IEEE TRANSACTIONS ON MAGNETICS, VOL. 46, NO.

7, JULY 2010 2611

Analytical Computation of the Magnetic Field Distribution

in a Magnetic Gear
Thierry Lubin, Smail Mezani, and Abderrezak Rezzoug
Groupe de Recherche en Electrotechnique et Electronique de Nancy, University Henri Poincaré, 54506 Nancy, France

We propose an analytical computation of the magnetic field distribution in a magnetic gear. The analytical method is based on the
resolution of Laplace’s and Poisson’s equations (by the separation of variables technique) for each subdomain, i.e., magnets, air gap, and
slots. The global solution is obtained using boundary and continuity conditions. Our analytical model can be used as a tool for design
optimization of a magnetic gear. Here, we compare magnetic field distributions and electromagnetic torque computed by the analytical
method with those obtained from finite-element analyses.
Index Terms—Analytical solution, electromagnetic torque, magnetic field, magnetic gear.

I. INTRODUCTION

AGNETIC gears offer substantial advantages compared

M to mechanical gears such as reduced maintenance, im-
proved reliability, minimum acoustic noise, and inherent over-
load protection. Physical and hermetic isolation between input
and output shafts are also specific to such contactless torque
transmission devices [1]–[5]. It has been shown that a trans-
mitted torque density of 100 can be obtained which
is comparable with that of mechanical gears. Recently, mag-
netic gear has been integrated into PM motors to obtain the so
called “pseudo-direct” drives that result in high-torque density
electrical machines [6]. The later actuators are suitable for high
torque low speed application like electric vehicles [7] or wind
power generation [8].
An accurate knowledge of the air-gap magnetic field distribu-
tion is necessary for predicting the performance of the magnetic
gear. The air-gap magnetic field can be evaluated by analytical
or semi-analytical methods or by numerical techniques like fi-
nite elements. Finite elements give accurate results considering
Q = 5,  = 0, p = 2, and
Fig. 1. Geometry of the studied magnetic gear (
geometric details and nonlinearity of magnetic materials. How- p =3 ).
ever, this method is computer time consuming and poorly flex-
ible for the first step of design stage of a magnetic gear. Ana-
lytical methods are useful tools for first evaluation of magnetic Laplace and Poisson’s equations are solved in each subdomain
gear performances and for design optimization since continuous (air-gap, magnet, and slots) and the solution is obtained using
derivatives issued from the analytical solution are of great im- boundary and interface conditions.
portance in most optimization methods. The paper is organized as follows. The problem description
As shown in Fig. 1, a magnetic gear contains a ring of fer- and the assumptions of the model are presented in Section II.
romagnetic pole-pieces which are difficult to handle (open slots Section III describes the analytical method for magnetic field
on two sides) in the analytical prediction of the air-gap magnetic calculation in the air gap, permanent magnets and in the slot
field. Analytical approaches for air-gap magnetic field computa- subdomains. The analytical results are then verified with the
tion in slotted machines can be found in the literature [9]–[14]. finite-element method in Section IV and Section V.
However, the publications focus essentially on the determina-
tion of cogging torque in permanent magnet motors. II. PROBLEM DESCRIPTION AND ASSUMPTIONS
The aim of this paper is to propose an exact analytical solu-
tion of the magnetic field distribution in a magnetic gear. The Schematic representation of the studied magnetic gear is
shown in Fig. 1. It consists of a pole-pairs inner rotor, a
pole-pairs outer rotor and ferromagnetic pole-pieces.
Manuscript received October 21, 2009; revised February 12, 2010; accepted Principle of operation of the magnetic gear is the modulation
February 19, 2010. First published March 08, 2010; current version published of magnetic field created by the pole-pairs PMs rotor (prime
June 23, 2010. Corresponding author: T. Lubin (e-mail: thierry.lubin@green. mover) by the Q pole pieces. The obtained field interacts with
uhp-nancy.fr).
Color versions of one or more of the figures in this paper are available online the pole-pairs PMs rotor to transmit torque to the load (outer
at http://ieeexplore.ieee.org. rotor or pole pieces) at a different speed. The combination
Digital Object Identifier 10.1109/TMAG.2010.2044187 results in the highest torque transmission capability of
0018-9464/$26.00 © 2010 IEEE
2612 IEEE TRANSACTIONS ON MAGNETICS, VOL. 46, NO. 7, JULY 2010

using separation of variables, we now consider the solution of

Laplace’s equation for the slots and the air-gap subdomains and
the Poisson’s equation for the PMs subdomains. For the sake
of clarity and simplicity of the general solutions in the different
subdomains, we adopt the following notations:

(2)

(3)

A. General Solution of Laplace’s Equation in the th Slot

Subdomain (Region i)
The th slot subdomain and the associated boundary condi-
Fig. 2. ith slot subdomain with its boundary conditions. tions are shown in Fig. 2. We have to solve the Laplace’s equa-
tion in a domain of inner radius and outer radius delim-
the gear. If the pole pieces are kept stationary, the resulting gear ited by the angles and
ratio is [1].
The geometrical parameters are
— for the inner rotor, the radius of the yoke and the outer
radius of the PMs ; (4)
— for the ring, the inner and outer radii of the slots and The boundary conditions for the th slot domain are (the tan-
, respectively; gential component of the magnetic field at the sides of the slot
— for the outer rotor, the inner radius of PMs and the inner are null)
radius of the yoke .
The slot opening angle is . The angular position of the th slot (5)
is defined as
The continuity of the normal component of the flux density
(1) between the th slot and the air-gap subdomains leads to

where is the initial angular position of the pole-pieces ring. (6)

The following assumptions are adopted: (7)
• end effects are neglected;
• permeability of the iron is infinite (the magnetic field in the The general solution of (4) can be found by using the method
iron vanishes); of the separation of variables, the solution can be written as [14]
• relative recoil permeability of the magnets is .
As shown in Fig. 1, the whole domain is divided into five
subdomains: the inner and outer air-gap subdomains (regions II
and III), the inner and outer PMs subdomains (regions I and IV),
and the slots subdomains (regions i). The th slot subdomain (8)
shape is shown in Fig. 2. The subdomains I, II, III, and IV have
annular shapes. where , , , and are arbitrary constants.
A magnetic vector potential formulation is used in 2-D polar Considering the continuity conditions (6) and (7), the general
coordinates to describe the problem. According to the adopted solution of the magnetic vector potential in the th slot domain
assumptions, the magnetic vector potential has only one com- is rewritten as
ponent along the z-direction and only depends on the and
coordinates. The notations used in the paper are
for the inner PMs subdomain
for the inner air-gap subdomain (9)
for the th slot subdomain
for the outer air-gap subdomain where is a positive integer, is defined by (3).
for the outer PMs subdomain. The constants , , , and are determined using a
Fourier series expansion of the inner and outer air-gap magnetic
vector potentials and over the slot in-
terval
III. ANALYTICAL MODEL
The solution of any partial differential equation (PDE) de-
pends on the domain in which the solution is to be valid as well (10)
as the boundary conditions that the solution must satisfy. By
LUBIN et al.: ANALYTICAL COMPUTATION OF THE MAGNETIC FIELD DISTRIBUTION IN A MAGNETIC GEAR 2613

Fig. 4. Schematic representation of f () along the inner air-gap at r = R .

with

(17)
elsewhere
Fig. 3. Inner air-gap subdomain (region II) with its boundary conditions.
where is the magnetic vector potential in the th slot
given by (9). The distribution of along the air-gap domain
interval is schematically shown in Fig. 4.
(11)
The general solution of (14) is well known [15] (periodic
Sturm-Liouville problem in an annulus). By taking into account
the boundary conditions (15) and (16), the general solution of
(12) the magnetic vector potential in the inner air-gap can be written
as

(13)

The expressions for the coefficients , , , and are

given in the Appendix.

B. General Solution of Laplace’s Equation in the Air-Gap

Subdomains
(18)
1) Inner Air-Gap Subdomain (Region II): The inner air-gap
subdomain and the associated boundary conditions are shown
in Fig. 3. The problem to solve is where is a positive integer, and are
defined by (2) and (3). The coefficients , , and
are determined using a Fourier series expansions of
and over the air-gap interval
(14)

The continuity of the tangential component of the magnetic (19)

field at leads to

(15) (20)

The boundary condition at the radius is more difficult

to handle because of the existence of the slots as shown in Fig. 1. (21)
Considering the continuity of the tangential magnetic field at the
interface between the slots and the air-gap and considering that
the tangential magnetic field is equal to zero elsewhere (infinite
permeability of the ferromagnetic pieces), the boundary condi- (22)
tion at can be written as

(16) The expressions for the coefficients , , , and

are given in the Appendix.
2614 IEEE TRANSACTIONS ON MAGNETICS, VOL. 46, NO. 7, JULY 2010

The Neumann problem (14), (15) and (16) can have a solution
only if [15]

(23)

The treatment of (23) yields to the following relations be-

tween the coefficient defined in (10) and (11)

(24)

The flux density distribution in the inner air-gap can be de-

duced from the magnetic vector potential by

(25)
Fig. 5. Outer air-gap subdomain (region III) with its boundary conditions.

(26)
with
The radial and the tangential components of the magnetic flux
in the inner air-gap are then (32)
elsewhere

By taking into account the boundary conditions (30) and (31),

the general solution of the magnetic vector potential in the outer
air-gap subdomain can be written as

(27)

(33)

where is a positive integer, and are

(28) defined by (2) and (3). The coefficients , ,
and are determined using a Fourier series expansions of
2) Outer Air-Gap Subdomain (Region III): The outer air-gap and over the air-gap interval
subdomain and the associated boundary conditions are shown in
Fig. 5. The problem to solve is
(34)

(29) (35)
The boundary condition for the outer air-gap domain at
is
(36)
(30)

The boundary condition at the radius can be written (37)

as

(31) The expressions for the coefficients , , , and

are given in the Appendix.
LUBIN et al.: ANALYTICAL COMPUTATION OF THE MAGNETIC FIELD DISTRIBUTION IN A MAGNETIC GEAR 2615

account the boundary conditions (39) and (40), the general so-
lution of the magnetic vector potential in the inner PMs subdo-
main can be written as

(41)
where

(42)

Fig. 6. Inner PMs subdomain (region I) with its boundary conditions. and

otherwise
(43)
where is a positive integer, is the number of pole-pairs of
the inner rotor and is defined by (2).
The coefficients and are determined using a Fourier
series expansion of over the interval

(44)

Fig. 7. Magnetization distribution along  -direction (inner PMs). (45)

C. General Solution of Poisson’s Equation in the PMs

Subdomains (Regions I and IV) The expressions for the coefficients and are given in
the Appendix.
1) Inner Rotor PMs Subdomain (Region I): The inner rotor 2) Outer PMs Subdomain (Region IV): The outer PMs sub-
PMs subdomain and the associated boundary conditions are domain and the associated boundary conditions are shown in
shown in Fig. 6. The problem to solve is Fig. 8. The problem to solve is

(38) (46)
where is the permeability of the vacuum and is the radial
magnetization of the magnets. The boundary conditions at and are,
The boundary conditions at and are, respectively
respectively
(47)

(39) (48)

(40) By taking into account the boundary conditions (47) and (48),
the general solution of the magnetic vector potential in the outer
The radial magnetization distribution is plotted in Fig. 7, PMs subdomain can be written as
where is the remanence of the magnets and is the initial
phase angle of the inner rotor. The radial magnetization can be
expressed in Fourier’s series and replaced in (38).
According to the superposition law, the general solution of
(38) is the sum of the general solution of the corresponding (49)
Laplace’s equation and a particular solution [15]. Taking into
2616 IEEE TRANSACTIONS ON MAGNETICS, VOL. 46, NO. 7, JULY 2010

TABLE I
PARAMETERS OF THE MODEL

Fig. 8. Outer PMs subdomain (region IV) with its boundary conditions.

where

(55)

(50) where

and

otherwise
(51)
where is a positive integer, is the number of pole-pairs of
the outer rotor, is the initial phase angle of the outer rotor (56)
and is defined by (2). The coefficients and
are determined using a Fourier series expansion of A similar expression can be obtained for the electromagnetic
over the interval torque computation in the outer air-gap.

IV. APPLICATION EXAMPLE 1

(52)
In order to validate the proposed model, the analytical re-
sults have been compared with 2-D finite element simulations
obtained using FEMM software [16]. In the finite-element anal-
(53) ysis, the surfaces of the inner and outer rotors yokes as well as
those of the ferromagnetic pole-pieces have been modeled by
homogeneous Neumann boundary conditions as in the analyt-
The expressions for the coefficients and are given ical study. The mesh in the air-gap and in the slot regions has
in the Appendix. been refined until convergent results are obtained.
D. Electromagnetic Torque The geometrical parameters given in Table I are considered
in the simulation studies. The analytical solutions in the air-gap,
The electromagnetic torque is obtained using the Maxwell in the PMs and in the slots domains have been computed with
stress tensor. A circle of radius in the inner air-gap sub- a finite number of harmonic terms and as indicated in
domain is taken as the integration path so the electromagnetic Table I.
torque is expressed as follows:
A. Flux Density Distribution
(54) Fig. 9 shows the magnetic flux lines for the magnetic gear
excited by both permanent magnet rotors. The phase angles
and of the rotors PMs are fixed to zero. The corresponding
where is the axial length of the magnetic gear. According to flux density distributions (radial and tangential components) in
(27) and (28), the analytical expression for the electromagnetic the middle of the inner air-gap and in the middle
torque becomes of the outer air-gap are plotted, respectively, in
LUBIN et al.: ANALYTICAL COMPUTATION OF THE MAGNETIC FIELD DISTRIBUTION IN A MAGNETIC GEAR 2617

Fig. 9. Magnetic flux line distribution for the studied magnetic gear excited by
both PMs rotors (' =0 ,' =0 and  =0 ).

Fig. 11. Flux density distribution in the middle of the outer air-gap (r =
6:3 cm): (a) radial component and (b) tangential component.

Fig. 12. Torque exerted on the inner rotor with the pole-pieces ring and the
outer rotor fixed ('=0 and  =0 ).

with those obtained by the FEM. Fig. 12 shows that a maximum

torque of 75 Nm is obtained for a value of around 51 .
Fig. 13(a) and (b) shows the variation of the electromagnetic
torque which is exerted, respectively, on the inner and outer ro-
Fig. 10. Flux density distribution in the middle of the inner air-gap (r = tors. The pole-pieces ring is fixed while the inner and outer ro-
5 1 cm)
: : (a) radial component and (b) tangential component. tors PMs rotate in opposite direction as

Fig. 10 and Fig. 11. One can observe a very good agreement (57)
between the analytical and finite element predictions for both
radial and tangential components.
The starting point for the phase angle of the inner rotor PMs is
fixed at that corresponds to a torque value exerted on
B. Torque
the inner rotor of 61 Nm (see Fig. 12). From Fig. 13(a) and (b),
Fig. 12 shows the variation of the torque which is exerted on one can observe as expected that the magnetic gear amplifies the
the inner rotor while keeping the pole-pieces ring and the outer mean torque value by the gear ratio of 3/2.
rotor fixed. The inner rotor rotates with a phase angle varying As shown in Figs. 13, the torque ripples are important in the
from 0 to 90 . The analytical results are in good agreement studied example. Indeed, the low order harmonics are present in
2618 IEEE TRANSACTIONS ON MAGNETICS, VOL. 46, NO. 7, JULY 2010

Fig. 14. Magnetic flux line distribution for the magnetic gear excited by both
PMs rotors (' =0 ,' =0 and  =0 ).

Fig. 13. Electromagnetic torque exerted on the inner rotor (a) and on the outer
rotor (b).

the flux density waveform of both air-gaps. This results in a low

filtering of the air-gaps which leads to high pulsating torques.
This effect can obviously be avoided by properly choosing
and for a given gear ratio [1]. Once again, very good agree-
ment is obtained between numerical and analytical results.

V. APPLICATION EXAMPLE 2
Another example is considered in this section. The same ge-
ometrical parameters as in Table I are used except for the pole-
pair numbers which are now , , which
leads to a gear ratio equal to 4.33.
The magnetic flux lines distribution for this configuration are
shown in Fig. 14 (the slot opening is fixed to ).
The radial and tangential flux densities on a circle of the inner
and outer air gaps are given in Fig. 15 and Fig. 16. A good
agreement is noticed between the analytical and the numerical
computations.
Fig. 17 shows the static torque versus position of the inner
rotor. Compared to the configuration given in the first design ex- Fig. 15. Flux density distribution in the middle of the inner air-gap (r =
ample, the maximum torque is lower. However, the transmitted 5 1 cm)
: : (a) radial component and (b) tangential component.
torque on the outer rotor is more important for the second ex-
Again, we notice a close agreement between the analytical
ample. Indeed, the gear ratio is more important for this config-
and numerical simulations.
uration (4.33 instead of 1.5).
Furthermore, the torque ripples are practically inexistent for VI. CONCLUSION
example 2 as it can be observed in Fig. 18. This confirms the
influence of the pole-pairs combination on the torque pulsations In this paper, an exact 2-D analytical method for predicting
[1]. the magnetic field distribution in a magnetic gear has been pre-
To show the influence of the slot opening, Fig. 19 gives the sented. The Laplace’s and Poisson’s equations in polar coordi-
variation of the electromagnetic torque exerted on the inner rotor nates have been solved by the technique of separation of vari-
against the “slot opening to tooth pitch” ratio. One can observe ables in the slots, air-gap and PMs subdomains. The solutions
that the maximum torque is obtained for a ratio of 0.5. have been obtained using boundary and continuity conditions
LUBIN et al.: ANALYTICAL COMPUTATION OF THE MAGNETIC FIELD DISTRIBUTION IN A MAGNETIC GEAR 2619

Fig. 16. Flux density distribution in the middle of the outer air-gap (r = Fig. 18. Electromagnetic torque exerted on the inner rotor (a) and on the outer
6:3 cm): (a) radial component and (b) tangential component. rotor (b).

Fig. 17. Torque exerted on the inner rotor with the pole-pieces ring and the Fig. 19. Maximal torque on the inner rotor versus “slot opening to tooth pitch”
outer rotor fixed (' = 0 and  = 0). ratio.

between the subdomains. Flux density and torque computations

are in close agreement with these of FE predictions. The pro- (A.3)
posed analytical model presents lower computational time than
FEM. Hence, it will be used in future work as a tool for design
optimization of a magnetic gear.
(A.4)
APPENDIX
For the determination of the integration coefficients, we have The development of (A.1) and (A.2) gives the following func-
to calculate integrals of the form tions that will be used in the expressions of the integration co-
efficients
(A.1) — for

(A.5)
(A.2)
(A.6)
2620 IEEE TRANSACTIONS ON MAGNETICS, VOL. 46, NO. 7, JULY 2010

— for • Expressions of the coefficients , , and

for the outer air-gap subdomain.
The development of (35) and (37) gives

(A.7) (A.17)

(A.18)

(A.8) The coefficient and defined in (34) and (35) can

be written as
The development of (A.3) and (A.4) gives the following
functions: (A.19)

(A.9)

(A.10) (A.20)

• Expressions of the coefficients , , , and for

where Q is the number of rotor slots. The development of (A.19)
the inner air-gap subdomain.
and (A.20) gives
The development of (19) and (21) gives

(A.11)

(A.12)

The coefficient and defined in (20) and (22) can be

written as (A.21)

(A.13)

(A.14)

(A.22)
where Q is the number of rotor slots. The development of (A.13)
and (A.14) gives
• Expressions of the coefficients and , for the inner
PMs subdomain (44) and (45)

(A.23)

(A.24)

(A.15)
• Expressions of the coefficients and , for the outer
PMs subdomain (52) and (53)

(A.25)

(A.26)

(A.16) • Expression of the coefficients , , , and for the

th slot subdomain.
LUBIN et al.: ANALYTICAL COMPUTATION OF THE MAGNETIC FIELD DISTRIBUTION IN A MAGNETIC GEAR 2621

The treatment of (12) and (13) yields to the following linear [3] S. Mezani, K. Atallah, and D. Howe, “A high-performance axial-field
relations: magnetic gear,” J. Appl. Phys., vol. 99, 08R303, 2006.
[4] P. O. Rasmussen, T. O. Andersen, F. T. Jorgensen, and O. Nielsen,
“Development of a high-performance magnetic gear,” IEEE Trans. Ind.
Appl., vol. 41, no. 3, pp. 764–770, May/Jun. 2005.
[5] K. T. Chau, D. Zhang, J. Z. Jiang, and L. Jian, “Transient analysis
of coaxial magnetic gears using finite element comodeling,” J. Appl.
Phys., vol. 103, p. 07F101, 2008.
[6] K. Atallah, J. Rens, S. Mezani, and D. Howe, “A Novel “pseudo” di-
rect-drive brushless permanent magnet machine,” IEEE Trans. Magn.,
vol. 44, no. 11, pp. 4349–4352, Nov. 2008.
[7] K. T. Chau, D. Zhang, J. Z. Jiang, C. Liu, and Y. Zhang, “Design of a
magnetic-geared outer-rotor permanent-magnetic brushless motor for
electric vehicles,” IEEE Trans. Magn., vol. 43, no. 6, pp. 2504–2506,
(A.27) Jun. 2007.
[8] L. Jian, K. T. Chau, and J. Z. Jiang, “A magnetic-geared outer-rotor per-
manent-magnet brushless machine for wind power generation,” IEEE
Trans. Ind. Appl., vol. 35, no. 3, pp. 954–992, May/Jun. 2009.
[9] B. Ackermann and R. Sottek, “Analytical modeling of the cogging
torque in permanent magnet motors,” Elect. Eng., vol. 78, no. 2, pp.
117–125, Mar. 1994.
[10] Z. Q. Zhu and D. Howe, “Instantaneous magnetic-field distribution
in brushless permanent-magnet dc motor, part III: Effect of slotting,”
IEEE Trans. Magn., vol. 29, no. 1, pp. 143–151, Jan. 1993.
[11] D. Zarko, D. Ban, and T. A. Lipo, “Analytical calculation of magnetic
field distribution in the slotted air gap of a surface permanent-magnet
motor using complex relative air-gap permeance,” IEEE Trans. Magn.,
(A.28) vol. 42, no. 7, pp. 1828–1837, Jul. 2006.
[12] Z. J. Liu and J. T. Li, “Analytical solution of air-gap field in permanent
magnet motors taking into account the effect of pole transition over
The treatment of (10) and (11) yields to the following linear slots,” IEEE Trans. Magn., vol. 43, no. 10, pp. 3872–3882, Oct. 2007.
relations: [13] Z. J. Liu and J. T. Li, “Accurate prediction of magnetic field and mag-
netic forces in permanent magnet motor using an analytical solution,”
IEEE Trans. Energy. Convers., vol. 23, no. 3, pp. 717–726, Sep. 2008.
[14] T. Lubin, S. Mezani, and A. Rezzoug, “Exact analytical method for
magnetic field computation in the air-gap of cylindrical electrical ma-
chines considering slotting effects,” IEEE Trans. Magn., vol. 46, no. 4,
pp. 1092–1099, Apr. 2010.
[15] S. J. Farlow, Partial Differential Equations for Scientists and Engi-
neers. New York: Dover, 1993.
[16] D. C. Meeker, Finite Element Method Magnetics ver. 4.2, (1 April 2009
Build) [Online]. Available: http://www.femm.info

Thierry Lubin was born in Sedan, France, in 1970. He received the M.S. degree
from the University of Paris 6, France, in 1994 and the Ph.D. degree from the
(A.29) University Henri Poincaré, Nancy, France, in 2003.
He is currently a Lecturer in electrical engineering at the University of Nancy
at the Groupe de Recherche en Electrotechnique et Electronique de Nancy. His
interests include modeling and control of electrical machine and applied super-
conductivity in electrical devices.

Smail Mezani was born in Algiers, Algeria, in 1974. He received the engineer
diploma and the magister degree from the University of Sciences and Tech-
nology Houari Boumediene, Algiers, Algeria, in 1996 and 1999, respectively.
He received the Ph.D. degree from the Institut National Polytechnique de Lor-
raine, France, in 2004.
From 2004 to 2005, he was a Research Associate with the Department of
(A.30) Electronic and Electrical Engineering, University of Sheffield, Sheffield, U.K.,
where he worked on magnetic gears and “pseudo” direct drive electrical ma-
chines. He is currently a Lecturer with the University Henri Poincaré, Nancy,
We have to solve a system of linear equations with the same France, at the Groupe de Recherche en Electrotechnique et Electronique de
number of unknowns. By rewriting the above equations in ma- Nancy where his research interests include the applications of superconductors
in electromechanical devices.
trix and vectors form, a numerical solution can be found by
using mathematical software (Matlab or Maple).
REFERENCES Abderrezak Rezzoug received the Doctorat d’Etat degree from the Institut Na-
[1] K. Atallah and D. Howe, “A novel high-performance magnetic gear,” tional Polytechnique de Lorraine (INPL), France, in 1987.
IEEE Trans. Magn., vol. 37, no. 4, pp. 2844–2846, Jul. 2001. He was with the INPL as an Assistant Professor until 1991. He then joined
[2] K. Atallah, S. Calverley, and D. Howe, “Design, analysis and realiza- the Université Henri Poincaré, Nancy, France, as a Professor. His interests are in
tion of a high-performance magnetic gear,” Proc. Inst. Elect. Eng., vol. superconducting applications to electrical devices, and the control and diagnosis
151, no. 2, pp. 135–143, 2004. of electrical machines.