Post on 18-Apr-2020
EMR’14CoïmbraJune 2014
Summer School EMR’14“Energetic Macroscopic Representation”
« EMR and Inversion -based control of 2 series connected 5 -phase PMSMs »
Prof. Eric SEMAIL, Dr. Ngac Ky NGUYEN, Dr. Xavier KEST ELYNL2EP, Arts et Métiers ParisTech
EMR’14, Coimbra, June 20142
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Outline -
1. EMR of a 5-phase PMSM
2. EMR of 2 series connected 5-phase PMSMs
3. Inversion-based control
4. Experimental results
PMSM : Permanent Magnet Synchronous Motor
EMR’14CoïmbraJune 2014
Summer School EMR’14“Energetic Macroscopic Representation”
« EMR of a 5-phase PMSM »
EMR’14, Coimbra, June 20144
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
Decrease the phase current comparing to a three phase machine for a given low DC bus (48V) reduce the per-phase converter rating (automobile applications).
Increase fault-tolerance capabilities (open-circuit and short-circuit).
Reduce harmonic torque ripple.
More possibilities of stator winding configuration .
- EMR of a 5-phase PMSM -
Why multiphase machine?
EMR’14, Coimbra, June 20145
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- EMR of a 5-phase PMSM -
Assumptionson the 5-phase Permanent Magnet Synchronous Machines (PMSMs):
• no reluctance, no saturation effects
• regular spatial distribution of windings
ia
ib
ic
id
ie
VDCN
5-leg VSI 5-phase machine
νeΝ
EMR’14, Coimbra, June 20146
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- EMR of a 5-phase PMSM -
vsiv
machi
mi
svsiv
DC bus
iDC
VDC
mvsi
si
mi
si
me
se
Tmm
ΩTsm
Ω
Ttot
Ω Tload
Ω
mvsiv5-leg VSI
Concordiatransformation
fictitious machines
mechanicalcoupling
load
shaft
5
1 1 0 1 021 2 2 4 4cos sin cos sin
5 5 5 521 4 4 8 82 cos sin cos sin
5 5 5 55 21 6 6 12 12cos sin cos sin
5 5 5 521 8 8 16 16cos sin cos sin
5 5 5 52
C
π π π π
π π π π
π π π π
π π π π
=
Ttot=Tmm+Tsm
Two equivalent(d,q) machines
One 5-phase machine = two (d,q) machines
two classicalvector controls
SM
MM
EMR’14, Coimbra, June 20147
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- EMR of a 5-phase PMSM -
Secondary (d,q) Machine
3, 7, 13… 5h±2
Main (d,q) Machine
1, 9, 11… 5h±1 family of odd harmonics
EMF : Electromotive force or back-EMF
EMR’14CoïmbraJune 2014
Summer School EMR’14“Energetic Macroscopic Representation”
« EMR of 2 series connected 5 -phase PMSMs»
EMR’14, Coimbra, June 20149
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- EMR for 2 series connected PMSMs -
Replace two hydraulic engines bytwo electric motors in orientation controlsystem where one machine pulls and the other pushes
Thales Alenia Space Project (Ariane 6): 20132017
One of applications : aerospace launcher
EMR’14, Coimbra, June 201410
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- EMR for 2 series connected PMSMs -
i1a
i1b
i1c
i1d
i1e
VDC
i2a
i2b
i2c
i2d
i2e
N
5-leg VSI 5-phase machine 1 5-phase machine 2
VDC
5-leg VSI
Healthy operation
One of inverter switch or DC source fault
i1a
i1b
i1c
i1d
i1e
VDC
i2a
i2b
i2c
i2d
i2e
N
5-leg VSI 5-phase machine 1 5-phase machine 2
VDC
5-leg VSI
How can we control separately two machines in case of fault ?
EMR’14, Coimbra, June 201411
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- EMR for 2 series connected PMSMs -
i1a
i1b
i1c
i1d
i1e
VDC
i2a
i2b
i2c
i2d
i2e
N
5-leg VSI 5-phase machine 1 5-phase machine 2
connection
Solution : Swapping connection + new control algorithm based on EMR
Swapping connection
two fictitiousmachines
MM1 & SM1
two fictitiousmachines
MM2 & SM2
MM1 & SM2 in series
MM2 & SM1 in series
MM1
SM1 SM2
MM2
Control objective: 8 currents (4 equivalent fictitious machines)
DOF : 4 (5-leg VSI)
Swapping
Require a special connection
EMR’14, Coimbra, June 201412
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- EMR for 2 series connected PMSMs -
vVSI
vM1
iVSI
VDC
IDC
DCiM1
iM2
vM2
iM1
eM1
iM2
eM2
MM1
SM1
SM2
MM2
TMM1
TSM1
TSM2
TMM2
Ω1
Ω1
Ω2
Ω2
TM1
Ω1
Ω1
Tload-1
Ω2
Ω2TM2
Tload-2
DC bus
5-leg VSI
Concordiatransformation
equivalent windings
swapping connection
fictitious machines
mechanical coupling
shaft
loads
iMM1
eMM1
iSM2
iMM2
iSM1
eSM1
eSM2
eMM2
mVSI
iMM1 iSM2*=
iSM1 iMM2=
EMR’14CoïmbraJune 2014
Summer School EMR’14“Energetic Macroscopic Representation”
« Inversion -based control »
EMR’14, Coimbra, June 201414
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Inversion-based control -
vVSI
vM1
iVSI
VDC
IDC
DCiM1
iM2
vM2
iM1
eM1
iM2
eM2
MM1
SM1
SM2
MM2
TMM1
TSM1
TSM2
TMM2
Ω1
Ω1
Ω2
Ω2
TM1
Ω1
Ω1
Tload-1
Ω2
Ω2TM2
Tload-2
DC bus 5-leg VSIConcordia
transformationequivalentwindings
swapping connection
fictitiousmachines
mechanical coupling shaft loads
iMM1
eMM1
iSM2
iMM2
iSM1
eSM1
eSM2
eMM2We have : 4 DOF (5-leg VSI)4 currents to control
mVSI
Possible
Ideally, SM1 and SM2 have to be canceled by machine design
Impact of SM1 and SM2 if they exist
Torque ripple of TM1 and TM2 has to be compensated by control iMM1 iSM2
*=
iSM1 iMM2=
EMR’14, Coimbra, June 201415
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Inversion-based control -
vVSI
vVSI-ref
vM1
vM1-ref
vM2-ref
iVSI
VDC
IDC
DCiM1
iM2
vM2
iM1
eM1iM2
eM2
MM1
SM1
SM2
MM2
TMM1
TMM2
Ω1
Ω2
TM1
Ω1
Ω1
Tload-1
Ω2
Ω2TM2
Tload-2
iMM1
iMM1-ref
iMM2-ref
iM1-ref
iM2-ref
eMM1
iMM2
eMM2
mVSI
Ω1-ref
Ω2-ref
TM1-ref
TM2-ref
TMM1-ref
TMM2-ref
SM1
SM2
TSM1
TSM2
Ω1
Ω2
iSM2
iSM1
eSM1
eSM2
stratégie
EMR’14CoïmbraJune 2014
Summer School EMR’14“Energetic Macroscopic Representation”
« Experimental results »
EMR’14, Coimbra, June 201417
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Experimental results -
Inverter
dSPACE
PMSM 1
PMSM 2
DC supply
Fig. 1. Laboratory experimental test-bench
EMR’14, Coimbra, June 201418
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Experimental results -
0 2 4 6 8 10 12 14-20
0
20
40
60
80
Time [s]
Mec
hani
cal s
peed
s [r
ad s
-1]
ω*m1
ωm1
ω*m2
ωm2
0 2 4 6 8 10 12-20
0
20
40
60
80
Time [s]
Mec
hani
cal s
peed
s [r
ad s
-1]
ω*m1
ωm1
ω*m2
ωm2
Ω1
Ω2
Ω1Ω2
Ω1ref = cycle
Ω2ref = 0
Ω1ref = cycle
Ω2ref = 40 rad/s
EMR’14, Coimbra, June 201419
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
0 2 4 6 8 10 12 14-20
0
20
40
60
80
Time [s]
Mec
hani
cal s
peed
s [r
ad s
-1]
ω*m1
ωm1
ω*m2
ωm2
- Experimental results -
Ω1
Ω2
Ω1ref = cycle
Ω2ref = cycle
independent behaviours of both machineswith a common supply
EMR’14, Coimbra, June 201420
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Experimental results -
EMR’14CoïmbraJune 2014
Summer School EMR’14“Energetic Macroscopic Representation”
« BIOGRAPHIES AND REFERENCES »
EMR’14, Coimbra, June 201422
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Authors -
Dr. Ngac Ky NGUYENArts et Métiers ParisTech, L2EP, FrancePhD in Electrical Engineering at University of Mulhouse (2010)Associate Professor since 2012Research topics: Multiphase machines control, fault tolerance systems, EMR
Prof. Eric SEMAILArts et Métiers ParisTech, L2EP, FrancePhD in Electrical Engineering at University of Lille 1 – Science and Technology (2000)Full Professor since 2010Research topics: Multiphase machines design and control, multi-converter modelling
Dr. Xavier KESTELYNArts et Métiers ParisTech, L2EP, FrancePhD in Electrical Engineering at University of Lille 1 – Science and Technology (2003)Associate Professor since 2004Research topics: Multiphase machines control, fault tolerance systems, EMR,
modelling and control of multi-physical systems
EMR’14, Coimbra, June 201423
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- References -
[Semail 05] E. Semail, E. Levi, A. Bouscayrol, X. Kestelyn, “Multi-Machine Modelling of Two SeriesConnected 5-phase Synchronous Machines: Effect of Harmonics on Control ”, EuropeanConference on Power Electronics and Applications, pp. 10, 2005.
[Mekri 12] F. Mekri, J-F. Charpentier, E. Semail, “An efficient control of a series connected two-synchronous motor 5-phase with non sinusoidal EMF suppliedby a single 5-leg VSI:Experimental and theoretical investigations”,Electric Power Systems Research 92 (2012) 11-19.
[Gataric 00] S. Gataric, “A polyphase cartesian vector approach to control of polyphase AC machines”,Industry Applications Conference, 2000. Conference Record of the 2000 IEEE, Vol. 3, pp. 1648-1654,Rome 2000.
[Jones 04] M. Jones, E. Levi, A. Iqbal, “A five-phase series-connected two-motor drive with current controlin the rotating frame ”,35th Annual IEEE power Electronics Specialists Conference, Vol. 5, pp. 3278-3284, Aachen, 2004.
[Bouscayrol 00] A. Bouscayrol, B. Davat, B. de Fornel, B. François, J. P. Hautier, F. Meibody-Tabar, M.Pietrzak-David, "Multimachine Multiconverter System: application for electromechanical drives",European Physics Journal - Applied Physics, vol. 10, no. 2, May 2000, pp. 131-147 [IF 0.465](common paper GREEN Nancy, L2EP Lille and LEEI Toulouse, according to the SMM project of theGDR-SDSE)
[Hautier 04] J. P. Hautier, P. J. Barre, "The causal ordering graph – A tool for modeling and control lawsynthesis",Studies in Informatics and Control Journal, December 2004, Vol. 13, no. 4, pp. 265-283.
EMR’14CoïmbraJune 2014
Summer School EMR’14“Energetic Macroscopic Representation”
« Thank you »
EMR’14, Coimbra, June 201425
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Colors -
violetpale
green gold
Powersystem
Powersource
Systemmodel
Systemcontrol
Controlstrategy
blueskyblue
Web X11 colour, standard colours on web pageshttp://en.wikipedia.org/wiki/Web_colors
• light green background RGB = (152,251,152) « pale green »• dark green border RGB = (0,128,0)« greeen »
• orange background RGB = (255,215,0)« gold » • red border RGB = (255,0,0)« red »
• purple background RGB = (238,130,238)« violet » • dark blue border RGB = (0,0,255)« blue »
• light blue background RGB = (135,206,235)« sky blue » • dark blue border RGB = (0,0,255)« blue »
• dark blue background RGB = (0,0,255)« blue » • dark blue border RGB = (0,0,255)« blue »
EMR’14, Coimbra, June 201426
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
element name
Name
No equation number in slides
x1
y1
- EMR pictograms -
element name element name
element name element name
power vectors(size b, full arrows)
signal vectors(size b/2, empty arrows)
(radius= a)
borders of powerelements = b pt
(a/2 x a) (a x a)(2a x a)
a
2a
EMR’14, Coimbra, June 201427
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
stratégie
- Control pictograms -
(support squarea x a)(same pictograms – same size -
with or without oblique bar)
borders of controlelements = b/2 pt
signal vectors(size b/2, empty arrows)
EMR’14, Coimbra, June 201428
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
Source
- estimation pictograms -
borders of estimationelements = b/2 pt
signal vectors(size b/2, empty arrows)
EMR’14, Coimbra, June 201429
« EMR and inversion-based control of 2 series connec ted 5-phase PMSMs »
- Example -
ich1
ubat
ifield
efield
battery choppers
DC machine
ubat
Bat.itot
Env.
uch1
iarm
iarm
earm Tem
Ω gear
Tgear
Ωwh
Fwh
Fres
vev
vev
gearbox
mch1
ifield
uch2
ich2
ubat
parallel connexion wheel
chassis
environment
mch2
ifield-ref
uch1-ref iarm-ref
Tem-ref Tgear-ref Fwh-ref vev-ref
uch2-ref
strategy driver requestSoCest
Ωwh-est vev-est