Ultra-Capacitors in Power Conversion -Applications, Analysis and Design from Theory to Practice- Dr. Petar J. Grbović HUAWEI Technologies Duesseldorf GmbH

Ultra-Capacitors in Power Conversion

-Applications, Analysis and Design from Theory to Practice-

Dr. Petar J. GrbovićHUAWEI Technologies Duesseldorf GmbH

Energy Competence Center Europe

Agenda

• PART ONE: Background of Power Conversion & Energy Storage

• PART TWO: Ultra-capacitor Energy Storage Device• PART THREE: Applications• PART FOUR: The Ultra-Capacitor Selection & Design• PART FIVE: Interface Power Converter

2

PART ONEBackground of Power Conversion & Energy

Storage 1. The need for Energy Storage2. Energy Storage Technologies & Devices3. Electrochemical Batteries4. Flywheel Energy Storage5. Ultra-capacitor Energy Storage6. Ultra-capacitors versus Batteries

3

• Electric systems A & B are interconnected via a power processor1. A =the grid & B=electric drive2. A= wind mill & B=the grid3. A=the grid & B= a critical load (data center, hospital, etc., etc,.)

• The power processor: a static power converter, transformer, installation, etc., etc.

• The power processor has no energy storage capability Instantaneous power of both systems are equal.

• The system is oversized• Efficiency is reduced• Peak power penalty• Interruption cost and penalty

tPtP BA

The Need for Energy Storage

4

AVPAVP

tPtP

BA

BA

• An energy storage device integrated within the power processor

• The energy storage device decouples the system A from the system B

• Instantaneous power• Average power of the systems A

& B remains the same

The Need for Energy Storage

2. What is an energy storage device that can be used in power conversion applications?

1. A power processor with energy storage capability

5

Energy Storage Technologies & Devices • Electric energy storage device is a device with ability to store electric energy

and keep it stored for undefined period of time1. Direct Energy Storage (Simple electric devices), or2. (Indirect Energy Storage) Complex electro-mechanical or electro-chemical

devices

Direct Energy Storage

Indirect Energy Storage

Magnetic Field

Electric Field Mechanical

ChemicalInductors Capacitors Kinetic Potential

SMES Ultra-capacitors

Flywheels HydroPumped

Compressed air

Batteries Fuel Cells

6

Energy Storage Technologies & Devices

• CAES & HPES• Large scale utility applications

• SMES• High power short term utility

applications• Electrochemical Batteries

• Long term, low & midium power applications

• Flaywheels & Ultra-capacitors• Short term, low & midium

power applications

7

Electrochemical Batteries• Secondary electrochemical batteries

1. Convert electric energy into chemical energy (charging),2. Store chemical energy, and3. Convert chemical energy into electric energy (discharging)

• The most popular energy storage devices• Composed of:

1. Two electrodes of different material, and2. Electrolyte

• Electro-chemical action is a slow process • Charge and discharge rate are limited

• Charge/Discharge power is limited• Life time and deep discharge cycling capability are limited

8

Electrochemical Batteries

Energy Density[Wh/kg]

Power Density[W/kg]

Life Time[Cycles]

Lead-Acid 20-35 25 200-2000

Lithium -Ion 100-200 360 500-2000

Lithium Polymer 200 250-1000 >1200

Nickel Cadmium 40-60 140-180 500-2000

Nickel-Metal Hydride

60-80 220 <3000

Sodium-Sulfur 120 120 2000

Zebra Batteries 25 80-150 <16 000

State of the art electro-chemical batteries

9

Ultra-capacitor Energy Storage

• An ultra-capacitor is a special kind of electrostatic capacitor• It is not an electrochemical battery! • Energy is stored directly as electric field between two charged plates

1. Capacitance of hundreds up to thousands of [F]2. The cell voltage 2.5 [V] 3. High energy density, 1 to 10 [Wh/kg] (>>> electrolytic capacitors )4. High power density 5 to 20 [kW/kg] 5. Fast charge/discharge6. High cycling capability <500 000

[cycles]

2002

1CuCE

7. Wide range of operating temperature, from – 40 [ºC] up to +60 [ºC]

www.ultracapacitor.co.kr.

10

Flywheel Energy Storage

• An electric motor/generator coupled with a large inertia rotor• Electric energy converted into mechanical and stored as kinetic

energy of a rotating mass (speed ω0 and inertia J)• Low speed (<10 000rpm) and high (>100 000rpm)

1. High energy density 2. Fast charge/discharge , high peak power3. Long life, high cycling capability4. Wide range of operating temperature

202

1 JE

• Similar characteristics as ultra-capacitors

• Efficiency of the motor/generator is a limiting factor

11

Comparison Power Density

Energy Density

Safety

Price per kWPrice per kWh

Cyclability

Operating Temperature

0

5

10

UC

Lead Acid

NiMH

Li-ion

Fly Wheel

12

Ultra-capacitors versus Batteries

If the charge/discharge time is shorter than TCR

A. An ultra-capacitor is the solution,B. Otherwise an electro-chemical

battery is the solution

• Currently TCR =10 to 30s

• Size and cost of an energy storage1. Energy Storage Capability2. Power Capability & Conversion Efficiency3. Cycling capability & Life time

13

PART TWOUltra-capacitor Energy Storage Device

1. A bit of history of ultra-capacitors2. How does ultra-capacitor work3. Material4. Technologies overview5. Advantages and disadvantages6. Ultra-capacitor macro model7. Charge and discharge Methods8. Frequency dependent losses and thermal aspects9. Integration of the ultra-capacitor into power conversion

system10. Ultra-capacitors today and tomorrow

14

Ultra-capacitor History• The double layer capacitor effect was described by Helmholtz in 1879• Almost a century after that, a first ultra-capacitor was patented by Standard

Oil Company in 1966• A decade after NEC developed and commercialized this device in 1978• The first high power ultra-capacitor was developed for military applications by

the Pinnacle Research Institute in 1982• Ten years after, in 1992, the Maxwell Laboratory had started development of

DoE ultra-capacitors for hybrid electric vehicles• Today, the ultra-capacitors are commercially available from numerous

manufacturers

15

How does ultra-capacitor work?

Ultra-capacitor is an electrochemical capacitor composed of:1. Two current collectors2. Two electrodes made of porous conducting material (activated carbon…)3. A separator, and 4. Electrolyte• The electrodes are separated by the separator and immersed in the electrolyte

One layer virtual capacitor

One layer virtual capacitor

16

• Each layer forms a capacitor The dielectric is a layer of ions. Relative

permeability εR~10 The dielectric thickness is d, diameter of the ions,

approximately d~10 Å Operating voltage is 1-3V, being limited by decomposition

voltage of the electrolyte The surface A is contact surface of the porous electrode,

theoretically A~2000m2/g for activated carbon electrode

Theoretically, specific capacitance C~125F/g @ 2.5V 390kJ/kg Overestimated capacitance, in reality it is 6 to 12F/g (5-10%)

• Electrode is conducting and very porous material• Positive and negative ions form a layer attached to the electrode surface

d

AC R02


17

1. Simple double layer model is not accurate • Very first work, Helmholtz in 1853

• A layer of electrolyte molecules attached to the electrode• Overestimated capacitance, no voltage dependency

• Gouy and Chapman, 1910 and few years later • Considered a space distribution of

the charge• Stern, 1924

• Real dimension of solvent molecules• The space charge in two layers compact

layer and diffused layer

d

AC R

kT

qzch

kT

nqAzC M

2

2' 0

2

kT

qzch

kT

nqAzC D

D 2

2' 0

2


18

2. The porous electrode contact surface is overestimated

• Theoretically A~2000m2/g for activated carbon electrode

• Practically, 10 to 20 % of the theoretical one• The electrode pores are not uniform

• Large pores• Medium, and • Small pores

• Ions do not penetrate in medium and small pores.• The effective surface is much smaller than the

theoretical oneZoomed in pore of porous activated carbon electrode. The ions penetrate only in large pores


19

Ultra-capacitor Material • Current collector

a) Metal foil (Al, etc., etc.)• Electrode

a) Carbon (Activated carbon, Carbon nanotubes, Fibers)

b) Metal oxides c) Polymers

• Electrolytea) Organic (Operating voltage ~2.8V, Good energy density, High series resistance,

Low power density, External balancing circuit is required)

b) Aqueous (Operating voltage ~1V, Low series resistance, Good power density, Good voltage balancing even without external circuit)

• Separatora) Organic (Polymer, Paper)

20

Technologies Overview

21

Advanatges & Disadvanatges

Advantages

1. High specific power >10 [kW/kg]2. Low internal resistance3. High output power4. High cycling capability >500 000 [cycles]5. No chemical action6. Long life ~20 years7. Low cost per cycle8. Very high rate of charge/discharge 9. Improved safety10. Simple charge/discharge method

Disadvantages

1. Low specific energy <10 [Wh/kg] for standard ultra-capacitors

2. Voltage varies with state of charge Need for interface converter

3. High self-discharge rate4. Low cell voltage. Need for series

connection of cells into a module5. Low internal resistance may create an

issue in case of short circuit

22

• Macro (Electrical) model• The control system analysis and

synthesis, and • Losses calculation and thermal design

• Two main effects1. Voltage dependent capacitance

2. The charge time/space distribution due to porosity of the electrodes

• Nonlinear transmission line • Nth order RLCG ladder network as an

approximation• The inductance L and shunt

conductance G (leakage ) neglected

0CufC

Ultra-capacitor Macro Model

23

• Nonlinear distributed network• Voltage dependent capacitance and resistance• Linearization around an operating point uC0=UC0

• Small signal Nth order ladder RC network

• ESR RC0(ω) and capacitance CC0(ω)• Parasitic inductance is neglected at the

frequency of interest

ESR versus frequency

0,3

0,35

0,4

0,45

0,5

0,55

0,6

0,001 0,01 0,1 1 10 100 1000

Fequency [Hz]

ESR

[m

Ohm

]

Capacitance versus frequency

0

500

1000

1500

2000

2500

0,001 0,01 0,1 1 10 100 1000

Fequency [Hz]

Cap

acit

ance

[F]

Ultra-capacitor Macro Model

An example: 2500F ultra-capacitor cell. The parameters at U0=1V

0

00 )(

1)(00

CC

CCCC

UuCUuCUuC Cj

RjZ

24

Frequency Dependent Capacitance and Energy Capability

• The capacitance depends on frequency CC0(ω)• Is it reflected to a real application and how?• Specific energy depends on frequency (pulse width), Fig A• The capacitor is charged with a short pulse and a long pulse, Fig B

• Final state of charge is expected to be the same in both cases , but it is not a case

A. Specific energy versus pulse width B. Pulse width effect on final state of charge

1

2

2

1

0

01200

C

CC U

UUCE

25

Simplified Model• First order nonlinear RC model

• ESR RC0 is assumed constant• The capacitance is voltage dependent

• The ultra-capacitor current

CCC ukCuC 0

dt

duuC

dt

du

du

udCuuC

t

Qi C

CIC

C

CCCC

26

Simplified Model

27

Capacitance versus Voltage

Capacitance & ESR versus Temperature

CCC ukCuC 0

25exp45.025155.00 RESR

Simplified Model

28

0

6

00

1

1035.65.24

96.0ln CC

T

C

TRP

60

max02 102 C

UR CP

tUii CPP ,,max0

Leakage current modeled by a Shunt Resistor

Leakage current modeled by a Shunt Current Source

(4% @ T=72h)

(2uA/F)

Simulation/Control Model

29

CC

CC

ukCi

dt

du

2

1

0

Large Signal (Nonlinear) Model

Small Signal (Linear) Model

000 CCCC iRuu

C

CC

CCC

CC

C uUkC

Iki

UkCdt

ud 2

0

00

0 2

2

2

1

000 CCCC iRuu

Ultra-capacitor Energy Capability

30

• Stored Energy

• “Energetic” Equivalent Capacitance

• Energy realized to the load

• Specific Energy

220 2

1

3

4

2

1CCECCCCC uuCuukCuE

CCCE ukCuC

3

40

DCHT

CCCCCCC

LOSSESCC

dttitRUUkUUC

EEE

0

200

3min

3max

2min

2max

0

0

3

2

2

M

uukCuSE

CCC

CC 2

3

4 20

Ultra-capacitor Energy Efficiency

31

• Charging /Discharge Efficiency

• Round Trip Efficiency

CHCH

CH

IN

CHINCH EE

E

E

EE

DCH

DCHDCH

DCHOUT

OUTDCH E

EE

EE

E

CHCH

DCHCH

CHCH

DCHDCH

IN

OUTRTP EE

EEE

EE

EE

E

E

0

Ultra-capacitor Power & Current

32

• Maximum Power

• Maximum Power IEC 62391-2

• Maximum Specific Power

• Maximum Specific Power IEC 62391-2

• Maximum 1s Current• Carbon Loading

Capability

0

2

0 4 C

CMAX R

uP

0

2

0 12.0C

CD R

uP

MR

uSP

C

C

0

2

max0 4

MR

uSP

C

CD

0

2

0 12.0

12

1

00

0max

C

C

RC

CuI

CLCL kCI 0max

Frequency Dependent Losses

• ESR RC0 depends on frequency

• Consider steady state operating voltage U0

• The system is linear (linearized)• However

• If the current is sinusoidal• Instantaneous terminal voltage and

power

tRR CC 00

tiRtu CCESR 00

)sin( 000 tIti CC

ESR versus frequency

0,3

0,35

0,4

0,45

0,5

0,55

0,6

0,001 0,01 0,1 1 10 100 1000

Fequency [Hz]

ESR

[m

Ohm

]

The capacitance C0=2500F at operating point U0=1V.

Nominal voltage UC0=2.5V

tiRtIRtUtu CCCCESRESR 00000000

)sin()sin(

titiftitutp CCCESR 000)(

00 CC RR

33


0

000 sink

kkCC tkIti

0

0000 sink

kkCCESR tkIkRtu

0

00000

00 sinsin)(k

kkCCk

kkC tkIkRtkItp

0

20002

1)(

kkCAV IkRTP

Voltage and current

The current spectrum

34

Case A: • The current is periodic .• The terminal voltage and instantaneous power

• Average power PAV over a period T


dejIjRtu tjCESR

02

1

dejIjRdejItp tjC

tj

024

1)(

dejIti tj

2

10

0)(1

lim2

2

T

TT

AV dttpT

P

dtdejIjRdejIdttpE tj

Ctj

RESR

024

1)(

Voltage and current

The current spectrum

35

Case B: • The current is not periodic . • The terminal voltage and instantaneous power

• Energy EESR determines the ultra-capacitor thermal image

Ultra-capacitor Thermal Model

36

Heat Generation1. Irreversible heat generation, 2. Reversible heat generation

Thermal Equivalent Model .

Cell Thermal Model

Simplified 1st order Model

Ultra-capacitor Thermal Model

37

Cell Temperature A general case

A Very Short Pulse non-repetitive Pulse

A Very Long Pulse, or

A Train of Repetitive Pulses at High Frequency

THTH

THC CsR

RsPs

10

TH

Ct

C

TH C

Edttp

Ct

0

0

0

1

THC Rtpt 0

TH

T

C

RP

RdttpT

tRP

0

0

1

Charge and Discharge Methodes

1. Constant voltage• The ultra-capacitor charge/discharged from/to

voltage VBUS via a resistor R0

• The resistor is must, but not practical• Low efficiency

2. Constant current• Very common method in applications• Maximum discharge power and current

Constant voltage charge/discharge circuit

Constant current charge/discharge circuit

0

2

0 4 C

CMAX R

uP

0

0

C

CMAX R

uI

• Maximum charge power and current

• limited by the current source terminal voltage U0max

0

max0max00

C

CMAX R

uUUP

0

00

C

CMAXMAX R

uUI

38

3. Constant power• The most common method in applications• Maximum discharge power

• • If the power source exceeds maximum power, the

voltage collapses• Maximum charge power is limited by the power source terminal voltage

• Charge/discharge current (resistance RC0 is neglected)

Constant power charge circuit 0

2

0 4 C

CMAX R

uP

0

max0max00

C

CMAX R

uUUP

Charge and Discharge Methodes

tPUC

UC

U

Pi

CC

C

C

CC

02

0

200

0 2

39

Ultra-capacitor Modules

40

• Cell voltage is very low, not practical for power conversion application

• N cells are series connected in a module,• M cells are parallel connected

Ultra-capacitor Modules

41

Module Voltage [V]

Capacitance [F]

SE [Wh/kg] Weight [kg]

Maxwell Technologies

BMOD0500 16 500 3.2 5.51

BMOD0083 48 83 2.6 10.3

BMOD0130 56 130 3.1 18

BMOD0094 P075 75 94 2.9 25

BMOD0063 P125 125 63 2.3 65

LS MTRON Ultra-capacitorsLS 16.2V / 500F 16.2 500 3.5 5.1LS 33.6V/250F 33.6 250 4 9.8LS 50.4V/167F 50.6 166 3.43 17.2LS 201.6V/41F 201.1 41 2.21 104

Integration of the Ultra-capacitor into the Conversion System

• The ultra-capacitor voltage varies with the state of charge (SOC)

• If the ultra-capacitor is directly connected 1. The ultra-capacitor is oversized

• Not cost and size effective2. The conversion system voltage

rating is N+1• Not cost effective• Efficiency issue

Not convenient to directly connect the ultra-capacitor to the conversion system

42

NNC

C

E

E

u

u

0

0

Integration of the Ultra-capacitor & the Conversion System

• The ultra-capacitor voltage has to be matched with the power processor dc bus voltage VBUS

• An ideal transformer with variable gain M

• Bidirectional “Loss-free” dc-dc converter emulates an ideal transformer

• Topology of dc-dc converter in numerous variety

00 CC

BUS

u

const

u

VM

43

• Mature technology• Numerous products in mass-production• The cost is driven by the market, mainly automotive industry• The cost prediction for 2010 was 1.28US$ per kJ on cell level

• The cost on module level >double • Packaging cost• Voltage balancing circuit cost

• Significant cost reduction is expected • Development of high energy density

devices• Higher capacitance• Higher cell voltage

• Mass-production and application

The Ultra-capacitors Today

44

The Ultra-capacitors of Tomorrow Development Objective

1. Energy density at least x10 of the existing (50Wh/kg to 100Wh/kg)2. Lower internal resistance, <50% of existing

• Higher power density

3. Higher cell voltage• Lower number of series connected cells into a string, higher reliability

4. Self balancing capability• No need for an external balancing circuit, lower complexity, lower cost, higher

reliability

5. Higher operating temperature• >85ºC, >10 years life time

6. Strongly voltage dependent capacitance (Voltage-SOC characteristic )• Smaller variation of the terminal voltage• Higher minimum discharge voltage , lower current rating, smaller interface dc-dc

converter

45

The Ultra-capacitors in Development

Existing Nano tube Nano gate EesE Mega Farad ultra-cap

Energy density [Wh/kg]

~5 20-40 50-80 280 500

Operating voltage [V]

~2.8 ~3 ~3.9 3000 <2.8

Internal resistance[mΩ]

~0.66+1000/C As existing As existing Low ???

1. Nano tube carbon electrode. Uniform pores and much higher effective surface. The concept developed at MIT [5]

2. Nano gate. The concept developed in Okamura Laboratory [6]3. EesE, high voltage multilayer ceramic capacitor. The dielectric is high

permittivity ceramic [7]4. Mega Farad Ultra-capacitor. Additional thin layer of high permittivity

material on top of the activated carbon electrode [8]

46

PART THREEApplications

1. Controlled electric drives2. Renewable energy3. Diesel electric generators4. STATCOM and power quality5. UPS6. Traction drives

47

Application 1: Controlled Electric Drives

The project “Application of ultra-capacitors in controlled electric drives” was sponsored by Schneider Toshiba Inverter, Pacy sur Eure, Franca and the Laboratoire d’Électrotechnique et d’Électronique de Puissance de Lille, l’Ecole Centrale de Lille, Villeneuve d’Ascq, France from 2007 until 2010.

• Electric drives convert electric energy into mechanical energy • Move an object

• > 60% of total electricity production is consumed by electric drives• Numerous applications

1. Hoisting and lift applications2. Machines with intermittent load3. Blowers and Pumps Applications4. Traction drives5. Home appliance drives

48


Application issues 1. Braking energy

• Dissipated on a brake resistor• 25 to 40% of consumed energy

2. Ride through capability• Critical in certain applications• 10kE up to 1ME per interruption• Ride through time up to 10 to 15s

3. High ratio of peak to average power• Peak power penalties• The installation size and cost• Voltage fluctuation and flickers

Power profile of a 40 t Hoisting application

Power profile of an intermittent load VSD

49


• Ultra-capacitor energy storage • To store and restore the

drive braking energy• Low voltage ride through

capability• To smooth the drive

power• The ultra-capacitor is

controlled independently from the motor control• As an option

A controlled electric drive with an ultra-capacitor as energy storage

50


Basic operating modes

• MM: The mains motoring

• B:Braking• STB: Stand by• MC0: Energy

recuperation • RT: Ride through • MM-CH: Mains and

charging• MPFM: Mains power

smoothing

51


A. Braking and energy recovery B. Low voltage ride through

Rated power PN=5.5kW, The grid voltage VMAINS=400V, dc bus voltage VBUS=650V, Ultra-capacitor C0=0.4F UC0=780V

52


A. Power smoothing function NO B. Power smoothing function YES

Rated power PN=5.5kW, The grid voltage VMAINS=400V, dc bus voltage VBUS=650V, Ultra-capacitor C0=0.4F UC0=780V

53

Application 2: Renewable Energy • Contribution of wind and solar renewable

sources to total electricity production is dramatically increasing

• Renewable sources are connected to the grid• The production output (power) is not constant

and deterministic• Wind speed• Radiation coefficient

• Significant influence on the grid voltage and frequency regulation and the grid stability• Time scale of couple of seconds

www. ceinsight.com

www.renewablepowernews.com

54

Application 2: Renewable Energy

Wind Energy • Wind speed is not constant and

deterministic • The turbine power strongly depends

on the wind speed

• The blade pitch• To control the turbine power• Smoothing of the wind fluctuations• But, not fast enough• The angle saturation is limitation

• A better method to filter the power fluctuations is a need

3WOUT CvP

Wind speed and turbine power [9] 55


• Ultra-capacitor energy storage• To smoot the wind turbin output

power1. Decentralized Connection• Connected on the wind mill level• The connection to the dc bus via an

interface dc-dc converter2. Centralized Connection • Connected on the wind park (farm)

level• Only one energy storage and

interface, but Increased the interface complex

(AC-DC & DC-DC) No redundancy

56


• pGRID smooth

• pWIND>pGRID Ultra-cap chargerd

• pWIND<pGRID Ultra-cap dischargerd57

Application 3: Diesel Electric Generators

Energy sources used to produce electric energy from fuel, such as diesel or natural liquid gas

• Diesel electric generators are the most common solution Diesel Internal Combustion Engine (ICE) drives a

three-phase generator The generator feeds different electric loads, motors

or independent network installations1. Rubber tyred gantry cranes (RTGC)2. Hybrid dampers (Diesel-Electric Traction Drives)3. Hybrid excavators4. Autonomous diesel-electric power supplies

1

2

3

4

www.hardydiesel.com

www.archtings.com

www.toreuse.com

www.allproducts.com

58

Application 3: Diesel-Electric Generators

• An ICE drives a three-phase generator that feeds three-phase motor(s) via a common dc link and dislocated inverter(s)

• A brake resistor burns the drive braking energy. An ICE is not regenerative device!• 30 to 40% of consumed fuel is burned on the resistor…40t RTGC consumption of

15 to 25 [L/hour] • Peak to average power is very high

• The ICE sized on the peak load. The system is oversized, not cost effective

59


• Long term UPS, Mobile generators…A Diesel ICE drives three-phase PMSG

• The load is directly connected to the generator• The output frequency and voltage must be constant the generator speed has

to be constant • The engine is oversized and runs at suboptimal operating point

• Low efficiency, high consumption and pollution• The output voltage THD with nonlinear load (diode rectifier)

www.hardydiesel.com

60


An ultra-capacitor as short term energy storage device • All the applications have similar

structure• A common solution can be

implemented• Ultra-cap connected to the dc

bus via a dc-dc converter• The ultra-capacitor absorbs

fluctuations of the load power• The generator power is smooth,

while the speed is variable• An optimal operating point.• Fuel saving up to 50% [10]

61


• pICE smooth function

• pLOAD>pICE Ultra-cap dischargerd

• pLOAD<pICE Ultra-cap chargerd62

Application 4: STATCOM & Power Quality

• STATic COMpensator (STATCOM) is a three-phase inverter that emulates synchronous voltage source connected to the Point of Common Coupling (PCC) of a power system

• Reactive and distortion power control• PCC voltage control and harmonic compensation

•

63


• The STATCOM is not a real voltage source. The dc bus is only a capacitor CBUS with limited energy• Not possible to control active

power flow on long term (<40ms)• Some applications need active

power flow control• Arc furnaces, large drives…...

• Smoothing active power fluctuation

1. The grid fault management • Handling the grid faults

without significant degradation of the supply voltage power quality

64


• An ordinary STATCOM equipped with ultra-capacitor energy storage• The ultra-capacitor SOC controlled via an interface dc-dc converter

• Dc bus voltage VBUS=const is controlled • The STATCOM control is independent

65

Application 5: UPS

• Uninterruptible power supplies (UPS):• Uninterrupted, reliable, and high-quality power supply for critical loads• Hospitals, date centers, telecommunication and military facilities

• Static (Static power convertors + Energy storage ) & Rotating (ICE & Generators• On-line, Off-Line & Interactive UPS

• Static UPS: Short and medium term applications, from couple of seconds up to couple of hours

• Hybrid UPS: Long term applications, up to couple of days or undefined period

66

Application 5: UPSShort Term UPS Applications

• From couple of seconds up to couple of minutes• Electrochemical batteries are still the main choice as energy storage• The battery size is defined by the power rating not energy capability

• Not cost effective in case of very short bridging time (<10 to 15s)

• The battery life time and maintenance limiting factor

• Ultra-capacitor as an alternative energy storage• Cost effective in case of short

bridging times• Long life• Fast recharging• Maintenance free

67

Application 5: UPSShort Term UPS Applications

68

Application 5: UPSLong Term Hybrid UPS Applications

• From couple of minutes up to couple of days

• Diesel ICE or Hydrogen as energy source• Warming (start up ) time is couple

of seconds• An additional energy storage with

fast response is the need to bridge the start up time

• Ultra-capacitor is a candidate• The philosophy as short term UPS

69

Application 5: UPSLong Term Hybrid UPS Applications

70

Application 6: TractionOverview

• Rail Vehicles1. Heavy-rail Catenary Supplied Vehicles2. Heavy-rail Diesel-supplied Vehicles3. Light Rail Rapid Transit Vehicles

• Road Vehicles4. Public Transportation Catenary Supplied Vehicles,5. Hybrid Electric Vehicles,6. Electric Vehicles

• Off-Road Vehicles 7. Heavy Tracks and Dampers

1

2

3

4

5http://en.wikipedia.org

71

Application 6: TractionHeavy-rail Catenary Supplied Vehicles

• The drive supplied from overhead line and rail via iron wheels• Supply voltage is 25kV 50Hz and 15kV 16 2/3 Hz

1. Step-down transformer 2. Rectifier3. Traction inverters and three phase motors4. Dynamic brake resistor nd dc-and dc-dc

converter • Brake resistor dissipates the braking

energy• Deceleration and down hill driving

• Braking energy is wasted• Low efficiency, high stress on the supply

grid

72

Application 6: TractionHeavy-rail Diesel-supplied Vehicles

• Traditionally used in North America and in some part of Europe1. Diesel internal combustion engine (ICE) with three phase generator2. Rectifier 3. Traction inverters and three-phase traction motors4. Dynamic brake resistor and dc-dc converter

• Brake resistor dissipates the braking energy • Deceleration, and down

hill driving• Wasted energy

• Low efficiency, High fuel consumption and ICE stress (ICE oversized)

73

Application 6: TractionLight Rail Rapid Transit Vehicles

• Light rail traction drives• Urban public transportation• The drive is supplied from overhead line, dc voltage,1.5kV and 3kV• On-board equipment is

similar to heavy traction vehicles equipment

• The same inconveniences• Braking energy is wasted

• Low efficiency• High stress on the supply grid, voltage

variations due to high supply impedance

74

Application 6: TractionPublic Transportation Catenary Supplied Vehicles

• Very similar to Light Rail Rapid Transit Vehicles• The on-board equipment supplied from overhead lines via two pantographs• The traction drive usually lower power rating than Public Transportation

Catenary Supplied Vehicles

• The overhead dc voltage 750 V

• The same inconveniences • Braking energy is wasted

• Low efficiency • High stress on the supply grid, voltage variations

due to high supply impedance

75

Application 6: TractionHybrid Electric Vehicles

• Vehicles driven by a combination of an internal combustion engine (ICE) and an electric drive

• The ICE operates at the maximum efficiency point1. ICE2.Traction motor/generator3.Traction inverter, and 4.Electric energy storage• Series hybrid• Parallel hybrid

Series hybrid vehicle

76

Application 6: TractionPure Electric Vehicles

• Vehicles driven by an electric drive only1. Energy storage 2. Electrochemical converter3. Electric traction drive

• An electrochemical converter converts stored chemical energy into electric energy

1. Electrochemical batteries• The energy storage and

electrochemical generator2. Hydrogen fuel cells

• Compressed hydrogen• Fuel cell

77

Application 6: TractionThe application issues

• High peak to average power ratio,• Highly positive power whenever the vehicle

accelerates, and• Highly negative power whenever the vehicle

decelerates• High energy sources

• Overhead line supply• Diesel ICE• Electrochemical batteries, fuel cells

• Sensitive on peak power• Oversized

• Do not accept negative power, partially or completely • Dynamic braking resistor and mechanical brake are used

78

Application 6: TractionA Common Solution

• General case vehicle• Common dc link• Voltage VBUS is accessible for “the

external world”• The vehicle is sensitive on peak power,

positive as well as negative • An external energy storage device is

connected to the common dc bus• Energy storage

• Short term (low energy) high power energy storage device

1. An ultra-capacitor, and 2. An interface dc-dc converter

79

Application 6: TractionA Common Solution

The primary source power pPS is smooth regardless on the traction drive load

80

Applications: Summary

• The same structure for all the power conversion applications:• Power conversion

system connected with an ultra-capacitor via a dc link dc-dc convertor

• The ultra-capacitor sizing and control parameters differ from application to application

81

PART FOUR:The Ultra-capacitor Selection & Design

1. Introduction2. Voltage Rating & Capacitance3. Losses & Conversion Efficiency4. The Module Thermal Design5. The Module Design-Voltage balancing6. The Module Design Summary

82

IntroductionThe Objective

• Design the ultra-capacitor for the application requirements

The Fact(s)• An ultra-capacitor is characterized by three main parameters

1) Terminal voltage uC0 • The module voltage rating

2) Capacitance C(uc)+C0

• Energy storage capability

3) Internal resistance RC0

• Losses, efficiency and thermal design

The main design stepsI. Compute the above parameters according to the applicationII. Design the ultra-cap module according to the parameters

83

Voltage Rating & Capacitance

The parameters to be selected according to the application requirements

1. UC0max Maximum operating voltage

2. UC0min Minimum operating voltage

3. UC0inM Intermediate operating voltage

4. UC0N The ultra-capacitor rated voltage

5. C0 The ultra-capacitor rated capacitance

RC0 Equivalent series resistance

C0 and RC0 are not independent parameters

84

Internal & External Voltage

85

External Voltage uC0 and Internal Voltage uC are different

ESR and current

Normalized Voltage difference

The difference cannot be neglected!

N

C CR

11068.0 3

0 CLNC kCi 0

NC

CLN

NC

C

U

kC

U

u

0

3

0

0 11068.0

Capacitance Definition

86

Nonlinear capacitance

Voltage to capacitance coefficient

CCCC ukCuC 00

0

0

0

0

11 kU

C

C

C

U

Ck

NC

N

NNC

NC

C

NC

NC

NC

NN

N

CC uU

kkCuk

U

CC

C

CuC

0

000

0

00

11

CN is the nominal capacitance at the nominal voltage UC0N,

k0 is normalized initial capacitance.

1. Maximum operating voltage UC0max

• depends on the dc-dc converter topology

• m is voltage gain of the interface dc-dc converter. • Directly connected non isolated dc-dc converter, m=1.

maxmax0 BUSC mVU

Voltage Rating

87

System Low Voltage DC

Single Phase 110V

Single/Three Phase 230V

Three Phase 400V

Three Phase 690V

VBUSmax 16 to 56V 150 to 450V 350 to 450 700 to 900V 1000 to 1200V

Application Area

Telecom, UPS,

AutomotiveDomestic

Domestic, Industry in

JapanDomestic and

Industry Industry

00max0max0max , CCCCC RiUUU

1. Minimum operating voltage UC0min

• Determined by the dc-dc converter current capability assuming constant conversion power PC0

• Internal Voltage• Practically, UC0min >=0.5 UC0max

max0

0min0

C

CC I

PU

Voltage Rating

88

0max0min0min0min , CCCCC RiUUU

3. Intermediate operating voltage UC0inM is “a long term” average voltage. It depends on the application profile.

Voltage Rating

89

UPS Application Controlled electric drive Application

3. Intermediate operating voltage UC0inM

• Charging energy

• Discharging energy

CHDCH

CCHCDCHinMC EE

UEUEU

2

min02

max00

CHCH T

CHCC

T

CHCH dtitRdttPE0

200

0

CHDCH T

DCHCC

T

DCHDCH dtitRdttPE0

200

0

Voltage Rating

90

4. The ultra-capacitor rated voltage UC0N

• Determines the ultra-capacitor life time

32

0

010exp exp, kk

U

ukuT

NC

CC

• The operating voltage uC0 is not constant

• Average life time for a given voltage profile uC0(t) and temperature profile (t)

• Voltage rating UC0N for expected life time

The ultra-capacitor operating temperature

uC0 The ultra-capacitor operating voltage

k1, k2, k3 Coefficients

ttuUT CNCAV ,0,0

The ultra-capacitor voltage profile

,0 AVNC TfU

Voltage Rating

91

5. The ultra-capacitor rated capacitance• Voltage dependent capacitor

• The initial capacitance C0

• UC0max, UC0max and ECH, from (4.1),(4.3) and (4.4).

• The ultra-capacitor approximated as a linear capacitor ( kc=0)

• The capacitance is (4.11) determined for the energy capability (4.4)• Is there another criterion to define the capacitance?

• Conversion efficiency, cost, or something else?

20

2max0

30

3max00

2

3

2

inMCC

inMCCCCH UUUUkEC

20

2max0

0

2

inMCC

CH

UU

EC

CCCC ukCuCCuC 00

Capacitance

92

Losses & Conversion Efficiency• The ultra-capacitor losses

• Valid only and only if the ultra-capacitor internal resistance is constant at the frequency of interest

• If not a case, the losses must be computed as frequency dependent losses

0

20002

1)(

kkCAV IkRTP

GDISCHARGINtPUC

C

CHARGINGtPUC

C

PRtP

CC

CCCCC

02

00

0

02

min00

0

200

2

2

93

Losses & Conversion Efficiency• The energy dissipated during one charge cycle

• The resistance RC0 depends on the capacitance C0

0

)(

0

00

0

0

C

CR

C

R CC

CHC

CCCT

CCHC

CCLOSSES EUC

UCCPRdttPEUC

CPRE

CH

2ln

222 2max00

2max00000

0 02

max00

0200

94

20

)2(

0

)1()0(0 C

k

C

kkR RCRCRCC

Losses & Conversion Efficiency

The ultra-capacitor module resistance versus capacitance. The module rated voltage UC0N=800V

95

200

0

17.042.1046.0

CCRC

200

0

68885637.1

CCRC

Losses & Conversion Efficiency• Round trip efficiency is a function of the capacitance C0

0.5 1 1.5 2 2.5 3

75

80

85

90

95

100

Eef

fici

ency

[%

]

Capacitance Co [F]

4kW5.5kW7.5kW11kW15kW

500 1000 1500

75

80

85

90

95

100

Eff

icie

ncy

[%]

Cost [€]

4kW5.5kW7.5kW11kW15kW

)(21100 0CE

E

CH

LOSSES

• Round trip efficiency is a function of the capacitor cost

96

The Module Design• Series-parallel Connection • N number of series

connected cells

• M number of parallel connected cells

97

cellCON

NC

U

UfloorN 0

N

C

CfloorM

cellN

N

The Module Thermal DesignThe problem(s) identification

• The internal temperature distribution has an enormous influence on the ultra-capacitor life time

• Understanding the ultra-capacitor thermal picture and an optimal cooling strategy are crucially important

Thermal Design Steps1. Thermal model identification

• Complex 3D model• Simplified 1D model

2. Losses calculation3. Hot spot temperature calculation4. Cooling system design for given temperature and power profile

98

The Module Thermal DesignSingle cell 1D model

• A cell is divided in sub-domains• Pj losses of j-th sub-domain

• Cj thermal capacitance of j-th sub-domain

• Rj j-th sub-domain to the cell case thermal residence

• j temperature of j-th sub-domain

• C temperature of the cell

• RC_A the case to ambient thermal resistance

• AMB ambient temperature

99

The Module Thermal DesignThe module 1D model

• PCn losses of the n-th cell

• CCn thermal capacitance of the n-th cell

• Rnn the n-th cell to the module case thermal resistance

• R1n the 1-st cell to the n-th cell cross coupling thermal resistance

• Cn temperature of the n-th cell

100

The Module Thermal Design

Thermal Design Steps

Step 1 If possible identify hot spot domain power and thermal resistance Pj, Rj

Step 2 Define hot spot maximum temperature jmax For expected life time (4.8)

Step 3 The cell case maximum temperature

Step 4 The module case maximum temperature

Step 5 The module to ambient thermalresistance

Step 6 Cooling system design

jjjC PR maxmax

nnnCM PR maxmax

n

AMBMAMBM P

R maxmax_

101

Voltage Balancing• An elementary cell voltage is low UC01~2.7V,

• Not sufficient for power conversion applications• UC0N is tens and hundreds volts• Tens and hundreds of cells series connected

• Total module voltage may not be equally distributed1. Dispersion of the cells capacitance, ~ +/- 20%

• Dynamic variation of the voltage

2. Dispersion of the cells equivalent series resistance• Dynamic variation of the voltage

3. Dispersion of the cells leakage current• Long term voltage variation

• A voltage balancing circuit is required to prevent the cell overvoltage and the life time degradation

dtiC

dtiC

dtiC C

n

CC 0

0

0

02

0

01

1......

11

00002001 .......... CnCCCCC iRiRiR

dtiC

dtiC

dtiC n

n

0

0

02

02

01

01

1......

11

102

String Voltage Balancing1. Passive resistive balancing circuit

• Additional leakage current , low efficiency, high reliability, no dynamic balancing capability

2. Passive voltage clamping circuit• The cell voltage is limited but not equally

distributed on the cells, limited dynamic balancing capability

3. Switched resistor balancing circuit• Voltage actively controlled, low efficiency, limited

dynamic balancing capability

4. Switch mode balancing circuit• Voltage actively controlled, good balancing

capability, high efficiency, high complexity

1 2

43

103

String Voltage Balancing• Switch mode balancing circuit,

• Active voltage sharing control, • Good balancing capability, • High efficiency, • High complexity

A. Centralized fly-back converter• No feedback control is required,• Multi-winding transformer is

complexB. Decentralized half bridge

• Simple inductors n-1• Feedback control is required,• High number of switches 2n-2• Complex control and driving circuit A B

104

The Module Design SummaryStep 1 Select the module rated voltage

Step 2Select the module capacitance for

given energy capability or efficiency required

Step 3 Number of series connected sub-cells

Step 4 Capacitance of a sub-cell

Step 5 Number of parallel connected cells

Step 6 Thermal design If needed, go to step 4, select new cell and repeat steps 5 and 6

Step 7 Voltage balancing circuit

10

0intNC

NC

U

UN

NCC 001

CC

CM

0

01int

,0 AVNC TfU

20

2max0

0

2

inMCC

CH

UU

EC

0C

105

Ultra-capacitor Testing• The internal resistance and capacitance-Charge/discharge Test

106

Ultra-capacitor Testing• The internal resistance and capacitance-Charge/discharge Test

107

Test Cycle Definition

ESR Determination Capacitance Determination

DCHC

CC I

uR

0

00

0201

0

CC

DCHCN UU

tIC

Ultra-capacitor Testing

108

Hold Method Hold Time Charge

CurrentDischarge Current Referent Voltages

Open circuit

THOLD IC0(CH) IC0(DCH) UC01 UC02

30min 75mA/F 75mA/F80% of UC0max

40% of UC0max

Test parameters defined by EC62391 standard


CurrentDischarge Current

Constant Voltage


30s 50mA/F 5mA/F60% of UC0max

30% of UC0max

Test parameters defined by EUCAR standard

Ultra-capacitor Testing

109


CurrentDischarge Current

Constant Voltag


300s90% of UC0max

70% of UC0max

Test parameters defined by IEC62576 standard

2min

2max0

3min

3max

0

minmax02

min2

max0

0

0

31

4

1

6

95.01

CCCCN

CCCC

NC

DCH

UUkUUU

k

UUkUUU

k

R

I

Charge/Discharge Current specified for minimum 95% efficiency

Ultra-capacitor Testing• Leakage Current & Self Discharge Test

110

Test set-up

Self Discharge: Measurement ProcedureLeakage Current: Measurement procedure

PART FIVE:The Interface Converter

1. Introduction 2. State of the Art 3. The Converter Design4. Design Example5. Control of The Interface Converter6. Controller Design Example

111

Introduction• Flexible and controllable interface between the ultra-capacitor and the power

processor• Required due to the ultra-capacitor voltage to SOC characteristic• For the system flexibility, controllability and efficiency• The interface is a bidirectional dc-dc power convertor• The dc-dc converter control• To control of the dc bus voltage VBUS, the ultra-capacitor SOC (uC0)…etc.• The control can be independent from the power processor control

112

A. Current source Converters Input Current iSW1

continuous, Output Current iSW2

discontinuous, Input Voltage uSW1

discontinuous Output Voltage uSW2

continuous,

C. Voltage Source Converters Input Current iSW1

discontinuous, Output Current iSW2

continuous, Input Voltage uSW1

continuous Output Voltage uSW2

discontinuous,

Background & Classification

113

A. Full scale rated converter

B. Fractional power rated converter

C. Non-isolated dc-dc converter• Galvanic connection between the input

and outputD. Isolated dc-dc converter

• The converter input and output coupled via a high frequency transformer

0CDCDC PP

0min0

0 1 C

BUS

CCDCDC P

V

UPP


114

A. Two-Level Converter Output Voltage 0 or VBUS

B. Multi-Level Converter N series connected semiconductors=VBUS

Output Voltage N steps between 0 and VBUS

C. Single-Cell Converter One semiconductor cell=Output current

D. Multi-Cell Converter N Paralleled cells =Output Current

Cell Current is a fraction of the output current


115

1. Single phase buck convertor • Simple topology• The voltage gain m 1• Full dc bus voltage rating• High switching loses• Limited switching frequency

2. Multi phase interleaved buck convertor• Parallel connected n single phase modules

with shifted switching• The output current ripple 1/n• The inductor ripple high• Low switching losses• The inductor losses are high

State of the Art

116

Cascaded two level single phase modules• The voltage gain (0<m<)• Double conversion, cost & losses

3. Boost-buck convertor• The input and output currents are continuous• Two inductors, one capacitor

• Size and cost• The devices voltage rating

4. Buck-boost convertor• The input and output currents are

discontinuous• Two capacitors, one inductor

• Size and cost

State of the Art

117

Multi level converters• The device voltage rating, VBUS/(n-1)

• Cost effective , high efficiency • Small output filter inductor• Small input filter capacitor• Complex control• The voltage gain m 1

5. Three-level flying capacitor convertor • The ultra-capacitor reference is minus dc bus• An additional flaying capacitor

6. Three-level flying output convertor• No additional components• The ultra-capacitor is floating, Common mode

voltage

State of the Art

118

Isolated Converters• Isolation• High voltage

ratio• Complex and

expensive

7. Single-Phase Double Active Bridge (DAB)

8. Three-Phase Double Active Bridge (DAB)

State of the Art

119

Application Summary

120

Non-Isolated Interface Converters when galvanic isolation not required, Two-Level Single-Cell & Multi-Cell Interleaved Converters:• Medium & High Power Applications• Low voltage applications, the DC bus voltage VBUS<1400V Switching frequency fSW<10kHz 600V, 1200V & 1700V IGBTs and PiN Diodes are used CollMOS used only in discontinuous conduction mode (DCM).

Three-Level & Multi-level Converters are used in following cases:• Low Voltage Applications, the DC bus voltage VBUS<1400V Low & Medium Power High frequency & High power density applications Switching frequency fSW>10kHz 600V & 1200V IGBTs and PiN Diodes are used CollMOS used only in discontinuous conduction mode (DCM).

• Medium Voltage Applications The dc bus voltage VBUS>1400V Switching frequency fSW<10kHz 1200V & 1700V IGBTs and PiN Diodes are used.

Current & Voltage Definition

121

ConstIi

ConstIi

CHCC

DCHCC

00

00

tIC

Utu

tIC

Utu

CHCC

DCHCC

0

0

0

0

0

0

1

1

0,

2

0,2

0

02

00

200

0

00

0

02

00

200

0

00

CHC

CHCC

C

C

CHCC

DCHC

DCHCC

C

C

DCHCC

PtPUC

UC

U

Pi

PtPUC

UC

U

Pi

0,2

0,2

0

0

020

0

00

0

0

020

0

00

CHCCHC

C

C

CHCC

DCHCDCHC

C

C

DCHCC

PC

tPU

ti

Ptu

PC

tPU

ti

Ptu

Constant Current Mode Constant Power Mode

Current

Terminal Voltage

Current

Terminal Voltage

The Converter DesignA design example: Three-level floating output dc-dc converter

A. PWM strategy• Minimization of LC0 and C1, C2 • Losses minimization

B. Selection and design of output filter inductor LC0

C. Selection of input filter capacitors C1 and C2

D. Selection of semiconductor switches SW and freewheeling diodes D

E. Cooling system design

122

A. PWM strategy• PWM carriers shifted for

radians• Effective output frequency• Reduced output current

ripple and input voltage ripple

• The input voltages vC1 and vC2 balanced controlling duty cycles d1 and d2

The Converter Design

123

B. Selection and design of the output filter inductor LC0

1. The inductance LC0

at saturation current

• ΔiC0max the current ripple, fSW switching frequency

2. The inductor losses

• Duty cycle when the ultra-capacitor is discharged with constant power PC0 from max operating voltage UC0max

lossescoreandcopperfrequencyHigh

CfCfCu

lossescopperfrequencyLow

BUS

CDCCLC

ddd

dddiRR

dV

PRdPP

SWSW

12/1,112

2/10,21

3

16,

22

222

max022

2

000

tPUC

UC

U

Vd

CC

C

C

BUS

02

max00

2max00

max0 2

16max0

max0

SWC

BUSC fi

VL

max0

min0

00 C

C

CsatC i

U

PI


124

C. Selection of the input filter capacitors C1,C2

1. The capacitance C

at

• ΔvBUSmax the dc bus voltage ripple, fSW switching frequency

2. The capacitor losses

• Duty cycle when the ultra-capacitor is discharged with constant power PC0 from max operating voltage UC0max

2

223,

2max

max

0

max

0

SWBUSBUS

C

SWBUSBUS

C

fvV

P

fvV

PC

ESR

CC

C

C

BUS

CC

C

C

BUS

BUS

CESR

BUS

CC R

tPUC

UC

U

V

tPUC

UC

U

V

V

PR

d

d

V

PP

02

max00

2max00

max0

02

max00

2max00

max0

2

0

2

0

2

21

1


125

D. Selection of the semiconductor switches1. Voltage rating

2. Current rating

3. The device technology • Defined by the device voltage rating

and switching frequencya) Si MOSFET, Si IGBTb) SiC JFET, SiC MOSFETc) Si PiNd) SiC SBD

VV

VV BUSDSW

2max

maxmax

max0

min0

0maxmax C

C

CDSW i

U

PII

V commutationover-voltage


126

D. Selection of the semiconductor switches4. The device losses

• The current is positive, the switches SW1A and SW2B, and diodes D2A and D1B are conducting

0

2

1

21

02

200

021

ASWBSW

SWITCHING

SWOFFONNN

C

CONDUCTION

BUS

CSW

BUS

CSWBSWASW

PP

fEEdIV

P

dV

Pr

V

PVPP

0

2

11

21

02

200

012

BDAD

SWITCHING

SWQNN

C

CONDUCTION

BUS

CD

BUS

CDBDAD

PP

fEdIV

P

d

d

dV

Pr

V

PVPP

QDD

OFFONSWSW

ErV

EErV

,,

,,,

0

0 The device static and switching parameters


127

The Converter Design ExampleNominal power PC0=5500W

DC bus nominal voltage VBUS=700VUltra-capacitor max voltage UC0max=700VUltra-capacitor min voltage UC0min=350V

Switching frequency fSW=25kHzThe current ripple ΔiC0=3A

IGBT/FWD 600V 30A

VSW rSW *EON+EOFF VD0 rD *EQ

0.8V 27mΩ 80µJ/A 0.9 V 20 mΩ 10µJ/A*Switching losses at VN=300V TJ=150˚C

FILTER INDUCTOR LC0 CAPACITOR C1, C2

High Flux Powder Core (2x) 58192-A2 MKP EPCOS B32774D4106L RDC RAC RC C1, C2 ESR

580µH 38mΩ 0.8Ω 3Ω 2x10µF 3.75mΩ

128

The Converter Design Example

0.50.6

0.70.8

0.9

10002000

30004000

50000

10

20

30

40

50

Duty cycle d (~uC0/vBUS)

The switch losses versus conversion power and duty cycle

Load Pco[W]

The switch losses [W

]

0.50.6

0.70.8

0.9

10002000

30004000

50000

5

10

15

20


The diode losses versus conversion power and duty cycle

Load Pco[W]The diode losses [W

]

The switches SW1A, SW2B and diodes D1B and D2A losses. The dc bus voltage VBUS=700V, switching frequency fSW=25kHz

129

The Converter Design Example

0.50.6

0.70.8

0.9

10002000

30004000

500096

97

98

99

100


Efficiency versus conversion power and duty cycle

Load Pco[W]The Efficiency [%

]0.5

0.60.7

0.80.9

10002000

30004000

50000

5

10


Inductor losses verus conversion power and duty cycle

Load Pco[W]

The losses [W

]

The inductor total losses and the converter efficiency . The dc bus voltage VBUS=700V, switching frequency fSW=25kHz

130

The Converter ControlThe Control Objective(s)

1. Low level converter control (sate of the art)• PWM and protection• Voltage balancing (three level converter )• Current balancing (multiphase interleaved converter)

2. The ultra-capacitor current control (state of the art)• Current limitation

3. The ultra-capacitor voltage control• State of charge (SOC) control

4. The conversion process control• The power processor input or output power (pIN,

pOUT) control• Not necessarily coupled with the power processor

control

131

The Converter ControlThe ultra-capacitor voltage control

1. SOC~ the ultra-capacitor voltage uC0

2. Whenever it is possible, regulate uC0 at intermediate level UC0inM

• UC0inM determined by the application parameters (4.3)

• Controlling uC0 at UC0inM gives maximum flexibility of the power conversion system

3. Limit the ultra-capacitor voltage at maximum operating voltage UC0max

4. Limit the ultra-capacitor voltage at minimum operating voltage UC0min

132

The Converter ControlThe conversion process control

• Numerous possibilities for the power process control• Control via the common dc bus voltage one of the

simplest control schemes [4]1. The power processor controls the dc bus voltage at the

reference VBUS(REF)

• VBUS(REF) is fixed or flexible, defined by the power processor or the conversion process

2. Whenever it is necessary, the ultra-capacitor converter controls the dc bus voltage to lower reference VBUS(min) or upper reference VBUS(max).• VBUS=VBUS(max)The ultra-capacitor is charged

• VBUS=VBUS(min)The ultra-capacitor is discharged

3. The bands V1=VBUS(REF)-VBUS(min) and V2=VBUS(max)-VBUS(REF) are small enough to be acceptable by the power processor

VBUS(REF)<VBUS(max)

VBUS(REF)>VBUS(min)

USF Under supply faultOVF Overvoltage fault

133

An example: Controlled electric drive with ultra-capacitor energy storage

The Converter Control

• The rectifier (AC-DC) controls the dc bus voltage at the reference VBUS(REF)

• The dc-dc converter controls the dc bus voltage at lower or upper reference and charges/discharges the ultra-capacitor

• The ultra-capacitor controller is independent of the rectifier controller• References must be set as VBUS(min) <VBUS(REF)<VBUS(max)

134


Control schema of an ultra-capacitor energy storage unit applied on a controlled electric drive [4]

135


BUS

LOAD

BUS

C

BUS

MAINSBBUS v

p

v

p

v

p

dt

dvC 0

MAINSMAINSMAINS ivp

0

000000

dt

diLiiup CCCCCC

00 2 CC

CC idt

duukC

000 CCCC iRuu

BUS

LOAD

BUS

C

BUS

RECESRBBUS v

p

v

p

v

pRvv 0

Large signal (non-linear) model

• Controlled power sources pMAINS, pC0 pLOAD

• Control variables iMAINS and iC0

• Disturbances vMAINS and pLOAD

136

Small signal (linear) model• Linearization & small signal model• The controllers synthesis

EDISTURBANC

MAINS

LOAD

vMAINSP

CONTROL

MAINS

C

MAINSBUSCBUS

C

BUS

C

sv

sp

sGsGsi

si

sGsG

sG

sv

su

0000

0

00

P

ZC

svsp

siC

CC s

sR

si

susG

REC

LOAD

REC

0

0)(0)(

0)(0

00 )(

)(

BUSBUS

ESRBUS

svsisi

LOAD

BUSP VsC

RsC

sp

svsG

RECRECC

1

)(

)(

0)(0)(

0)(0

BUSBUS

CCCESRBUS

svsp

siC

BUSCBUS VsC

RIURsC

si

svG

RECLOADREC

000

0)(0)(

0)(0

)0(

1

)(

)(


137


Experimental waveforms of the dc bus voltage VBUS [100V/div], the mains current iMAINS [5A/div] the ultra-capacitor current iC0 [5A/div] and voltage uC0 [100V/div]. The load is cycling (10% to 100% to 10%) VBUS(REF)=650V, VMAINS=400V, CBUS=820µF,PLOAD=5500W. a) fBOOST=50Hz and b) fBOOST=1Hz [4]

138

PART SIX:Key Message

139

Key Message• In this course we have discussed ultra-capacitors as emerging

technology for short term energy storage• Candidate for power conversion applications

• We have identified ultra-capacitors as major candidate for short term energy storage in numerous power conversion applications

• Compare to electrochemical batteries, ultra-capacitors have: • Higher power density• Higher cycling capability• Longer life time• Higher operating temperature• Higher safety

140

Key Message• However, current performances are not sufficient for replacement of

electro-chemical batteries in some power conversion applications• Energy density• Cost per kWh• Module design

• The aspects to be significantly improved in future development:• Energy density >(20-50)Wh/kg• Higher operating voltage and self balancing capability• Flat voltage to SOC transfer function

141

Selected References1) B. E. Conway, “Electrochemical supercapacitors, scientific fundamentals and technological

applications, “Kluwer Academic/Plenum Publisher, New York 1999.2) F. A. Burke, “Ultracapacitors: Present and Future,” Proceedings of the Advanced Capacitor World

Summit 2003, Washington, D.C., August 2003.3) J. M. Miller, U. Deshpande, “Ultracapacitor technology: state of technology and application to

active parallel energy storage systems,” 17th International Seminar on Supercapacitors and Hybrid Energy Storage Systems, pp.9-25, December 2007.

4) Petar J. Grbović, “Ultra-capacitor based regenerative energy storage and power factor correction device for controlled electric drives” PhD Dissertation, Laboratoired’Electrotechniqueetd’Electronique de Puissance (L2EP) Ecole Doctorale SPI 072, Ecole Centralede Lille, July 2010, Lille.

5) R. Signorelli, J. Schindall, D, Sadoway, J. Kassakian, “High energy and power density nanotube ultracapacitor design, modelling, testing and predicting performance,” 19th International Seminar on Supercapacitors and Hybrid Energy Storage Systems, December 2009.

6) M. Okamura, K. Hayashi, T. Tanikawa and H. Ohta, “The nanogate-capacitor has finally been launched by our factory,” In Proc of The 17th International Seminar on Double Layer Capacitors and Hybrid Energy Storage Devices December 10-12, 2007.

142

7) R. D. Weir and C. W. Nelson, “Electrical energy storage unit (EESU) utilizing ceramic and integrated circuit technologies for replacement of electrochemical batteries,” US Patent 7 033 4060 B2, April 2006.

8) B. G. Ezzat, “New mega-farad ultracapacitors,” IEEE Trans. Ultrasonics, ferroelectrics, and frequency control, Vol. 56, No1, pp. 14-21, January 2009.

9) G.Cimuca, S. Breban, M. M. Radulescu, C.Saudemont, and B.Robyns, “Design and Control Strategies of an Induction-Machine-Based Flywheel Energy Storage System Associated to a Variable-Speed Wind Generator”, IEEE Trans Energy Conversion, Vol. 25, No. 2, pp. 526-534, June 2010.

10) Sang-Min Kim and Seung-Ki Sul, ”Control of rubber tyred gantry crane with energy storage based on supercapacitor bank,” IEEE Trans. Power Electronics, Vol. 21, No. 5, pp. 1420-1427, September 2006.

11) Petar J. Grbović, Philippe Delarue and Philippe Le Moigne, “The Ultra-capacitor Based Regenerative Controlled Electric Drives with Power Smoothing Capability,” IEEE Trans. Industrial Electronics, IEEE Trans. Industrial Electronics, Vol.59, No. 12, pp.4511-4522, December 2012.

12) Petar J. Grbović, Philippe Delarue and Philippe Le Moigne, “A Three-Terminal Ultra-Capacitor Based Energy Storage and PFC Device for Regenerative Controlled Electric Drives,” IEEE Trans. Industrial Electronics, Vol. 59, No. 1 pp. 301-316, January 2012.

Selected References

143

13) Petar J. Grbovic, Petar J. Grbović, Philippe Delarue and Philippe Le Moigne, “The Ultra-capacitor Based Regenerative Controlled Electric Drives with Power Smoothing Capability,” IEEE Trans. Industrial Electronics, accepted for publication.

14) Petar J. Grbović, Philippe Delarue and Philippe Le Moigne, “A Three-Terminal Ultra-Capacitor Based Energy Storage and PFC Device for Regenerative Controlled Electric Drives,” IEEE Trans. Industrial Electronics, Vol. 59, No. 1 pp. 301-316, January 2012.

15) Petar J. Grbović, Philippe Delarue and Philippe Le Moigne, “Modeling and control of the ultra-capacitor based regenerative controlled electric drive system,” IEEE Trans. Industrial Electronics, Vol. 58, No. 8, pp. 3471-3484, August 2011.

16) Petar. J. Grbović, P. Delarue, P. Le Moigne and P. Bartholomeus, “The ultra-capacitor based controlled electric drives with braking and ride-through capability: Overview and analysis,” IEEE Trans. Industrial Electronics,Vol. 58, No. 3, pp. 925-936, March 2011.

17) Petar J. Grbović, Philippe Delarue, Philippe Le Moigne and Patrick Bartholomeus, ”A bi-directional three-level dc-dc converter for the ultra-capacitor applications,” IEEE Trans. Industrial Electronics Vol 57. No.10, pp. 3415-3430, October 2010.

Selected References

144

http://dx.doi.org/10.1109/TIE.2010.2048838

18) Petar J. Grbović, Philippe Delarue and Philippe Le Moigne, “Interface Converters for Ultra-capacitor Applications in Power Conversion Systems”, EPE PEMC (ECCE Europe) 2012, IEEE Energy Conversion Congress and Exposition, Novi Sad, Serbia, 2-6 September, 2012.

19) Petar J. Grbović, Philippe Delarue and Philippe Le Moigne, “Selection and Design of Ultra-Capacitor Modules for Power Conversion Applications: From Theory to Practice” IPEMC (ECCE Asia) 2012, IEEE Energy Conversion Congress and Exposition, Harbin, Chine, 2-5 June, 2012.

Selected References

145

Thank you very much for your [email protected]

146

mailto:[email protected]

Documents

Ultra-Capacitors in Power Conversion -Applications, Analysis and Design from Theory to Practice- Dr. Petar J. Grbović HUAWEI Technologies Duesseldorf GmbH