Polytech’ Paris Sud June 2014 - EMRwebsite - Home · 21 u 21 p: number of downstream power flows m < p objective = distribution of energy into p different power flows y 2p u 1p

Polytech’ Paris SudJune 2014

«« EENERGY NERGY DDISTRIBUTION AND ISTRIBUTION AND

SSTRATEGY TRATEGY »»

Prof. A. Bouscayrol(University Lille1, L2EP, MEGEVH, France)

based on the course of Master“Electrical Engineering & sustainable Development”

University Lille1

2

«« Energy Distribution and Strategy Energy Distribution and Strategy »»

EMR, Paris, June 2014

- Objective: example of HEV control -

BAT

ICE

VSI EM

FuelParallel HEV Trans.

fast subsystemcontrols

EMcontrol

ICEcontrol

Transcontrol

Energy management(supervision/strategy)

driver request

slow systemsupervision

3



- Different control levels (2) -

BAT

ICE

VSI1 EM1

FuelParallel HEV Trans.

fast subsystemcontrols

EM1control

ICEcontrol

Transcontrol


driver request

slow systemsupervision

EMR

Inversion-based control

How to manage energy?

Interactionbetween

both levels?

4



- Outline -

1. Control decomposition• Local control structure• Strategy level for energy management

2. Various inversions of distribution elements• Different kinds of energy management• Inversion of the different cases

3. Analysis and strategy• Analysis of the system possibilities• Strategy level

4. Application to the automatic subway VAL 206


1.1. «« Local controlLocal controland energy management and energy management »»

6



Systemcause effect

- Principle of Inversion-based methodology -

desired effectControl

right cause

measurements?

control = inversion of the causal path

1. Which algorithm? (how many controllers)2. Which variables to measure?3. How to tune controllers?4. How to implement the control?

Inversion-based methodology

automatic controlindustrial electronics

input output

[Hautier 96]

7



- EMR and Inversion-based methodology -

desired effectright cause

measure?

SS1

causeeffect

inputoutputSS2 SSn

C1 C2 Cn

measure?measure?

EMR = system decomposition in basic energetic subsystems (SSs)

Remember,divide and conquer!

Inversion-based control: systematic inversion of each subsystems usingopen-loop or closed-loop control

8



- Mirror effect -

desired effectright cause

SS1

causeeffect

SS2 SSn

C1 C2 Cn

The control scheme is developed as a mirror of the model

[Hautier 96] [Bouscayrol 03]

9



- Inversion of EMR elements -

coupling element distribution criteria

conversion element direct inversion +disturbance rejection

accumulation elementcontroller +

disturbance rejection

Legend

Control = light blue Parallelogramswith dark bluecontour

directinversion

indirectinversion

sensor

mandatory link

facultative link

10



2. Tuning path

y3 y4 y5 y6y1S1

y2 y7z23 z67

S2x1 x2 x3 x4 x5 x6 x7

1. EMR of the system

- Maximum control scheme -

x7-refx6-refx5-refx4-refx3-ref

3. Inversion step-by-step Strong assumption: all variables can be measured!

4. Simplification of control

5. Estimation of non-measured variables

6. Tuning of controllers 7. Strategystrategy

11



v

DCM1

DCM2

EP1

EP2

Bat. Fres

ME1

ME2

EnvBat

Strategy

- Management of the energy distribution -

12



- Different control levels -

Energy management of HEVs:Energy management of local subsystemsEnergy management of the whole system (co-ordination of subsystems)

Two control levels can be organized:- local control- system supervision

Dynamic and causal modelsQuasi-static

models

compatibilityin term of

inputs/outputs

compatibilityof the

control levels[Delarue & al 2005]

13



- Articulation of control levels -

Local control:Fast energy management in each subsystemObjective: ensure the best efficiency of each component in function

of the requestInversion of each element, step-step-step, using dynamic causal models

Energy management:Coordination of energy between all subsystemsObjective: ensure the best energy distribution in function of the global

demandGlobal view for decision using quasi-static or static modelsCould be considered as an inversion of a global model of the system

14



- Articulation of control & models -

BAT

ICE

VSI EM

Fuel Trans.

EMcontrol

ICEcontrol

Transcontrol


driver request

Strategy delivers referencesfor local controls

Inversion of causaldynamical models

Inversion of a globalstatic model

15



r-est

is-dq-est

es-dq-est

iim

isd

gear

Tim

is-dq

es-dqis-dq

vs-dquinv

rd

d/s

Park’s transformation

stator windings

EMconversion

d/s_est

uinv-refTim-refvs-dq-ref is-dq-ref

isd-refr-ref

- From dynamical model of IM drive… -

VDC

ivsi

inverter

Global behaviorof the drive

(machine + inverter + control)

16



gear

Tim

- … to static and dynamical models -

VDC

ivsi

drive

Tim-ref

0 2000 4000 6000 8000-150

-100

-50

0

50

100

150

49

49 49

49

49

64

6464 6477 7777 77

86

86

86

86

9292

9292

96

96

9696

9696

96

9898

9898

98

Tim (Nm)

gear (rad/s)

Efficiency map

x+-x gear

TimVDC

ivsi

Tim-ref

staticmodel

17



gear

Tim

- Coherence of dynamical and staticmodel -

VDC

ivsi

Tim-ref

gear

Tim

gear-ref

VDC

ivsi

gear

Tim

VDC

ivsi

backward

Static model: no dynamical effect, I/O could be changed

What happens if EM = inversion of a static modeland IBC = inversion of dynamical modelswith differents I/O?

Bad articulation between control levels

gear

TimVDC

ivsi

Tim-ref

static Dynamic

18



- Energy management of HEVs -

[Salmasi 2007]


driver request

Rule-based

deterministic rule-based

fuzzyrule-based

state machine / power follower/ thermostat control…

predictive / adpative /conventional…

Optimization based

global optimization

real-time optimization

Dynamic programming / stochastic DP /Game theory / Optimal control….

Robust control / Model predictive / decoupling control / l control…


2. 2. «« Inversion of variousInversion of variouscoupling elements coupling elements »»

20



- Upstream coupling elements -

upstream common part

u1

y1

m: number of upstreampower flows

y21u21

p: number of downstreampower flows

m < pobjective = distribution of energy

into p different power flows

y2pu1p

Example: differential

diffldif rdif

rwh lwhwh

TT T2

2

diff

Tgear

lwh

rwh

Tldiff

Trdiff

m=1, p=2 (m<p)objective = distribution of torque

Tldiff

rwh

Trdiff

lwhTgear

wh

21



- Downstream coupling elements -

downstream common part

y2

u2


y21u21


m > pobjective = collect energy

into m different power flows

y2pu1m

Example: chassis of a train

m=4, p=1 (m>p)objective = collecting bogie forces

4 traction bogies

Fbog1

vtrain

Ftot

Fbog4

Fbog2

Fbog3

vtrain

vtrain

vtrain

vtrain

22



- Neutral coupling elements -

u21y21



m = pobjective = reorganize

power flows

u21y21

Example: Park’s transformation

m=2, p=2 (m=p)objective = convert AC voltages and currents

into DC voltages and currents

u13is1

u23is2

d/s

vsdisd

vsqisq

y11

u11

y1m

u1m

23



- Inversion of neutral coupling elements -

u21y21

u2py2p

y11

u11

y1m

u1m

u1m y2p-ref

no measurementno controller / direct inversion

y21-refu11

Example: Park’s transformation

u13is1

u23is2

d/s

vsdisd

vsqisq

vsq-ref

vsd-refu13-ref

d/s

u23-ref

use of the inverse matrix

24



- Inversion of upstream coupling elements -

u1

y1

y21u21

no measurement no controller

(p - m) weighting variables

y2pu2p

u1

y2p-ref

y21-ref

kW1…kW(p-m)

refp'Wiref

'W y)k(...yku 22111 1

Implement a compromise or

prioritize outputs.

Tgear

Tldiff

rwh

Trdiff

lwhTgear

wh

21 )1(2 refdiffWrefdiffWgear TkTkT

Tdiff-ref1

kw

Tdiff-ref2


25



u1m

u11

- Inversion of downstream coupling elements (1) -

Case 1: all inputs are used

y2

u2

y2-ref

kD1…kD(m-p)

no measurement no controller

(m - p) distributionvariables

ref'

pDmm

ref'D

yku...

yku

21

2111

y11

u11

y1m

u1m

Example: chassis of a train

Fbog1

vtrain

Ftot

Fbog4

Fbog2

Fbog3

vtrain

vtrain

vtrain

vtrain

Ftot-ref

Fbog1

Fbog4

Fbog2

Fbog3

kD1 kD3kD2


3. 3. «« Analysis of the system possibilities Analysis of the system possibilities and strategy level and strategy level »»

27



- Critical factors -

nO = number of objectives

nC = number of constraints(in order to achievethe objectives)

nT = number of tuning variables(in order to act onthe system)

s13s12s11

s23s22s21

Example: 3-leg Voltage-Source-Inverter

VDC

ivsi

u13

u23

i1

i2

nO = 2 (u13 and u23)

nC = 3 (3 commutation cells)

1,2,3ifor 21 ii SS

nT = 6 (6 switching orders Sij)

28



- Ideal case -

Example: 3-phase induction machine

They are enough tuning inputsto achieve the objective and

realize the constraints

COT nnn

isdgear

Tim

is-dq

es-dqis-dq

vs-dq

iim

uinv

rd

d/s

Park’s transformation

stator windings

EMconversion

nO = 1 (Tim)nC = 1 (flux r)

nT = 2 (u13 and u23)

COT nnn

Field-oriented control is a way to impose flux and torque by voltages (using a decoupling (d,q) frame)

29



- Optimisation case -

s13s12s11

s23s22s21

Example: 3-leg Voltage-Source-Inverter

VDC

ivsi

u13

u23

i1

i2

nO = 2 (u13 and u23)nC = 3 (3 commutation cells)nT = 6 (6 switching orders Sij)

COT nnn

They are more tuning inputsthan the objective and

the constraints

• all tuning inputs must be used• the degrees of freedom can beused for Optimisation

• a strategy level is required todefine the degrees of freedom

Various optimization strategies for VSI suchas 3rd harmonic injection, flat-top modulation…

30



- Compromise case -


nO = 2 (Tdiff1 and Tdiff1)nC = 0 nT = 1 (Tgear)

COT nnn

They are not enough tuning inputsthan the objective and

the constraints

• a compromise must be defined(priorities)

• a strategy level is required todefine the priority

Tgear Tdiff-ref1

kW

Tdiff-ref2

Ex: kW=1 priority to Tdiff1

Tldiff

rwh

Trdiff

lwhTgear

wh

31



- Coupling elements and strategy -


All downstream and upstreamCoupling elements

Requires criteria(distribution or weighting)

• these criteria variablesmust be defined by the strategylevel

Tgear Tdiff-ref1

kW

Tdiff-ref2

Strategy

Ex: during straight lines, Tgear = average value of both resquets

Tldiff

rwh

Trdiff

lwhTgear

wh

32



- Strategy level -

Strategy level:global management of the whole systemHas to define:• local control references• distribution coefficients• weighting coefficients• optimization inputs• compromises

For energetic system:definition of priority between performances and efficiency

Inversion of a global static model(different possibilities)

Example: Flux weakening in EVs

rated

Umax

U

U depends on the battery voltage Vbat depends on velocity vEV

ref

vEV-measVbat-meas

static plots

global view


4. 4. «« Application to the controlApplication to the controlof an automatic subway of an automatic subway »»

part II: real systempart II: real system


based on PhD of J. N. Verhillein collaboration with Siemens Transportation Systems

34



ifield1

iarm

etot

iarm

iarm

earm1bog1

Tmach1

MCC in series bogies

Seq1

Seq2

uchop2

mchop2

ufilter

parallel

ufilter

ufilter

choppers

itotufilter

ufilterifilter

filter

ifilter

rail

VDC

ES

- EMR of electrical part -

35



iarm

etot

iarm

iarm

earm1bog1

Tmach1


Seq1

Seq2

uchop2

mchop2

ufilter

parallel

ufilter

ufilter

choppers

itot

ifield1

ufilter

ufilterifilter

filter

ifilter

rail

VDC

ES

ifield1-refTmach1-ref

ifield2-ref

Tmach2-refiarm-ref2iarm-ref1iarm-ref

uchop1-ref

uchop2-ref

uchop3-ref

ufilter-mes

iarm-ref=kW iarm-ref1+(1-kW) iarm-ref2

- Control of electrical part -

36



iarm

etot

iarm

iarm

earm1bog1

Tmach1


Seq1

Seq2

uchop2

mchop2

ufilter

parallel

ufilter

ufilter

choppers

itot

ifield1

ufilter

ufilterifilter

filter

ifilter

rail

VDC

ES


ifield2-ref


uchop1-ref

uchop2-ref

uchop3-ref

ufilter-mes


Strategyhas to definethe 3 criteria

37



bog1

Tmach1


Seq1

Seq2

parallel choppersfilterrail

VDC

ES


ifield2-ref


uchop1-ref

uchop2-ref

uchop3-ref

ufilter-mes


Tmach2-ref

uchop2-ref

iarm-ref2

Actual control: no possibility of independent torque

38



non-linear

devicesmotor

MOTEUR

TRANSMISSION

1/2 ESSIEU

1/2 ESSIEU

REDUCTEUR

PNEU

REDUCTEURPNEU

PONT + DIFFERENTIEL

backslash

Tred

slip

F

- EMR of mechanical part -

vVAL

bk1

motor

gear1

m

Tmach1

DCM

m Tdif

Tgear dif

bk1 Tgear1Tdif1 gear1

SM

vVAL

Fres

Twh1 Fcont1

vwh1 Fcont1

vVAL Fbog1

differential brake gearbox wheels contact chassis environ.

gear2

bk2

bk2

gear2

Tgear2 Troue2 Fcont2

vwh2Fcont2

vVAL

Tdif2

Tgear1

Tgear2

39



- Control of mechanical part -

DCM1

DCM2

40



Tmach1ref

vVAL-ref

Tmach2ref


DCM1

DCM2

41



Tmach1ref

vrame-ref

Tmcc2-ref Master-slave control


DCM1

DCM2

42



Tmach1ref

vrame-ref

Merging of control blocs

vrame-estarb-mes

- Actual control of mechanical part -

Tmach1ref Actual control:no local management of anti-slip

DCM1

DCM2

43



- Simulation model -

Actual control

Matlab/SimulinkTM electrical part mechanical part

44



armature current (A)field current

(A)

experimentalsimulation


- Validation model (1) -

45




velocity (m/s)

Test at normal operating

- Validation model (2) -

0 4 8 120

4

8

12

16

velocity (m/s)

Test at maximal torque



4. 4. «« Application to the controlApplication to the controlof an automatic subway of an automatic subway »»

part III: antipart III: anti--slip strategyslip strategy


based on PhD of J. N. Verhillein collaboration with Siemens Transportation Systems

47



VAL Automatic Subway:Lille, Paris airport, Chicago, Taïpeh...

Actual controlVAL 206

• identical torque for 4 motors• global anti-slip (cancellation of torque)

vrefTref

4-mes

Independent controls with same power structure?Performances with local anti-slip?

vref

T1-ref 1-mes T2-ref 2-mes T3-ref 3-mes T4-ref 4-mes

?

STSSTS

- VAL 206 and control -

48



vsub_refFtot1_ref

Fbog_ref

strategywh22_meswh21_meswh12_mes

wh11_mes

Tdcm_ref

vsub_mesTdcm_ref Actual control:slip detection

Torque set to zero

vsub_refFtot1_ref

Fbog2_ref

kD

wh2_ref

shaft2_refTdcm2_ref

kD2kW2

wh1_ref

kD1kW1

Fbog1_ref

newstrategy

wh22_meswh21_meswh12_meswh11_mes

shaft1_mes

Tdcm1_ref

shaft2_mes

vsub_mes

shaft1_ref

New strategy:slip detection

Reduction of torqueof the slipping wheel

Increase of othertorques

- Anti-slip control -

49



- Philosophy of the strategy -

kD

kD2kW2

kD1kW1

newstrategy

wh22_meswh21_meswh12_meswh11_mes

Normal operation:equal distribution of traction forces

21

2121 WWDDD kkkkk

iBogtot

totBogBogBogBog

FF

FFFFF41

4321

Slipping on bogie 1:cancelation of Fbog1 and newdistribution of other forces

01

21

11

22

WD

WD

D

kkkk

k

iBogtot

totBogBogBog

Bog

FF

FFFF

F

31

0

432

1

50



- Simulation results -

vsub_refvsub

vsub_refvsub

actual control New anti-slip strategy

Loss of adhesion of wheel no. 1

Ftot (Nm)traction force (Nm)traction force (Nm)

velocity (m/s) velocity (m/s)

51



mechanical power traintraction drive

drive control

controlpower

power traincontrol

vev-refTim-ref

Tim

Tim

model

mechanical power train

AC load drive

drive control

ref= mod

- HIL simulation of the VAL traction system -

52



velocities (m/s)

vsub

vsub_ref

current (A)

iarm

- Experimental results -

current (A)ifield2

ifield1

Loss of adhesion on wheel 1Fbog1 is decreased by acting on ifield1Fbog2 is increased by acting on iarm

53



- Conclusion -

• Energetic Macroscopic Representation of VAL traction• Inversion-based control from EMR• Best repartition of traction effort• Best performances during slipping phenomena

• real-time validation of the control• dual drive system• to be implemented on the actual system

HIL simulation

New anti-slip control

Polytech’ Paris SudJune 2014 «« Conclusion Conclusion »»

Inversion based control = inversion of EMRinversion of dynamic causal modelsa unique Maximum Control Schemes

Strategy Level = inversion of a global static modelglobal model and local dynamic models must have same I/Odifferent strategies are possible

Distribution elementsHighlight degree of freedom, to be defined in the strategy level

Efficient energy management of EVs and HEVsWell-defined local controls of subsystems

coherent articulation between control levelsefficient strategy level

55



- References -

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, p. 131-147 (common paper GREEN Nancy, L2EP Lille and LEEI Toulouse, according to the SMM project of the GDR-SDSE).

P. Delarue, A. Bouscayrol, A. Tounzi, X. Guillaud, G. Lancigu, “Modelling, control and simulation of an overall wind energy conversion system”, Renewable Energy, July 2003, vol. 28, no. 8, p. 1159-1324 (common paper L2EP Lille and Jeumont SA).

A. Bouscayrol, M. Pietrzak-David, P. Delarue, R. Peña-Eguiluz, P. E. Vidal, X. Kestelyn, “Weighted control of traction drives with parallel-connected AC machines”, IEEE Transactions on Industrial Electronics, December 2006, vol. 53, no. 6, p. 1799-1806 (common paper of L2EP Lille and LEEI Toulouse).

J. N. Verhille, A. Bouscayrol, P. J. Barre, J. C. Mercieca, J. P. Hautier, E. Semail, "Torque tracking strategy for anti-slip control in railway traction systems with common supplies", IEEE-IAS’04, proceeding vol. 4. pp. 2738-2745, Seattle (USA), October 2004 (common paper L2EP Lille and Siemens Transportation Systems)..

J. N. Verhille, A. Bouscayrol, P. J. Barre, J. P. Hautier, “Validation of anti-slip control for a subway traction system using Hardware-In-the-Loop simulation”, IEEE-VPPC’07, Arlington (USA), September 2007 (common paper L2EP Lille and Siemens Transportation System)

L. Horrein, V. Derache, A. Bouscayrol, J. N.Verhille, P. Delarue, “Different models of the traction system of an automatic subway”, ElectrIMAC’11, Cergy Pointoise (France), June 2011 (common paper of L2EP and Siemens Transportation Systems).

Polytech’ Paris Sud June 2014 - EMRwebsite - Home · 21 u 21 p: number of downstream power flows m < p objective = distribution of energy into p different power flows y 2p u 1p

Documents