By:Netra Gyawali

Yasuharu Ohsawa

Kyoto University

Date: 2009/11/81

Control Strategies of Hybrid Microgrid

Why Renewable?

2

Environmental Concern

Rapid Depletion of Fossil Fuel 

Sources

Liberalized Electricity Market

Recent Advancement of Power Electronics 

and Control

Wakeup Call

Storage Technologies

Action

Renewable Electricity Generation

Motivation

EmbryonicEmbryonic GrowthGrowth MatureMature AgingAging

Wave Energy

Solar ThermalElectric

AdvancedBiomass

Photovoltaics

Wind

Geothermal

NaturalGas

TraditionalCoal

LargeHydro

Nuclear

Oil

CleanCoal

(IGCC)

Energy Productivity/Enabling Technology

Stirling Engines

Generation Technologies:

Fuel Cells

Landfill Gas

New Nuclear

Wind Power

Size of Wind Power System Vs. Time

Source: NREL Report 2008

Wind Power Issues•Wind Energy is one of  the Promising RE for Future

•Five fold increase in 2001‐2007 and Expected to increased threefold in 2008‐2015 

•The technology is in advanced stage and Size also getting larger

•Cost

1980: 40 cents/kWh

2009: 7- 9 cents/kWh

Source: U.S. DOE

Wind Power ( in Japan)

Source: CRIEPI

Recent Declaration of  Japanese  Government Recent Declaration of  Japanese  Government 20% reduction of 20% reduction of COCO2 2 2025 2025 will remarkable change this graphwill remarkable change this graph

Wind Power Issues

Until Now (Low Penetration Level)

Negative Load

From Now on (High penetration Level)Dispatchable

Ride through capability

Regulates Plant Voltage and Power

7

Stand-AloneStand-Alone Grid ConnectedGrid Connected

Wind Power

Wind Power CharacteristicsA variable output source embedded in a variable electricity system : seconds, minutes, hours, days, months, seasons and years

Yearly variation  (Source: ISET (2004) )

8

Hourly variation (Source: www.energinet.dk)

Low Capacity factor, Operational and control Challenges

Isolated mode and also in Weak grid  8Storage

Wind Power and Energy Storage

Short term storageinstantaneous power balance; Buffer storage   (ms‐s)

Midterm  Spin reserve, Load  Shaving  etc.(minutes‐hours)

Long term storage for energy management (day‐month)

Source: ESA 20089

Energy Storage

Conventional Technology: Lead Acid Battery, Pump Storage, Flywheel

Emerging Technology:Ultracapacitor, SMES, H2/Fuel Cell etc.

Proposed System:H2/ Fuel Cell as Mid term and Long term StorageUltracapacitor as Short term Storage

Choice

10

WTG Standard Models

Type A‐ Fixed speed(Conventional  squirrel caged IG)

Type‐B:  Fixed speed

(Wound rotor  IG)

generator

PlantFeeders

PF controlcapacitor s

generator

Slip poweras heat loss

PlantFeeders

PF controlcapacitor s

actodc

Type C‐ Variable speed(doubly‐fed induction generator IG)

Type‐D:  Variable speed(IG, PDSM, SM)

generator

partia l power

PlantFeeders

actodc

dctoac

generator

full power

PlantFeeders

actodc

dctoac

WTG Model (Contd..)

Aerodynamic characteristics• Mechanical power (Pm)

PPmm = = ½½ ×× (air density) (air density) ×× (swept area) (swept area) ×× CCpp ×× (wind speed)(wind speed)33

• Rated Power –Maximum power generator can produce.

• Cp (Power Coefficient)Function of blade pitch and tip-speed ratio

(< Betz Limit - 59% Max )• During a typical dynamic

simulation, blade pitch and tip speed ratio vary, thus Cp and Pmwill also vary

• Cut-in wind speed where energy production begins

• Cut-out wind speed where energy production ends.

Typical Power Curve

WTG Model (Contd..)

Power ‐ Turbine speed characteristics•Locus of Max. Power shifts with the turbine speed

Variable WTG system•Capturing Max. Power available• Absorption of turbulent power

Typical Power vs. Turbine speed Characteristics

500 1000 1500 2000 2500 30000

0.2

0.4

0.6

0.8

1

1.2

Constast rotor speed mode

12 m/s

11 m/s

10 m/s

9 m/s

8 m/s

7 m/s 6 m/s 5 m/s

Locus of Max. Power(Variable rotor speed mode)

Wind turbine characteristics

Pow

er (p

u)

Turbine speed referred to generator side (rpm)

Wind turbine characteristics

WTG Model (Contd..)

Power coefficient as function of pitch angle and the tip and ratio• Typical Cp curve (left) The dashed magenta line shows operating points that correspond to the steady‐state power curve (right)

• During a typical dynamic simulation, blade pitch and tip speed ratio vary, thus Cp and Pm will also vary

Frac

tion

of r

ated

MW

Tip-Speed Ratio (λ)

Pitch Angle (β)

Coefficient of Performance(Cp)

Pitch AngleTrajectory forIncreasing Wind Speed

Pitch angleTip‐speed ratio (λ)

Coeff.  of Power(Cp)

Pitch angle Trajectory for increasing wind speed

Wind speed (m/s)

WTG Model (Contd..)

where Pm Mechanical output power of the turbine 

(W)Cp Performance coefficient of the turbineλ Tip speed ratio (rωgen/vwind)

Air density (kg/m3)A Turbine swept area (m2)V wind Wind speed (m/s)A Tip speed ratio of the rotor blade tip speed 

to wind speedβ Blade pitch angle (deg)J Moment of inertiaD frictional constantTe Electrical torquer    Radius of turbineCi constant coefficient

. ( , )

( , )

..

i

m p windc

pi

i

windgen

mm

gen

gen gen m e

P c Av

cc c c c e c

vr

PT

dJ D T Tdt

λ

λ β ρ

λ β β λλ

λ λ β β

ω

ω

ω ω

−

=

⎛ ⎞= − − +⎜ ⎟⎜ ⎟

⎝ ⎠

= −+ +

=

=

+ = −

5

3

21 3 4 6

3

0 5

1 1 0 0350 08 1

Mathematical  model of WTG (Summarized)

WTG Model (Simulink Model)

WTG Model (Contd..)

Electrical Model of IG (Fifth order model)

Fuel Cell Model

Solid Oxide Fuelcell (SOFC )•Capable to resist High thermal Stress•Suitable for high Power application•High Efficiency •Allows the internal reforming of gases ( Pure H2 is not needed)

Fuel cell Model

FiMM

dtdp

V

FiMM

dtdp

V

FiMM

dtdp

V

outOH

inOH

OHa

outO

inO

Oc

outH

inH

Ha

2

2

2

22

2

22

2

22

2

+−=

−−=

−−=

( )[ ]

inohm

lcon

.OHOH

fc

fc-

Nernst

cellcellstack

actohmicactNernstcell

iRViiBV

iiAV

pppF

RT+

).-(T×..=E

VNVVVVEV

=−=

−=

−

=−−−=

)/1ln()/ln(

/ln2

1529810582291

0act

50

3

222

ntange curre ExchIntting curre LimiI

tsA,B Conssstration lo ConcenV

Ohmic loss Vtion loss ActivaV

Here

l

con

ohm

act

0

tan

,

.

''

''

,

constFaradayF

speciesthiofrateflowMassiM

channelthiofVolumeiV

species'th'iofessurePartial prip

Here

Mass Flow/Partial Pressure Dynamics

Electrochemical Model 

20

Fuel Cell Model (Contd..)

50 100 150 200 250 300 3500

50

100

150

200

250

300

350

Vol

tage

[V]

Static Voltage vs. Current Respone of the SOFC Array

time [s]

Activation loss

Ohmic  loss

Concentration loss

Static V‐I Characteristics of SOFC

Equivalent Electrical Model

Power Interface ModelSimplified  Model of single 

Ultracapacitor

UC in series‐parallel

Source: BOOTCAP Double layer UC

Power Interface ModelDC‐DC Boost converter

Circuit Model

Average Model

State Space equation

[ ]10,0

/1

1)1(

)1(0

][,

__

_

_

=⎥⎦

⎤⎢⎣

⎡=

⎥⎥⎥⎥

⎦

⎤

⎢⎢⎢⎢

⎣

⎡

−−−

−−

=

=

=

+=

CL

B

RCCd

Ld

A

vixwhere

xCv

BvAxdtdx

dd

dddd

dd

Toutddddl

Toutdd

indd

Power Interface Model (contd..)

State Space models :

Thyrister Rectifier

Voltage Source Inverter

Proposed System

System Components

• Wind Turbine Generator (WTG) System– Wind Turbine and Generator – Power electronics interfaces (ac/dc to dc/ac )

• Electrolyzer/Fuel cell/ Ultracapacitor system– Hydrogen Storage and regulation– Oxygen flow regulation– DC/DC interfaces– DC/Ac interfaces

24

Proposed System

25

Proposed Scheme

26

Modeling and Design of Controller

1. WTG Controller– Pitch angle control– Rotor speed control (MPPT)– DC voltage control– AC voltgae/Reactive power control

2. ELZ/FC/UC Controller– DC voltage control– AC bus voltage control– Frequency regulation– ELZ/FC current control– UC charge control

27

WTG Controller

WT

Pitch‐anglecontrol

MPPTcontrol

Vw

ωr

P

Generator Side

DC/ACConverterDC/AC

Converter

PCC Side

IG

28

FC/UC Controller

dq-to-abc DutyCy

LPF

PI

PI

PI

PI

PI

1s

1s

V_d2*

V_d2

I_uc

I_dc*

I_fc *

Duty cycle (D2)

Duty cycle (D1)

Vd_ref

V2

Vq_ref=0

Vq

E_q

E_d

377 theta

gate signal

I_fc

I_i

To DC2

To DC1

To VSC#2

29

Simulation

IG Voltage and Current Built‐up Process

30

0 0.1 0.2 0.3 0.4 0.5-500

0

500V

olta

ge [V

]Terminal AC Voltage

time [s]

0 0.1 0.2 0.3 0.4 0.50

50

100

150

200

Spe

ed [r

ad/s

]

Rotor Speed Dynamics

time [s]

Simulation (Power Flow Dynamics)

31

5 10 15 20 25-50

0

50

Pow

er [k

W]

Active Power

time [s]

5 10 15 20 25-40

-20

0

20

40

Pow

er [k

W]

time [s]

PloadPig

PucPfcPelz

Sequence of Sequence of disturbancedisturbance

•• t=5 st=5 s, load of 30 , load of 30 kW is introducedkW is introduced

••t=10st=10s, wind velocity , wind velocity increased from 7m/s increased from 7m/s to 10m/s.to 10m/s.••t=15 st=15 s, 15 kW load , 15 kW load is releasedis released

Simulation (Voltage Dynamics)

32

5 10 15 20 25440

460

480Vol

tage

[V]

Magnitude of PCC Bus Voltage

5 10 15 20 25620

640

660

680

Vol

tage

[V]

DC BUs Voltage

5 10 15 20 25220

240

260

280

Vol

tage

[V]

UC Voltage

time [s]

Simulation (Partial Pressure Dynamics )

33

0 5 10 15 20 256.98

7

7.02

7.04P

ress

ure

[atm

]H2 Pressure in Tank

0 5 10 15 20 250

0.1

0.2

Mol

ar F

low

[Mol

/sec

] Molar production/Consumption Rate

time [s]

ELZFC

Conclusion

• Wind power being non-dispatchable in nature, long-term, medium-term and short-term energy storage system are necessary for power and energy management.

• In a hybrid /stand-alone system, fuel-cell/hydrogen storage/electrolyzercan be used as mid-term and long-term energy storage and UC as Transient Load Mitigation.

• Simulation Results show that effective voltage and frequency regulation is achieved while fulfilling the operational requirements. Thus they validates the applicability of the proposed scheme in the real system

• Further simulation on worst case scenario and model Validation with real/experimental system is suggested to get further insight of the scheme

• The constraints for utilization of UC and FC/H2 /ELZ with wind power system is the cost conversion effiecieny. The research on these technologies are going on, and in future, the cost effective and efficient solution is expected to achieve.

34

Thank you

Lighting the World

35