of the Squirrel-Cage

Induction Machine Including Instantaneous
Rotor Bar Currents
Alfredo R. Muñoz and Thomas A. Lipo, Fellow, IEEE

Abstract—In this paper, a new detailed mathematical deriva- inductance matrices. The resultant set of differential equations
tion of the squirrel-cage induction machine d–q model is intro- is then solved for each rotor position.
duced. The model is based on coupled magnetic circuit theory and In general, for a machine having rotor bars, this approach
complex space-vector notation and takes into account the actual
nonsinusoidal rotor bar distribution. It is shown for the first time leads to a model having nonlinear simultaneous dif-
that, given the structural symmetry of the induction machine, ferential equations (plus the mechanical equation), requiring
both stator and rotor circuits can be modeled by the simple set huge computational power [6], [7]. Although the number of
of only four coupled differential equations, i.e., the d–q model. equations can be somewhat reduced when there is an integer
More importantly, the number of equations does not depend on number of rotor bars per pole, the task of solving such a system
the number of rotor bars, and the model is valid even if the
number of bars per pole is not an integer number. This enormous of differential equations remains a formidable one.
simplification is achieved without loss of generality nor loss of any In this paper, it is shown that by using a particular space-
information contained in the full set of equations, and it is valid vector transformation, the machine can be fully modeled
for any operating condition. The actual n rotor bars and end-ring using only four coupled differential equations. This enormous
currents are fully included in the model, and they are obtained simplification is achieved without loss of generality nor loss
directly by using a simple vector transformation. In addition, the
three-phase rotor equivalent parameters are obtained. Second- of any information contained in the full set of equations, and
order effects, such as skin effect in the rotor bars, can be taken it is valid for any operating condition. More importantly, the
into account by simply modifying the bar and end-ring resistance number of equations does not depend on the number of rotor
values. An equivalent circuit based on the model is also derived. bars, and it is valid even if the number of bars per pole is not
Index Terms—Complex vector, d–q model, induction machine, an integer number. The model predicts the three-phase rotor
modeling, rotor bar, rotor current, simulation, squirrel cage. equivalent parameters in terms of the actual rotor bar resistance
and inductance values, as well as giving the actual rotor bar
and end-ring currents. Since the model uses the actual rotor
I. INTRODUCTION
bar and end-ring parameters, second-order effects, such as skin

T HE well-known – model of induction machines is

based on the assumption that both stator and rotor wind-
ings are sinusoidally distributed in space. Although the rotor
effect, can be easily incorporated.

cage is clearly nonsinusoidally distributed, it is claimed that it

can be replaced by an equivalent distributed winding [1]–[3]. II. STATOR MODEL
A formal derivation of such an equivalence, however, is rarely The coupled magnetic circuit theory and complex space-
found in the literature, and in those few cases where a more vector representation will be used throughout the derivation.
formal approach has been adopted, the derivation becomes This technique is chosen because of the great deal of simpli-
very complex, requiring simplifying assumptions [4], [5]. fication that can be achieved, as well as its generality. The
In the past, several authors have used the coupled magnetic following general assumptions are made:
circuit theory to model a squirrel-cage induction machine
• negligible saturation;
[6]–[12]. The method consists of describing the machine as
• uniform air gap;
a set of multiple coupled circuits defined by self- and mutual-
• stator windings sinusoidally distributed;
Paper IPCSD 98–78, presented at the 1998 Industry Applications Society • negligible interbar current.
Annual Meeting, St. Louis, MO, October 12–16, and approved for publica- Although a sinusoidally distributed stator winding is as-
tion in the IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS by the Electric
Machines Committee of the IEEE Industry Applications Society. Manuscript sumed, other winding distributions could also be analyzed by
released for publication November 30, 1998. simply using superposition. This approach is justified by the
A. R. Muñoz is with the Department of Electrical and Computer Engineer- well-known fact that different space harmonic components do
ing, University of Wisconsin, Madison, WI 53706-1691, USA, on leave from
the Electrical Engineering Department, Technical University Santa Marı́a, not interact [12].
Valparaı́so, Chile (e-mail: [email protected]). The total stator flux vector can be separated into a part
T. A. Lipo is with the Department of Electrical and Computer Engi- due to the stator currents and one due to the mutual coupling
neering, University of Wisconsin, Madison, WI 53706-1691, USA (e-mail:
[email protected]). with the rotor circuit. For clarity of explanation, each term
Publisher Item Identifier S 0093-9994(99)07496-4. will be developed separately.
0093–9994/99$10.00  1999 IEEE
MUÑOZ AND LIPO: COMPLEX VECTOR MODEL OF THE SQUIRREL-CAGE INDUCTION MACHINE 1333

portions of the end rings between them. Each loop is magnet-

ically coupled to all the other loops (circuits) and to all three
stator phases.
The stator flux linkage due to the rotor currents is

..
. (6)

where represents the mutual inductance between the stator

phase and the rotor loop . Neglecting the MMF drop in the
iron, these inductances are conveniently calculated by means
of winding functions [13], [14]. According to this theory the
mutual inductance between two arbitrary machine windings
“ ” and “ ” is given by

(7)
Fig. 1. Elementary rotor loops and current definitions.

where is the angular rotor position with respect to some

A. Stator Flux due to Stator Currents arbitrary reference, is a particular point along the air gap,
is the inverse air-gap function, and is the
For sinusoidally distributed windings, the stator flux due to
winding function defined as the spatial distribution of MMF
the stator currents is known to be [1]
due to a unit current flowing in the winding. If the air gap is
assumed to be uniform and small compared to the rotor radius,
the inverse air-gap function becomes simply a constant ,
(1) and it can be taken out of the integral. In this case, to find
the mutual inductance, we only need to define the winding
functions.
Consider the stator windings and rotor bar disposition shown
where and are the leakage and magnetizing in-
in Fig. 2. Taking as a reference the magnetic axis of phase ,
ductances of the stator windings. Applying the three-phase the normalized stator winding function for this phase is [14]
space-vector definition to (1) yields

(2) (8)

where is the complex number and is the The normalized winding functions of phases and are simply
complex space vector, thus displaced by 120 with respect to .
The normalized winding function for the th rotor loop,
given by the MMF distribution produced by 1 A of current
(3)
flowing through the th loop, is shown in Fig. 2. For all the
other loops, the normalized winding functions are similar,
where changing only the relative phase angle as defined by the
position of each loop with respect to the reference point.
(4)
Mathematically, this function is defined by
is the space-vector representation of the stator current.
The magnetizing inductance for a winding having
(9)
turns per phase is given by [1]

(5) where is the angle between two adjacent rotor bars and
and define the position of the bars forming the loop.
where is the stack length, the mean air-gap radius and The mutual inductance between phase and the th rotor
the air-gap length. loop is

B. Stator Flux Due to Rotor Currents

The rotor cage is normally modeled as identical and
equally spaced loops [4], [5], [8]. As shown in Fig. 1, each (10)
loop is formed by two adjacent rotor bars and the connecting
1334 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 35, NO. 6, NOVEMBER/DECEMBER 1999

where is the complex number . The mutual stator–rotor

flux linkage complex vector is obtained by adding all three
rows of (15) and multiplying by 2/3, thus

.. (16)
.

Since , the first term on the right-hand side of

(16) is always zero, therefore, simply becomes

(17)

where

.. (18)
(a) .

defines the space-vector representation of the rotor currents.

The total stator flux, given by the sum of (3) and (17), is

(19)

where . Note that, in this derivation, no

assumption has been made regarding the waveform of the rotor
(b)
currents, and (19) is valid for any type of excitation, as well
as during transient operation.
Fig. 2. (a) Stator windings distribution. (b) Rotor loop winding function.
In space-vector notation, the stator voltage is known to
be [1]
Since each rotor loop is symmetrically located along the
rotor periphery, can be expressed in terms of an arbitrary (20)
rotor angle and the angle between adjacent rotor bars as where is the stator resistance. Taking the time derivative of
(11) (19) and substituting into (20) yields

By using trigonometric identities, (10) can be written as (21)

(12) This expression has the same structure as the well-known
where – model available in the literature, the main difference being
the phase angle introduced by the complex exponential term,
(13) which is due to the arbitrary choice of reference for the mag-
netic axis of the first rotor loop. Also, since a nonsinusoidal
and rotor bar distribution is considered, the number of rotor bars
appears explicitly.
(14)

The mutual inductance of the th rotor loop with respect to III. ROTOR MODEL
phases and are found by phase shifting by 120 . Given the structural symmetry of the rotor, it is convenient
Using the Euler relation and substituting the mutual induc- to model the cage as identical magnetically coupled circuits.
tance expressions into (6) yields One particular advantage of this approach is that it is applica-
ble to rotors with a nonintegral number of bars per pole. For
simplicity, each loop is defined by two adjacent rotor bars and
the connecting portions of the end rings between them.
For the purpose of analysis, each rotor bar and segment of
.. end ring is substituted by an equivalent circuit representing the
. (15) resistive and inductive nature of the cage [4], [5], [15]. It is
also convenient to carry out the analysis using mesh currents as
MUÑOZ AND LIPO: COMPLEX VECTOR MODEL OF THE SQUIRREL-CAGE INDUCTION MACHINE 1335

where represents the zero voltage vector due to the squirrel

cage and is the equivalent rotor resistance in the subspace
defined by the space-vector transformation.
The total rotor flux can be
divided into two components, one due to the stator currents,
and one due to the rotor currents . Again, for clarity
of explanation, each component will be developed separately.

A. Rotor Flux Due to Stator Currents

The mutual coupling between rotor and stator is given by

.. .. .. .. (25)
. . . .

Because of energy considerations, for

and for ; and they correspond to those defined in
Section II-B. Using complex notation and multiplying the rows
of (25) by yields (26), shown at the bottom of
the page. The complex vector is obtained by adding all
the rows of (26) and multiplying by , thus

Fig. 3. Rotor cage equivalent circuit.

the independent variables. Such an equivalent circuit is shown (27)

in Fig. 3.
In general, there are independent meshes defined by
the rotor loops plus one formed by any one of the end Since and using the definition of
rings. However, in the absence of an axial flux component, the stator current space vector , (27) reduces to
the circumferential current in the end ring is identically
zero, hence, it will not be considered here. (28)
The loop equation for the th rotor circuit is
B. Rotor Flux Due to Rotor Currents
(22)
The total flux linked by the th rotor circuit and due only
where represents the bar resistance, is the end-ring to the rotor currents is given by
segment resistance, is the total flux linked by the th
loop, and is the loop current. Since each loop is assumed
to be identical, (22) is valid for every loop, therefore, the rotor (29)
voltage equation in matrix form is where represents the mutual magnetic coupling between
rotor loops and and is the self-inductance of the th
loop.
.. .. .. .. .. .. .. (23) Because of the structural symmetry of the rotor, (29) is valid
. . . . . . .
for every loop, hence, can be written in matrix form as

with .
The rotor equation in space-vector form is obtained by .. .. .. .. .. ..
. . . . . .
applying the transformation given in (18)–(23) yielding

(24) (30)

.. .. .. .. .. .. .. (26)
. . . . . . .
1336 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 35, NO. 6, NOVEMBER/DECEMBER 1999

with . The self- and mutual inductances

defined in (30) are obtained by means of winding functions.
The self-inductance of the th loop is found to be [6]

(31)

and the mutual inductance between the th and the th rotor

loops is

(32)

As expected, both and are constants that only depend

on the rotor dimensions and not on their relative position.
Substituting (31) and (32) into (30) and multiplying each row
by yields (33), shown at the bottom of the page,
where we have used the identity . (a)
Adding all the rows in (33) and reducing terms, we obtain
the rotor flux complex vector due to the rotor currents

(34)

The proportionality constant corresponds to the equivalent

rotor inductance. Note that its value is expressed only in terms (b)
of rotor dimensions and bar and end-ring inductance values. Fig. 4. (a) Common reference frame definition. (b) Complex vector equiv-
The total rotor flux vector, defined as the sum of (28) and alent circuit.
(34), is
Finally, substituting (35) into (24) yields the rotor equation
(35) in the subspace defined by the rotor space-vector transforma-
tion
C. Rotor Equivalent Resistance
(37)
The equivalent rotor resistance is obtained from the first
term on the right-hand side of (23). The procedure is identical
to the one used to obtain the rotor flux, therefore, it will not Equations (21) and (37) correspond to the complex vector
be repeated here. It suffices to say that after multiplying each model of the squirrel-cage induction machine and fully define
row of the first term in (23) by , adding and the electrical behavior for both steady-state and transient
simplifying it yields operation.

(36) IV. COMPLEX VECTOR EQUIVALENT CIRCUIT

It is known that the sinusoidal coupling between the stator
Note that equivalent rotor resistance is expressed in terms and rotor circuits can be eliminated by referring all the
of the actual bar and end ring resistance values, making equations to a common reference frame [16]. Fig. 4 shows
it straightforward to include changes in their values due to an arbitrary – reference frame rotating at angular speed .
temperature or skin effect. The required – variables in this common reference frame

.. .. .. .. .. .. (33)
. . . . . .
MUÑOZ AND LIPO: COMPLEX VECTOR MODEL OF THE SQUIRREL-CAGE INDUCTION MACHINE 1337

are defined by the vector transformations where and

(38) (46)
.. .. ..
. . .
Applying this transformation to (21) and (37) yields
where the superscript denotes complex conjugate. Since the
(39) input voltage vector in (45) is identically zero, the only source
of excitation comes from the mutual rotor–stator coupling
terms. Given the structure of , regardless of the instanta-
neous value of the stator currents, only the second and th
(40) rows of (45) will yield a nonzero rotor current [17]. The
resultant differential equations are of the form
where
(47)
(41)
for and . Their solutions are independent of the
stator currents and correspond to a trivial case (e.g., identically
(42) zero for all time), therefore, they can be eliminated from the
solution set. On the other hand, the equations corresponding
The corresponding equivalent circuit representing the ma- to the second and last rows of (45) are
chine in an arbitrary rotating reference frame is shown in
Fig. 4.
The analysis carried out so far has implicitly assumed (48)
a machine having only one pair of poles, the extension
to multipole structures is quite straightforward, and it only
requires one to substitute mechanical angles by their electrical and their solutions depend on the instantaneous stator currents
equivalents. and . From here, it follows that the inverse current
It is important to point out that, in contrast to the usual relation can always be written simply as
full set of differential equations required to solve for rotor bar
currents, the use of the space-vector transformation defined
in (18) gives a much simpler description of the machine, (49)
regardless of the number of rotor bars, thus providing a faster .. ..
. .
analysis tool.

V. INVERSE CURRENT TRANSFORMATION Thus, due to lack of excitation, only the second and last
column of the inverse transformation matrix are of interest.
In general, all rotor currents need to be mapped into a full Furthermore, from (43), it is clear that corresponds to
-dimensional vector space. This new -dimensional space is the space-vector definition given in (18) and is the com-
defined by the transformation [18] plex conjugate. Hence, taking the inverse transformation and
substituting yields

.. .. .. .. .. .. (43)
. . . . . .
(50)
.. .. ..
. . .
where represents the th rotor current in the new space.
Upon applying (43) to the voltage equations of a symmet-
rical squirrel-cage machine results in [17], [18] Equation (50) shows that, for a symmetrical squirrel-cage
machine, only the complex conjugate subspaces are externally
(44) excited through the stator coupling, while the rest can never
be excited. Therefore the original -dimensional space can be
where , fully represented by the bidimensional subspace spanned by
and is the transformation matrix (43). (18). Finally, the inverse transformation is simply
If, in addition, the stator currents are also mapped into a
new vector space defined by a transformation , similar to
but substituting by and by 3, results in
(51)
(45)
1338 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 35, NO. 6, NOVEMBER/DECEMBER 1999

where represents the real part and and are the real
and imaginary components of the complex vector . Note that
this result is completely general and it does not assume any
symmetry in the rotor current distribution.

VI. ELECTROMAGNETIC TORQUE

Neglecting saturation, the electromagnetic torque can be
expressed as the partial variation of the coenergy with respect
to rotor position [1], [6]

(52)

where and corresponds to

the rotor–stator mutual inductance matrix. After some matrix
manipulation, the mutual inductance can be written as

(53)

Note that only the exponential terms depend on the rotor

position, hence, the partial differentiation is quite easily ob-
tained. After differentiating and premultiplying by and
postmultiplying by yields

(54)

where represents the imaginary part.

It is important to point out that (54) is similar to the usual
expression derived for wound-rotor machines. The term
arises from the arbitrary constant in the definition of the rotor
current space vector and the inductance term includes the
equivalent of the turns ratio found in the usual model. Fig. 5. Complex vector and full matrix model results (superimposed). Rotor
bar current, stator current, rotor speed, and electromagnetic torque during free
acceleration.
VII. SIMULATION RESULTS
To validate the proposed complex vector model, a set of
simulations was prepared. The main objective of the simula- inversions combined to the rapid change in mutual inductance.
tions is to compare the results of the proposed model to those Decreasing the integration step helps minimize this error,
of the full matrix model of the induction machine. The full however, this increases the computation time even more. The
matrix model has been well documented in the literature, and complex vector model, on the other hand, is free of such
its correctness has been proven over the years. The most recent problems.
examples can be found in [8], [15]. For simulation purposes, a
typical 5-hp four-pole machine was used. The results from the
VIII. THREE-PHASE EQUIVALENT PARAMETERS
full matrix model and the complex vector model are shown
superimposed in Figs. 5 and 6 for sinusoidal excitation. Fig. 7 Another important advantage of the complex vector model
shows the results for nonsinusoidal excitation. The agreement is its ability to predict the parameters of an equivalent wound-
in all cases is perfect. The computation times required in each rotor machine. The equivalent parameters are most easily
case are shown in Table I. The reduction in computational obtained by introducing a slight modification to the rotor
time achieved by the complex vector model is overwhelming. current space-vector definition
Note that both models fully predict the rotor bar and end-
ring currents. However, at high speeds, the full matrix model (55)
is prone to larger numerical errors due to the multiple matrix
MUÑOZ AND LIPO: COMPLEX VECTOR MODEL OF THE SQUIRREL-CAGE INDUCTION MACHINE 1339

TABLE I
REAL TIME REQUIRED TO RUN SIMULATIONS

TABLE II
LEAKAGE INDUCTANCE CALCULATIONS

Fig. 6. Complex vector and full matrix model results (superimposed). Rotor
currents in two adjacent rotor bars and in the segment of end-ring between
them during the initial transient.

Fig. 8. Change in machine dynamics with rotor leakage inductance (24 rotor
bars). 1: Full matrix model. 2: Space-vector model. 3: d–q model using
equivalent parameters derived from space-vector model. 4: d–q model using
equivalent parameters derived from traditional harmonic analysis [2].

and the equivalent rotor leakage inductance

(57)

Alternate forms to obtain the equivalent rotor parameters

of a squirrel-cage machine are presented in [2] and [4].
The technique used is based on harmonic analysis, and to
obtain a closed-form solution for the rotor leakage flux,
some approximations need to be made. Specifically, some
Fig. 7. Complex vector and full matrix model results (superimposed). Stator
and rotor bar currents for nonsinusoidal excitation (six step). of the spatial harmonic components produced by the rotor
fundamental time harmonic component are neglected. This
Substituting into (21) and (37) and comparing to the yields an approximate value of equivalent leakage inductance
equations of a sinusoidally wound-rotor induction machine given by
leads to the equivalent stator referred rotor resistance

(56)
(58)
1340 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 35, NO. 6, NOVEMBER/DECEMBER 1999

On the other hand, the complex vector model is based on [6] H. A. Toliyat, “Analysis of concentrated winding induction and reluc-
winding functions which take into account all the harmonic tance machines for adjustable speed drive applications,” Ph.D. disserta-
tion, Dep. Elect. Comput. Eng., Univ. Wisconsin, Madison, 1991.
components at once. Also, the novel expression for the leakage [7] H. R. Fudeh and C. M. Ong, “Modeling and analysis of induction
inductance derived here has the virtue of simplicity without machines containing space harmonics,” IEEE Trans. Power App. Syst.,
using the concepts of differential and zig-zag leakages. vol. PAS-102, pts. 1–3, pp. 2608–2628, Aug. 1983.
[8] X. Luo, Y. Liao, H. A. Toliyat, A. El-Antably, and T. A. Lipo,
The difference between the exact value of (57) and the “Multiple couple circuit modeling of induction machines,” IEEE Trans.
approximate one (58) is the coefficient of the last term with the Ind. Applicat., vol. 31, pp. 311–317, Mar./Apr. 1995.
[9] A. K. Wallace, R. Spée, and H. K. Lauw, “Dynamic modeling of
exact expression being twice as large. The relative importance brushless doubly-fed machines,” in Conf. Rec. IEEE-IAS Annu. Meeting,
of the error depends on the expression between the square San Diego CA, 1989, pp. 329–334.
brackets. When the number of rotor bars is large, and [10] J. Stepina, “Matrix calculation of inductances for the general theory of
electrical machines,” Elect. Mach. and Power Syst., vol. 11, pp. 53–63,
the error is small. Conversely, for rotors with fewer bars, the 1986.
error may be significant. As an example, the computations for [11] S. Williamson and R. C. Healey, “Space vector representation of
rotors having 24 and 48 bars are presented in Table II. advanced motor models for vector controlled induction motors,” Proc.
Inst. Elect. Eng.—Elect. Power Applicat., vol. 143, no. 1, pp. 69–77,
The effect of using the approximate value of leakage induc- 1996.
tance on the torque and speed curves for the 5-hp machine used [12] S. A. Nasar, “Electromechanical energy conversion in nm-winding
in the simulation is shown in Fig. 8. In this figure, the dotted double cylindrical structures in presence of space harmonics,” IEEE
Trans. Power App. Syst., vol. PAS-87, pp. 1099–1106, Apr. 1968.
line corresponds to the results of the – model with com- [13] T. A. Lipo, “Theory and control of synchronous machines,” ECE 511
puted using (58), while the solid line corresponds to the super- Class Notes, Elect. Comput. Eng. Dep., Univ. Wisconsin, Madison,
1991.
imposed solutions of the full matrix model, the complex vector [14] N. L. Schmitz and D. W. Novotny, Introductory Electromechanics.
model, and the – model with computed using (57). New York: Ronald, 1965.
[15] H. A. Toliyat and T. A. Lipo, “Transient analysis of cage induction
machines under stator, rotor bar, and end ring faults,” IEEE Trans.
IX. CONCLUSION Energy Conv., vol. 10, pp. 241–247, June 1995.
[16] P. Krause and C. Thomas, “Simulation of symmetrical induction ma-
A detailed mathematical model of a squirrel-cage induction chinery,” IEEE Trans. Power App. Syst., vol. PAS-84, pp. 1038–1053,
machine has been developed. It has been shown that the use Nov. 1965.
of complex space vectors leads to a simple – model without [17] A. Muñoz-Garcı́a, “Analysis and control of a dual stator winding squirrel
cage induction machine for high performance drives,” preliminary Ph.D.
giving up any information regarding the actual rotor bar and dissertation proposal, Univ. Wisconsin, Madison, 1997.
end-ring currents. Note that this is true during both transient [18] D. C. White and H. H. Woodson, Electromechanical Energy Conversion.
New York: Wiley, 1959.
and steady-state behavior, and it does not assume sinusoidal
currents.
The number of equations needed to fully represent the
machine is independent of the number of rotor bars considered
and the resulting equations have a similar structure to the Alfredo R. Muñoz was born in Valparaı́so, Chile.
one found in the usual – model. Because of the minimum He received the B.S. degree in electrical engineering
in 1981 from the Technical University Santa Marı́a,
number of equations used in the final model, the computation Valparaı́so, Chile, and the M.S. degree in electrical
time required to simulate the machine is drastically reduced. engineering in 1995 from the University of Wiscon-
The three-phase equivalent rotor parameters in terms of actual sin, Madison, where he is currently working toward
the Ph.D. degree.
rotor bar and end-ring resistance and inductance values are From 1981 to 1986, he was with Schlumberger
also obtained. Overseas. In 1987, he became a full-time Lecturer
and, in 1989, an Assistant Professor in the Electri-
REFERENCES cal Engineering Department, Technical University
Santa Marı́a. Since 1993, he has been on leave as a Research Assistant at the
[1] D. W. Novotny and T. A. Lipo, Vector Control and Dynamics of AC University of Wisconsin, Madison. He has been involved in several research
Drives. Oxford, U.K.: Clarendon, 1996. projects in the areas of variable-frequency drives and power electronics. He
[2] A. S. Langsdorf, Theory of Alternating Current Machinery, 2nd ed. was a Fulbright Fellow from 1993 to 1995. His main research interests include
New York: McGraw-Hill, 1955. electric machines, ac drives, and power electronics.
[3] P. L. Alger, Induction Machines, 2nd ed. New York: Gordon and
Breach, 1970.
[4] T. A. Lipo, Introduction to AC Machine Design. Madison, WI: Univ.
of Wisconsin Press, 1996.
[5] A. K. Wallace and A. Wright, “Novel simulation of cage windings based
on mesh circuit model,” IEEE Trans. Power App. Syst., vol. PAS-93, pp. Thomas A. Lipo (M’64–SM’71–F’87), for a photograph and biography, see
377–382, Jan./Feb. 1974. p. 486 of the March/April 1999 issue of this TRANSACTIONS.