Analytical design of a high-torque flux-switching permanent magnet machine by a simplified lumped parameter magnetic circuit model

Abstract

This paper presents how to analytically design a high-torque three-phase flux-switching permanent magnet machine with 12 stator poles and 14 rotor poles. Firstly, the machine design parameters are studied addressing on high output torque and its flux distribution is also investigated by finite-element method (FEM) analysis. Then a simplified lumped parameter magnetic circuit model is built up for analyzing design parameters. And a design procedure is also presented. The analytically designed machine is verified by FEM simulations.

Full text

ΦAbstract -- This paper presents how to analytically design a
high-torque three-phase flux-switching permanent magnet
machine with 12 stator poles and 14 rotor poles. Firstly, the
machine design parameters are studied addressing on high
output torque and its flux distribution is also investigated by
finite-element method (FEM) analysis. Then a simplified
lumped parameter magnetic circuit model is built up for
analyzing design parameters. And a design procedure is also
presented. The analytically designed machine is verified by
FEM simulations.
Index Terms — Finite element method, permanent magnet
machine, magnetic circuits, flux-switching, high torque.
I. NOMENCLATURE
Br Magnet remanence
Bt Average flux density in stator tooth top
cs Ratio of stator tooth width to stator pole pitch
Do Machine outer diameter
Fpm Magnet MMF
g Airgap length
Hrb Rotor iron-back thickness
Hsb Stator iron-back thickness
Ht Stator tooth height
Hrt Rotor tooth height
J Current density
kf Winding factor
kpm Magnet temperature coefficient
kσ Flux leakage factor
L Machine active axial length
lpm Magnet thickness
Pg
Permeance in airgap
Pgil Magnet inner air-leakage permeance
Pgol Magnet outer air-leakage permeance
Ppm Magnet permeance
Pr Rotor pole number
Prb Rotor back-iron permeance
Prt Rotor tooth permeance
Ps Stator pole number
Psb Stator back-iron permeance
Pst Stator tooth permeance
S Electrical loading
T Torque
Wrt Rotor tooth width
Ws Stator slot opening
Wst Stator tooth width
λ Ratio of stator inner diameter to outer diameter
τr Rotor pole pitch
τs Stator pole pitch
μ0 Permeability of free space
μpm Magnet relative permeability
ωm Machine synchronous angular speed
This work was supported by Research Council of Norway (NFR).
The authors are with the Department of Electrical Engineering,
Norwegian University of Science and Technology, O.S. Bragstads plass 2E,
level 4, No 7491, Trondheim, Norway. (anyuan.chen@elkraft.ntnu.no).
(robert.nilssen@elkraft.ntnu.no). (arne.nysveen@elkraft.ntnu.no).
II. INTRODUCTION
LUX-SWITCHING permanent magnet (FSPM)
machines having PMs in the stator with doubly salient
stator and rotor structure like a switched reluctance
machine combine the advantage of a conventional PM
machine and a switched reluctance machine. They have
therefore high reliability, high torque /power density and
relatively high efficiency, hence preferable for reliability
premium applications. Today FSPM machines have been
presented for different applications, such as in aerospace,
automotive and wind energy applications [1]-[3]. Several
papers have investigated different FSPM machines with
various stator and rotor pole combinations and their
characteristics [4]-[11]. In [4] and [9] a FSPM machine with
12 stator poles and 14 rotor poles (12/14 poles) as shown in
Fig. 1 has been investigated. Compared with a 12/10 pole
machine, this machine can provide higher torque density
with less torque ripple.
Today FSPM machines are generally designed as an
initial machine, in which Hsb = lpm = Wrt = Ws = Wst = τs /4 as
shown in Fig. 2, thereafter the optimal parameters and /or
performance were studied by either finite element method
(FEM) simulations or lumped parameter magnetic circuit
model [8] [11][14]. Such initially designed FSPM machines
usually have highly saturated stator iron teeth that is
normally beneficial for a 12/10 pole machine to improve the
output torque. But for a 12/14 pole machine, the high
saturation will lead to a torque decrease due to the high flux
leakage between the stator and rotor [9]. So a new approach
is required to design a high-torque 12/14 pole machine. This
paper introduces a simplified lumped parameter magnetic
circuit model to analytically design the machine. Firstly the
machine design parameters are studied addressing on high
output torque. Then the flux distribution of a typical 12/14
FSPM machine is investigated by FEM simulations, based
on which a lumped parameter magnetic circuit model is built
up for finding optimal design parameters. Finally, the
analytically designed machine is verified by FEM
simulations.
Fig. 1. Cross section of a 12/14 pole machine.
Analytical Design of a High-Torque Flux-Switching
Permanent Magnet Machine by a Simplified
Lumped Parameter Magnetic Circuit Model
Anyuan Chen, Robert Nilssen and Arne Nysveen
F
XIX International Conference on Electrical Machines - ICEM 2010, Rome
978-1-4244-4175-4/10/$25.00 ©2010 IEEE

Fig. 2. Part of an initial machine in a plain form.
III. MACHINE CONSTRUCTION
Fig. 1 shows the machine construction. Each phase
winding of the machine consists of four coils and each coil is
concentrated around two stator teeth with a magnet inset in
between. The magnets are circumferentially magnetized and
the magnetization is reversed in polarity from one magnet to
the next. For each phase the flux in coils 1 and 2 are
respectively the same as that in the corresponding phase
coils 3 and 4 due to the symmetrical machine construction.
The coil-flux linkage of each phase (the summary of four
coils) is essentially sinusoidal with respect to the rotor
position and has a period of τr as shown in Fig. 3. And it
reaches the peak value when the rotor is at the d-axis
position of the phase as shown Fig. 4 (a). At this position
the fluxes in the four coils of the phase are the same, as can
be seen in Fig. 5 in which the fluxes in coils A1, A2, A3 and
A4 are the same.
0 2 4 6 8 10 12 14 16 18 20 22 24 26
-1
-0.5
0
0.5
1
Rotor position [mec hanical degree]
Normalized f lux (mWb)
Fig. 3 Coil-flux linkage in one phase
(a) (b)
Fig. 4. Rotor at (a) d- axis position where the coil-flux linkage is maximum,
(b) q-axis where the coil-flux linkage is zero.
Fig. 5. Flux distribution at the d-axis position of phase a.
IV. MACHINE DESIGN
A. Design parameters
If neglecting machine losses the torque of a 12/14 pole
FSPM machine can be expressed as [9]
2
22
2
4rt o s
s
TkPBDLSc
P
σ
πλ
=. (1)
where Bt is the average flux density in the stator tooth tops at
the d-axis position, and kσ is the leakage factor representing
the effective flux for torque production at the d-axis position
and evaluated here by
34
3
p
p
p
k
σ
Φ−Φ
=Φ (2)
where Φp3 and Φp4 are respective the flux through the teeth
P3 and P4 in Fig. 8.
The parameters Do and L are generally constrained by the
available volume of a specific case and are therefore fixed.
In this paper they are respectively 100 mm and 200 mm.
Bt is an important design parameter. Ideally without
considering iron saturation, the higher Bt, the higher torque
from (1) would be produced. In reality, along with an
increase of Bt the value of kσ will decrease because of the
increased iron saturation. This has been proven by FEM
analysis as presented in Fig. 6, in which Bt is varied by using
different magnet materials with various Br from 0.6 -1.2 T,
whilst keeping the machine dimension parameters
unchanged. Since the output torque depends on both Bt and
kσ, their product value that directly indicates the torque
capability of the machine is also shown in the figure. It is
observed that the product reaches its peak value when Bt is
1.8 ~1.9 T. With a further increased Bt the leakage flux Φp4
increases more than the total flux Φp3 due to the iron
saturation as shown in Fig. 7. As consequence, the effective
flux, Φp3 - Φp4, for torque production decreases, hence the
leakage factor k
σ determined by (2). In this paper Bt is
chosen to be 1.8 T, which is typically the saturation flux
density of iron materials.
1.55 1.6 1.65 1. 7 1.75 1.8 1.85 1.9 1. 95
0.65
0.7
0.75
0.8
0.85
Leakage fact or
Average flux density in stator toot h Bt (T)
1.55 1.6 1.65 1. 7 1.75 1.8 1.85 1.9 1. 95 1.2
1.25
1.3
1.35
1.4
Product of kdel*Bt
(
T
)
Leakage factor
Product of B t and kdel
Fig. 6. kσ and the product of Bt*kσ as function of Bt from FEM analysis.
1.55 1.6 1.65 1.7 1.75 1.8 1.85 1.9 1.95 2
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Average flux density in stator tooth B t (T)
Flux (mWb)
Tota l flux
Effec tive flu x
Leakage flux
Fig. 7. The total, effective and leakage flux at different flux density Bt
A1
A2
A3
A4

λ and cs are selected as design variables here, then the
other machine parameters can be expressed in terms of them
as follows:
The tooth width is calculated as
s
tss
Wc
τ
=. (3)
where τs is the stator pole pitch and calculated by
/
s
os
D
P
τπλ
=. (4)
The magnet width is determined by
()
1/2
pm o
WD
λ
=− . (5)
The stator iron-back thickness Hsb is chosen so that the
maximum flux density in the stator back iron is the same as
Bt. By so the iron does not get saturated and the stator
winding can have the maximum available area. FEM
simulations show that Hsb should be around 70% of Wst to
avoid the saturation. The result is the same as that for a
12/10 pole machine [11]. The rotor iron-back thickness Hrb
is chosen to be the same as Hsb.
The height of the rotor tooth Hrt determines the rotor
saliency. Generally the reluctance torque of the machine is
negligible. However, the output torque can be slightly
increased with higher Hrt. The research based on a 12/10
pole machine shows the maximum torque is obtained when
the rotor tooth height is around twice the stator tooth width
[11]. This conclusion is employed for this machine design
because of the similar construction of the two machines and
the negligible reluctance torque value.
The magnet thickness lpm may be started with a small
initial value, for example, 1 mm here, so that Bt is less than
1.8 T. Its final value will be found out later in designed
machine section. Then the electrical loading S is determined
by the available copper area as
/( )
cu f o
SAJk D
πλ
=. (6)
where Acu is the copper area in the stator and given by
()
22
2
22
oo
cu sb s t st pm
DD
A
HPHWl
λ
ππ
⎛⎞⎛⎞
=−− − +
⎜⎟⎜⎟
⎝⎠⎝⎠ .(7)
where Ht is the stator tooth height and calculated by
()
1/2
tosb
H
DH
λ
=− − . (8)
The rotor tooth width Wrt, unlike the stator parameters,
can be freely chosen without influencing the other design
parameters. Although Wrt does not directly appear in torque
equation (1), it affects the airgap reluctances, further the
leakage factor k
σ and the output torque. The principle for
selecting Wrt is to make kσ value as high as possible. For a
12/14 machine with lpm = Wst = τs /4, the optimal Wrt is found
to be τr /3 [14]. In the discussed case both lpm and Wst are
varied with different λ and cs, the rotor tooth width is chosen
so that the left edge of the rotor tooth at the d-axis aligns
with the left edge of the stator tooth as shown in Fig. 8 (T2
and P3). By so the maximum overlapped area between the
stator and rotor teeth is obtained for the chosen Wst and Wrt
at the d-axis. Then the rotor teeth width is determined by (9).
2/2
rt st pm r
WWl
τ
=+− (9)
where
()
2/
ro r
D
gP
τλ π
=− . (10)
It should be noted that Wrt determined by (9) may not
give the maximum leakage factor for the designed machines.
If necessary, the rotor tooth width may be optimized by
FEM analysis afterwards.
B. Flux distribution
So far all the machine parameters in (1) are known
except kσ. Since Bt and kσ are calculated at d-axis position, it
is of interest to investigate the flux distribution at that
position. Fig. 5 shows the flux distribution of a typical 12/14
pole machine at the d-axis (phase a here). The four coils of
phase a have the same flux linkage, and the flux in each coil
is mainly from the three magnets near the phase coil as
shown in Fig. 8 (a). So it is sufficient only to analyze the
flux paths of these three magnets for evaluating Bt and kσ.
The flux distributions in the airgap between the stator and
rotor teeth are approximated in Fig. 8 (b). The flux
distributions in the arigap are slightly varied depending on
the specific values of Wst, Wrt, and lpm.
C. Lumped parameter magnetic circuit model
Based on the flux paths in Fig. 8, a lumped parameter
magnetic circuit model of one fourth the machine is built up
at no load condition as shown in Fig. 9. Compared with the
model in [8] and [14] where half the machine is modeled,
this model is simplified. In the presented model the end
effect is not considered because the machine axial length is
relatively long compared with its diameter (L/Do=2).
The permanent magnets are simply modeled as a MMF
by (11) and their permeances are calculated by (13) .
0
/( )
pm pm a pm
FlB
μμ
= (11)
where Ba is the magnet remanence at the ambient
temperature Tθ and it is determined by
(a)
(b)
Fig. 8. (a) Flux distribution at the d-axis position. (b) Approximation of the
flux paths in a plain form.
A
1

Fig. 9. Magnetic circuit with PMs represented by flux sources.
()
0
1( )
ar pm
B
BkTT
θ
=+ −. (12)
where Br is the magnet remanence at room temperature T0.
0/
p
mpm pmpm
PLWl
μμ
= (13)
The permeances of the iron parts, Pst, Prt and Psb, are
determined by
0/
r
PAl
μμ
=. (14)
where A and l are respectively the cross-sectional area and
the length of the corresponding iron part, μr is the relatively
permeability of the iron part and determined by iteration
form the B-H curve of the lamination material.
To calculate the airgap permeances Pg, Pgil and Pgol the
method presented in [8] is employed here.
D. Approximation of Magnetization curve
An expression for the relation between the flux density
and the field intensity (B-H curve) is required for calculating
the magnetic reluctance of the iron. In reality, it is very
difficult to find an expression that can exactly represent the
curve. Fortunately, equation (15) can be used to approximate
the magnetization curve [12].
0
() ( (coth( ) ))
i
ii i s
i
Ha
BH H M aH
μ
=+ − (15)
where Ms is saturation magnetization, and α is a material
dependent parameter, Bi and Hi are respectively the flux
density and field intensity in the corresponding iron part.
By varying the values of Ms and α, the shape of the curve
obtained from (15) can be changed. After several iterations,
the curve with Ms =1.5 MA/m and α =550 shown in Fig. 10
can be used as an approximation of the iron magnetization
curve.
012345678910
x 10
4
0
0.5
1
1.5
2
2.5
Magnetic field intens ity H (A/m)
Flux densi ty B (T)
Magnetization c urve
Approximated curve
Fig. 10. Magnetization curve and its approximation.
E. Magnetic circuit equations
Nodal analysis is employed to solve the magnetic circuit.
Each permanent magnet is represented by a flux source and a
flux resistance in parallel as shown in Fig. 9, in which Φpm is
given by
p
mpmpm
PFΦ= (16)
There are totally 17 nodes in the magnetic circuit and the
equations between the relationship of the these magnetic
variables are established as
(1) (1,1) (1, 2) . . . (1,17)
(2) (2,1) (2, 2). . . (2,17)
. . . . . .
.
. . . . . .
.
. .
.
(17)
s
s
s
PP P
PP P
Φ
⎡⎤
⎢⎥
Φ
⎢⎥
⎢⎥
=
⎢⎥
⎢⎥
⎢⎥
⎢⎥
Φ
⎢⎥
⎣⎦
(1)
(2)
.
.
. . . . .
(17,1) (17,2) . . . (17,17) (17)
F
F
PP P F
⎡
⎤⎡ ⎤
⎢
⎥⎢ ⎥
⎢
⎥⎢ ⎥
⎢
⎥⎢ ⎥
⎢
⎥⎢ ⎥
⎢
⎥⎢ ⎥
⎢
⎥⎢ ⎥
⎢
⎥⎢ ⎥
⎢
⎥⎢ ⎥
⎣
⎦⎣ ⎦
.(17)
where
Φs(1) - Φs(17) respectively represent the flux flowing into
the corresponding nodes from the flux sources, here Φs(1) =
Φs(4) = Φs(5) = Φpm and Φs(2) = Φs(3) = Φs(6) = - Φpm, the
others are zero.
P(m,n), m ≠ n, is the negative permeance value between
nodes m and n.
P(m,m) is the sum of the permeance of those branches
connected to node m.
F(1)- F(17) are respectively the magnetic potential at nodes
1-17.
When solving (17), the initial permeance value of each
iron part is set μr = 4000. Afterwards, μr
k
for kth iteration of
each iron part is updated based on previous calculation as the
follows:
1. Calculating the magnetic field intensity over the
corresponding iron part by
1
1
k
ki
i
F
Hl
−
−Δ
=. (18)
where ΔFi
k-1 is the magnetic potential drop over the
corresponding iron part, this value can be calculated from
the magnetic potentials at the nodes.
2. Updating μr according with the magnetization curve in Fig.
10 by
1
1
1
1
((coth()))
k
ki
is k
ki
rk
i
Ha
HM aH
H
μ
−
−
−
−
+−
=. (19)
3. Repeating the procedure with the updated μr value until
Bi
k and Hi
k are satisfactory with (15). Bi
k is obtained from
/
kk
ii
B
FP A=Δ . (20)
It should be mentioned that the magnetic saturation over
teeth T4 is underestimated since the flux from P7 is not
considered. Fortunately, the saturation is negligible due to
the large airgap reluctance because of the small or even no
overlapped iron area between teeth P6/P7 and T4 as shown
in Fig. 8.

F. Design procedure
To calculate the maximum output torque from (1) with
certain λ and cs, the value of kσ should be known when Bt is
1.8 T. This is achieved by gradually increasing lpm based on
the given initial value, then recalculating Wrt from (9) and
further all the permeances in Fig. 9. Thereafter, solving the
model to figure out Φp3 and Φp4 and further k
σ and Bt.
Repeating the process until Bt = 1.8 T. Now lpm and kσ are
known and the output torque can be evaluated by (1).
Fig. 11 presents the design procedure.
G. Designed machine
Fig. 12 shows the leakage factor as function of λ and cs.
For each cs the value of kσ increases along with an increase
of λ, and for each λ there is an optimal cs where kσ reaches its
maximum value.
Fig. 13 presents the output torque as function of λ and cs.
There is an optimal λ and cs giving the maximum output
torque. Fig. 14 and Fig. 15 respectively show the maximum
output torque with respect to split ratio λ and stator tooth
factor cs. It is found that the optimal λ is around 0.5 and cs is
around 0.25 for the discussed case here. Table I lists the
parameters of the designed machine.
It should be noted that the magnet demagnetization and
the maximum allowable temperature of the winding
insulation should be considered when selecting the current
density. This is out of the scope and therefore is not
discussed in this paper.
Fig. 11. Machine design process.
0.2
0.25
0.3
0.35
0.4
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
Stator tooth factor
Split ratio
Leakag factor
Fig. 12. Leakage factor kσ as function of λ and cs.
0.2
0.25
0.3
0.35
0.4
0.2
0.4
0.6
0.8
1
0
10
20
30
Stat or tooth factor
Split ratio
Output torque (Nm)
Fig. 13. Output torque as function of λ and cs.
0.3 0.35 0.4 0.45 0.5 0.55 0.6 0. 65
16
18
20
22
24
26
Split ratio
Maximum torque (N.m)
Fig. 14. Maximum output torque at different split ratio λ.
0.2 0.22 0. 24 0.26 0.28 0.3 0.32 0.34
16
18
20
22
24
26
Stat or tooth fac tor
Maximum torque (N.m)
Fig. 15. Maximum output torque at different stator tooth factor cs.
Table I Machine parameters
Paramete
r
value
D
o100 m
m
L
200 m
m
g
0.5 m
m
λ
0.5
c
s
0.25
P
s
12
P
r
14
W
rt 3.4 m
m
W
s
t3.2 m
m
H
s
b2.2 m
m
H
t23 m
m
H
rt 6.4 m
m
l
p
m2.4 m
m
B
r
1.16 T
k
f
0.6
J
4 A/mm
2
T
θ
150º
k
p
m-0.0004
K
-
1

V. FEM SIMULATION
To verify the result, the machine with the parameters
given in Table I is investigated by 2D-FEM simulations, in
which the B-H curve in Fig. 10 is employed for the iron
material and Br determined by (12) is set to 1.09 T to take
the temperature influence into account. Fig. 16 shows the
flux distribution of the machine at the d-axis position with
no load (J = 0), from which Bt and kσ are obtained. Fig. 17
presents the output torque from the simulation. And Table II
lists the results from both the lumped parameter magnetic
circuit model and the FEM simulations. The torque
calculated from the circuit model is about 3.3% higher than
that from the FEM simulations. They match each other
satisfactorily.
Fig. 16. Flux distribution of the designed machine at no load
0 5 10 15 20 25 30 35 40
0
5
10
15
20
25
30
Rotor position in mech. degree
Output t orque (N.m)
Fig. 17 Output torque of the machine from FEM
Table II Comparison of Bt, kσ and T
Analytical FEM
B
t (T) (no load) 1.8 1.76
kσ (no load) 0.67 0.69
T
(Nm) 25.4 24.6 (average)
VI. CONCLUSION
This paper has introduced a simplified lumped parameter
magnetic circuit model for analytically designing a high-
torque 12/14 pole FSPM machine. And the design procedure
of how to find out the optimal design parameters is also
presented. The design machine has been verified by FEM
simulations.
VII. REFERENCES
[1] Arwyn S. Thomas, Z. Q. Zhu, Richard L. Owen, Geraint W. Jewell
and David Howe, “Multiphase Flux-Switching Permanent-Magnet
Brushless Machine for Aerospace Application”, IEEE Trans. Ind.
Appl., vol. 45, No. 6, pp. 1971-1981, November/December 2009.
[2] Fang, Z. X., Wang, Y., Shen, J. X., Huang, Z. W., “Design and
analysis of a novel flux-switching permanent magnet integrated-
starter-generator” , PEMD 2008.
[3] Jiangzhong Zhang, Ming Cheng and Zhe Chen, “Optimal design of
stator interior permanent magnet machine with minimized cogging
torque for wind power application”, Energy Conversion and
management 49, 2008, pp. 2100-2105.
[4] J. T. Chen, Z. Q. Zhu, A. S. Thomas and D. Howe, “Optimal
combination of stator and rotor pole numbers in flux-Switching PM
brushless AC machines”, the proceeding of ICEMS,2008, vol. 44, pp.
4659 – 4667.
[5] W. Z. Fei and J. X. Shen, “Comparative study and optimal design of
PM switching flux motors”. UPEC’06, 6-8 Sept. 2006, vol. 2 pp. 695-
699.
[6] Yu Chen, Z. Q. Zhu and David Howe, “Three-Dimensional Lumped-
Parameter Magnetic Circuit Analysis of Single-Phase Flux-Switching
Permanent-Magnet Motor”, IEEE Trans. Ind. Appl., vol. 44, No. 6,
pp. 1701-1710, November/December 2008.
[7] Richard L. Owen, Z.Q. Zhu, Arwyn S. Thomas, Geraint W. Jewell and
David Howe, “ Alternate Poles Wound Flux-Switching Permanent-
magnet Brushless AC Machines”, IEEE Trans. Ind. Appl., vol. 46,
No, 2, pp. 790-797, March/April, 2010.
[8] Z. Q. Zhu, Y. Pang, D. Howe, S. Iwasaki, R. Deodhar, and A. Pride,
“Analysis of electromagnetic performance of flux-switching
permanent magnet machines by non-linear adaptive lumped parameter
magnetic circuit model,” IEEE Trans. Magn, vol.41, No.11, pp. 4277-
4287, November 2005.
[9] Anyuan Chen, Njål Rotevatn, Robert Nilssen and Arne Nysveen,
“Characteristic Investigations of a New Three-Phase Flux-Switching
Permanent Magnet Machine by FEM Simulations and Experimental
Verification” in the proceeding of ICEMS2009,15-18, Nov., 2009
Tokyo, Japan.
[10] Emmanuel Hoang, Michel Lecrivain, Mohamed Gabsi, “A new
structure of a switching flux synchronous poly-phased machine with
hybrid excitation”, the proceeding of Power electronics and
applications, 2007 European conference, pp. 1-8.
[11] Z.Q. Zhu, Y. Pang, J. T. Chen, Z. P. Xia, D. Howe, “Influence of
design parameters on output torque of flux-switch permanent magnet
machines”, IEEE VPPC, 2008, September 3-5, 2008, Harbin, China.
[12] P. Thelin and H-P Nee, “Calculation of the Airgap Flux Density of
PM Synchronous Motors with Buried Magnets Including Axial
Leakage and Teeth Saturation”, EMD’99, Canterbury, United
Kingdom, September 1999, pp.339-345.
[13] Wei Hua, Ming Cheng, Z. Q. Zhu and David Howe, “Analysis and
optimization of back EMF waveform of a flux-switching permanent
magnet motor”, IEEE Transactions on Energy Conversion, VOL. 23,
NO. 3, September 2008, pp 727-733.
[14] J. T. Chen and Z. Q. Zhu, “ Influence of Rotor Pole Number on
Optimal Parameters in Flux-Switching PM Brushless AC Machines by
Lumped Parameter Magnetic Circuit Model”, IEMDC 2009, May 3-6,
2009, pp 1451-1458.
VIII. BIOGRAPHIES
Anyuan Chen received the B.Sc. degree in electrical engineering from
Wuhan Institute of Technology, Wuhan, China in 1991, and then worked as
(senior) electrical engineer at several companies. In 2004 he received the
M.Sc. in Electrical Power Engineering from the Royal Institute of
Technology (KTH), Stockholm, Sweden. Now he is working toward the
Ph.D. degree in Norwegian University of Science and Technology (NTNU),
Trondheim, Norway.
His research interests include permanent magnet machine design and
electric drives.
Robert Nilssen received the M.Sc. and Dr.ing from the Norwegian Institute
of Technology (NTH), Trondheim Norway, in 1983 and 1988, respectively,
specializing in the field of Finite Element Analysis.
He was an advisor to Norwegian Research Institute of Energy supply
and SINTEF. Currently he is a professor in Electrical Engineering at the
Norwegian University of Science and Technology (NTNU). He current
research interests include design of electromagnetic components and
electrical machines, optimization and modeling. He is a co-founder of
several companies.
Arne Nysveen (M’00-SM’06) received the M.Sc. degree in Electrical
Power Engineering and the Dr.ing degree from the Norwegian Institute of
Technology (NTH), Trondheim, Norway, in 1988 and 1994, respectively.
From 1995 to 2002, he was a Research Scientist with ABB corporate
Research, Oslo, Norway, where his main research dealt with subsea power
supply and electrical power apparatus. Since 2002, he has been a professor
at the department of Electrical Power Engineering, Norwegian University of
Science and Technology (NTNU), Trondheim. He holds several patents on
subsea power equipment and electric machinery.