Roy Davis Ballard

8/3/2019 Roy Davis Ballard

1/19

Ballard Pow er Systems

Using Simplorer and Maxwell Tools to ValidateElectric Drives for EV Applications, and Interactions

w ith Rapid Prototyping Tools

Roy I. Davis

Ballard Power Systems

Dearborn, Michigan

Automotive Electromechanical Simulation Workshop

9 October 2003


2/19

18 December 2003 2

Electric Drives for EV Applications - Outline

n Brief Overview of Ballard

n Electric Drives Design Process

n How Ansoft Tools Fit In at Ballard

n PM Motor Design Evaluation

n Control Design Motor Designn Inverter Design Custom Power Module Design

n Rapid Prototyping Control Design

n Testing Motor Design Evaluation


3/19

18 December 2003 3

Electric Drives Design Process

Inverter

Controller

Motor TransmissionDC Supply LoadVDC V AC MM M T

IDC IAC NM NT

Customer Specification

Motor Requirements

Motor Mechanical Design

Motor Thermal Design

Inverter Requirements

Motor Electromagnetic Design

Cost EvaluationVolume EstimateMass Estimate

TransmissionAssumptions

Transmission Requirements

Custom Power Module DesignInverter Design

Performance Evaluation


4/19

18 December 2003 4

Inverter

Controller

Motor TransmissionDC Supply LoadVDC V AC MM M T

IDC IAC NM NT

Customer Specification

Motor Requirements

Motor Mechanical Design

Motor Thermal Design

Inverter Requirements

Motor Electromagnetic Design

Cost EvaluationVolume EstimateMass Estimate

TransmissionAssumptions

Transmission Requirements

Custom Power Module DesignInverter Design

Performance Evaluation

How Ansoft Tools Fit in at Ballard

Simp lo re r

RMxpr t

M ax w e ll 2 D

M ax w e l l The r m a l

M ax w e ll 3 D

S imp lo re r


5/19

18 December 2003 5

PM Motor Design Evaluation

n Maxwell 2D used to perform analysis

Calculation of Average Torque and Ripple Torque

Average value and peak-to-peak magnitude match well between test andFEA calculations.

70

75

80

85

90

95

-0.10 -0.08 -0.06 -0.04 -0.02 0.00 0.02 0.04 0.06 0.08

Time, seconds

Torque,

Nm

Maxwell 2D Torque vs. Time

Time, seconds

Torq

ue,

Nm

Test Data (2.8 revolutions)FEA Results (3.7 revs)


6/19

18 December 2003 6



Calculation of Back EMF

Harmonic content and magnitude matches well with experiments.

Test Data

-40

-30

-20

-10

0

10

20

30

40

-0.010 -0.008 -0.006 -0.004 -0.002 0.000 0.002 0.004 0.006 0.008

Time (s)

Back-E

MF(V)

FEA Results

Equal time and voltage scales


7/19

18 December 2003 7


n RMxprt used to enter motor design information


Calculation of Ld and Lq (Energy or Vector Potential Method)

Flux Linkage Method:Ld versus Id

0.0E+00

5.0E-05

1.0E-04

1.5E-04

2.0E-04

2.5E-04

0 100 200 300 400 500 600

Id(A)

Ld(H)

Ld (25 Turns) FEA Ld Test

Flux Method: Lq versus Iq

0.00E+00

5.00E-05

1.00E-04

1.50E-04

2.00E-04

2.50E-04

3.00E-04

3.50E-04

4.00E-04

4.50E-04

0 100 200 300 400 500 600 700

Iq (A )

Lq(H)

Lq (25 Turns) FEA Lq,Test


8/19

18 December 2003 8

Control Design Motor Design - General

n Motor equivalent circuit parameters with saturation

Needed for control development

Use Maxwell 2D to update estimates made earlier in design process

Used to populate Simplorer model of PM motor

n Ansoft tools are used to explore and understand:

PTPA (selection of id/iq trajectories) Maxwell & Simplorer Field weakening control Maxwell & Simplorer

Interactions of control with dc bus variations due to load pointchanges (dynamic) - Simplorer

Interactions of sensors and interface limitations on controlperformance Simplorer

Etcetera

n Matlab/Simulink used as primary control development tool withautocode function for dSpace rapid prototyping hardware.


9/19

18 December 2003 9

Control Design Motor Design - Simplorer

GND

MS3 ~BA

C

Viscous_Load

+

Omega_GND

+

Omega_Motor

Ballard_IPM

EQUBL

EQUBL

EQUBL

+

V

+

V

+

V

vdr

vqr

Rotor_Speed_MechVq_feedfwd

Vd_feedfwd

Vq_feedback

Vd_feedback

Q_reference

Q_feedback

D_reference

D_feedback

A_voltage_reference

B_voltage_reference

C_voltage_reference

Theta_Rotor_Mech

D_Q_Controller_Rotor_Frame

GAIN

Norm_to_DC_Bus_A

GAIN

Norm_to_DC_Bus_B

GAIN

Norm_to_DC_Bus_C

OMEGA

SINE

COSINE GAIN

NEG

P_over_2

IQ_Rotor

ID_Rotor

CONST

IQ_R_Ref

CONST

ID_R_Ref

+

V

EQUBL

Reciprocate

Lin_IPM_25kW.IQ_Rotor.VALIQ_R_Ref.VALLin_IPM_25kW.I1Lin_IPM_25kW.I1ID_R_Ref.VALID_Rotor.VALLin_IPM_25kW.I1IA_Filter.VAL

t [s]

0.4k

-0.3k

0

-0.2k

-0.1k

0.1k

0.2k

0.3k

0 5.6m1m 2m 3m 4m

A

A

Transform_ab_alphabeta

IA_Meas

IB_Meas

PHIMLOAD

LIMIT

LIMIT

LIMIT

Ibeta

theta_electrical

Ialpha

V1D

V1Q

GAIN

Half

I1D

I1Q

FA

FB

Falpha

Fbeta


10/19


11/19

18 December 2003 11

Control Design Motor Design Co-simulate

Kp

Ki R

1

S

1

S

1

L q

eL d

K p

K i

R

1

S

1

S

1

Ld

eL q

P M S M p h y s i c a l m o d e l

eL d

eL q

R

R

C ross-

Coup l ingI n p u t

Decoup l ing

Res is tance

Decoupl ing

Res is tanceDecoupl ing

e

pm e pm

B a c k - E M FDecoupl ing

S t a t e F e e d b a c kCur ren t Regu la to r

^

^

^

^

^

iq

*

id

*

-

-

--

+

+ +

+

+

++

++

++

+++

+-

-

iq

iq

id

id

v d

vq

Current Regulator/Motor Block Diagram Simulink Current Regulator


12/19

18 December 2003 12


n Space Vector Modulation (SVM) with Overmodulation Modes 1-2and Deadtime Compensation

Simulink Model of Modulation Systems


13/19

18 December 2003 13


GND

PHIDEGMS3 ~B

A

C

Viscous_Load+

Omega_GN

+

Omega_Moto

Ballard_IPM

EQUBL

EQUBL

EQUBL

+

V

+

V+

V

OMEGA

+

V

V_DC

IQ_Rotor.VALID_Rotor.VALLin_IPM_25kW.Viscous_Load.TLin_IPM_25kW.Lin_IPM_25kW.I-1 * Lin_IPM_25

t [s]

60

-0.12k

0

-0.1k

-80

-60

-40

-20

20

40

0 23m2.5m 5m 7.5m 10m 13m 15m 18m 20m

A

A

IA_Meas

IB_Meas

PHIMLOAD

LIMIT

LIMIT

LIMIT

Ballard Powe r Syste ms

Roy Davis

Simulation of IPM Drive

7 April 2003

SiM2SiMSIMPLORER Link Interface

SiM2SiM

SIM2SIM1

EQU

FML1

PHI_MOD:=MOD(Lin_IPM_25kW.PHIDEG,36

CONST

CONST1

CONST

CONST2

CONST

CONST3


14/19

18 December 2003 14


GND

PHIDEGMS3 ~B

A

C

Viscous_Load+

Omega_GN

+

Omega_Moto

Ballard_IPM

EQUBL

EQUBL

EQUBL

+

V

+

V+

V

OMEGA

+

V

V_DC


t [s]

60

-0.12k

0

-0.1k

-80

-60

-40

-20

20

40

0 23m2.5m 5m 7.5m 10m 13m 15m 18m 20m

A

A

IA_Meas

IB_Meas

PHIMLOAD

LIMIT

LIMIT

LIMIT


Roy Davis


7 April 2003


SiM2SiM

SIM2SIM1

EQU

FML1


CONST

CONST1

CONST

CONST2

CONST

CONST3


15/19

18 December 2003 15


GND

PHIDEGMS3 ~B

A

C

Viscous_Load+

Omega_GN

+

Omega_Moto

Ballard_IPM

EQUBL

EQUBL

EQUBL

+

V

+

V+

V

OMEGA

+

V

V_DC


t [s]

60

-0.12k

0

-0.1k

-80

-60

-40

-20

20

40

0 23m2.5m 5m 7.5m 10m 13m 15m 18m 20m

A

A

IA_Meas

IB_Meas

PHIMLOAD

LIMIT

LIMIT

LIMIT


Roy Davis


7 April 2003


SiM2SiM

SIM2SIM1

EQU

FML1


CONST

CONST1

CONST

CONST2

CONST

CONST3


16/19

18 December 2003 16

Inverter Design Custom Pow er Module Design

n Use Maxwell 3D to model power module layout Objective is to compute parasitic module inductance to support

module design layout decisions

2nd objective is to provide inductances for bridge circuit simulation

Presently developing this process difficult to get accurate resultsUppe

Lowe

E1

E2 S2

S14.7

3.3EMF := 17

EMF := 5

Switching

NEG

E3

-5

I1

Lstray

E4

n Also developing process for parameterizing the detailed modelsof IGBTs in Simplorer using this test circuit.


17/19

18 December 2003 17

Inverter Design Custom Pow er Module Design

n Simplorer model of inverter, dc source, and motor used to:

Evaluate component stresses under realistic conditions

Switches

Diodes

Capacitors

Investigate effects of different/variable PWM algorithms Determine losses


18/19

18 December 2003 18

Rapid Prototyping Control Design

n We use dSpace for rapid prototyping control design and motorperformance evaluations on the dyno

Clean interface to Simulink/RTW with Control Desk

n Simplorer model of motor and inverter easily integrated withSimulink model of control

Helps to debug any problems arising from interactions with inverter

PM motor model easily allows use of saturable inductances asfunctions of current

Easy to do what-if investigations

Much easier and seemingly more reliable than Matlab Power System

Toolbox to include detail model of inverter and PM motor.


19/19

18 December 2003 19

Testing Motor Design Evaluation

n Unexplained or different behavior that occurs during testing canbe examined in Maxwell or Simplorer to determine cause.

n Provides a feedback link to design tool usage that allowsrefinement of rules of thumb or development of design rules tobe applied with the numerical design tools.

Documents

Roy Davis Ballard