onics in systems - universidad técnica federico santa maría · onics in systems by prof. frede...
TRANSCRIPT
1
Power Electronics in Renewable Energy Systems
by
Prof. Frede Blaabjerg, IEEE Fellow
Aalborg UniversityInstitute of Energy Technology
Denmark
2
1. Aalborg University2. Energy Demand3. Energy Technology4. Wind Power5. Solar Power6. Conclusions
Outline
Power Electronics in Renewable Energy Systems
3
1.Aalborg University
Denmark
4
Aalb
org
Un
ivers
ity
Where
are
we
from?
Aalborg University - Denmark
5
· Revenue : 180 mio Euro
· Employees : 1500
· Brutto area 140.000 m2
Research 73.744 m2
Education 53.188 m2
Administration 1.731 m2
Other 11.439 m2
· Ph.D. : 570
· Departments : 12
· Schools : 4 (Appr. 30 Bachelor, 60 Master)
Aalborg University - DenmarkEngineering, Science and Medicine
6
En
erg
y D
em
an
d
2. Energy Demand
7
En
erg
y D
em
an
dGeneral Trends
Energy comsumption increasesMore people (born, longer life-time etc.)More equipmentHigher living standardMore production
Global Energy Market becomes deregulated(electrical
power, natural
gas, etc.)
Global ressources limitedClimate
Change
a global issue
New power sources interesting (e.g. renewable)More efficient use of the existing sources
�• From production
to end user
�• Power balance extremely
important
�• New energy
storage
devices
Therefore
Power Electronics is the Enabling
Technology!
8
Power capacities perspective
until
2020
�• The installed
capacity
has to increase
by over 80%
�• Renewable
energy
source
still modest in 2020
Distributed Power Generation Systems (DPGS) necessary
9
En
erg
y T
ech
no
log
y
3. Energy Technology
10
En
erg
y T
ech
no
log
yClassical
Power Systems
Traditional
Power System
11
En
erg
y T
ech
no
log
yFuture Power System
Less
central power plants
and more Distributed
Power Generation
Smart Grid !
1212
Centralized
in 80�’s
Danish Experience
Wind Power Systems
High
coverage
of
dispersed
generation
today
Ener
gy
Dem
and
and T
echnolo
gy
1313
Danish Experience
Wind Power Systems
more than
21% wind
power penetration!
Target 36% from renewables
by 2030
Ener
gy
Dem
and
and T
echnolo
gy
14
En
erg
y T
ech
no
log
yRenewable
Energy Systems
Important issues for power converter�• reliability and thereby security of supply�• efficiency�• cost�• volume�• power electronics
enabling
technology
�• protection�• control
active
and reactive
power
�• ride-through
and monitoring
15
En
erg
y T
ech
no
log
yPower Electronic System Development
More integrationLower volumeHigher power densityLower cost
0
50
100
150
200
250
300
350
1968 1983 1988 1994 1998 2004
Rela
tive
Unit
Components Functions Weight Volume
Industrial Adjustable
Speed Drives (power electronics)
16
Win
d P
ow
er
4. Wind Power
17
Aerodynamic
Gearbox
Generator PowerElectronicInterface
Transformer
GridWind
control
Win
d P
ow
er
Power Conversion
18
Installed Wind Power in the World- Annual and Cumulative -
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
20,000
1983 1990 1995 2000 2007
Year
MW
per
yea
r
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
100,000
Cum
ulat
ive
MW
Source: BTM Consult ApS - March 2008
Win
d P
ow
er
Wind Power Directions
Windforce 10: • 2010 180 GW• 2020 1200 GW
Source: BTM Consult
2008
Global Wind Power StatusCumulative MW by end of 2001, 2004 & 2007
0
10,000
20,000
30,000
40,000
50,000
60,000
Europe USA Asia Rest of World2001 (24,927 MW) 2004 (47,912 MW) 2007 (94,005 MW)Source: BTM Consult ApS - March 2008
19
Win
d P
ow
er
Wind Turbine Development
�•
Bigger
and more efficient
! �•
3.6-6MW prototypes running
(Vestas, GE, Siemens Wind,Enercon)
�•
2MW WT are
still the "best
seller" on
the market!
2020
EU Status
Wind Power SystemsEner
gy
Tec
hnolo
gy
21
�•
2 squirrel-cage
induction generators (power ratio 1:4)
�• small �–very
low
wind
speed�• large �–
rest of the range
�• variable capacitor
bank�• pasive/active
stall
control
Advantages�• robust -> to grid
faults
�• cheap
Drawbacks�• Requires
a stiff
grid
for stable operation
�• does
not support speed control�•
its
mechanical
construction
must be
able
to support high
mechanical
stress
Fixed
Speed Wind TurbinesW
ind
Po
wer
Wind Turbine Concepts
22
Win
d P
ow
er
Variable Speed Wind Turbines �–
Road maps
Wind Turbine Concepts
23
�• wound-rotor
induction
generator �• Variable pitch
�–
variable speed
�•
30% slip variation around
synchronous
speed �• power converter
(back2back / direct
AC/AC) in rotor circuit
Drawbacks�• use
slip-rings
-> maintenance
�•
power converter
sensitive to grid
faults
-> complicated
protection schemes
Advantages�• smooth
reactive
power control
�• smooth
grid
connection�• reduced
mechanical
loads
on
the WT tower
Variable Speed Wind TurbinesW
ind
Po
wer
Wind Turbine Concepts
24
�• variable pitch
�–
variable speed�• with/without
gearbox
�• generator�• synchronous
generator,
�• permanent magnet generator�• squirrel-cage
induction
generator
�• power converter�• diode rectifier+boost
DC/DC+inverter
�• back2back�• direct
AC/AC (matrix, cycloconverters, etc)
Variable Speed Wind TurbinesW
ind
Po
wer
Wind Turbine Concepts
25
Variable Speed Wind TurbinesW
ind
Po
wer
Synchronous generator with field winding
Permanent Magnet Synchronous Generator
Squirrel-Cage Induction Generator
Wind Turbine Concepts
26
�• multiple stator
windings�• paralleled
power converters
�• better
efficiency
at low
wind
speed�• redundancy
�• used
by some
manufacturers
Win
d P
ow
er
Variable Speed Wind Turbines
Wind Turbine Concepts
27
DemandsReliableMinimum maintenanceSolution competitive economicallyLow power lossesPhysical size limitedWeight limited (if in nacelle)
Back-to-back two-level
voltage
source
converterProven technologyStandard power devices (integrated)
Decoupling between grid and generator (compensation fornon-symmetry
and other
power quality
issues)
Need for major energy-storage in DC-link (reduced life-time andincreased
expenses)
Power losses (switching and conduction losses)
Win
d P
ow
er
Back-to-back
VSC
Power Electronic Converters
28
Win
d P
ow
er
Multilevel
Converter
a) Three-level diode clampedb) Three-level with bidirectional switch interconnectionc) Three-level flying capacitord) Three-level using 3 two-level converterse) Three-level H-bridge cascaded
High voltage capability -> direct connection to MV networksReduced harmonic contentReduced switching lossesIncreased number of switches -> higher conduction lossesunbalances in DC voltage for some of the topologies
Power Electronic Converters
29
Contr
olof
Win
d T
urb
ines
Full Scale
Power Converter
Wind Turbine
Targets for control:maximum power point operationpower limitations for high wind speedsreactive power control
PMSG control level:Maximum power pointControl of grid side converter
DC-link voltageunity power factor
Wind turbine control level:pitch controlpower limitation control
Permanent Magnet Synchronous Generator
3030
Grid C
onnec
tion R
equirem
ents
Operational ranges
Danish Grid Code for Distribution Networks
Great Britain�’s Grid Code for Transmission Networks
Danish Grid Code for Transmission Networks
German Grid Code for Transmission Networks
3131
Power production regulation at Wind Farm Level
Priority 1
Priority 3
Priority 4
Priority 5
Priority 6
Priority 7
Grid C
onnec
tion R
equirem
ents
Active Power Control
Danish Grid Code for Transmission Networks
3232
Low Voltage Fault Ride-through CapabilityG
rid C
onnec
tion R
equirem
ents
x= 300-500 ms
Danish Grid Code
Successive & non-symmetrical faultsLVRT
E-On Grid Code
Grid support by 100% reactive current injection
3333
Vestas V120 off-shore
turbine
Rated
power: 4,500 kW
Rotor diameter: 120 m Hub
height: 90 m
Turbine concept: Gearbox, variable speed, variable pitch
control
Generator: HV DFIG
Vestas A/S Denmark
Target market: Big off-shore
farms
Win
d T
urb
ine
Tec
hnolo
gie
s
Current Development
3434
Enercon
E-112 gearless
turbine
Rated
power: 4,500 -
6,000 kW
Rotor diameter: 114 m Hub
height: 124 m
Turbine concept: Gearless, variable speed, variable pitch
control
Generator: Enercon
ring generator
Enercon
GmbH Germany
Target market: Big on-shore
and off-shore
farms.
Win
d T
urb
ine
Tec
hnolo
gie
s
Current Development
35
So
lar
Po
wer
5. Solar Power
36
·
Monocrystalline Silicon
�•
Efficiency: 12 �–
18 %�•
Shape: round / quadratic
�•
Colour: black / dark-
blue / blueish
�•
Peak power app.: 120 Wp/m2
�•
Price app. 4-5 �€
/Wp
(continuously decreasing by ca 7% p.a.)
Solar cells technologies
Polycrystalline Silicon�•
Efficiency: 10 �– 22 %�•
Shape: quadratic�•
Colour: blueish, shimmer
�•
Peak power, app.: 100 W/m2
�•
Price app. 3-4 �€/Wp
Amorphous silicon�•
Efficiency: 4 �–
9 %�•
Shape: slim ribbons�•
Colour: black / dark brown
�•
Peak power, app.: 50 Wp/m2
�•
Price app. 5-6 �€/Wp�•
can be foldable
Thin film�•
Amorphous Si, cadmium telluride, copper indium diselenide
and many new others!
�•
Efficiency up to 11 % �•
Can be deposited on any surface
�•
Colour depends on materials
�•
Can be clear films mounted on windows or roof tiles
Very
fast development
on
Thin
FilmOrganic
cells
with
very
low
manufacturing
cost
but still short life
time are
emergingMulti
junction
cells
with
efficiency
higher
than
40% are
reported!
37
Concentrating Solar Photovoltaic
Power -
CPV
�•
Dish
technology
�•
Two-axis
tracking
dishes
�•
CPV panels in the focus
of the
dish
�•
Sun light concentrated
with
lenses
or
optical
concentrators
up to x500 typ. with
tracking
�•
High
efficieny/high
temp
silicon
solar cells or advanced III-IV multi-junction technology (~40% eff)
�•
Considerable
lower
solar cell
material�•
Potential lower
overall cost
than
PV�•
200-500 kWe
-commercial,MW
plants
-
near
term
Source: Amonix
�•
Fresnell
lenses
concentrator
with
tracking
�•
25 kWp
unit/850W/m2
�•
26.6% efficiency
triple-
junctionsolar
cells
�•
X 250 concentration
�•
DPGS/Stand alone/pumping/
�•
desalinization/H2 production
�•
CPV farm in Alice Spring, Australia of 20kW units
�•
CPV developments
�•
18 Mwe
installed
up to 2006
�•
6 years
field
experience
(young!)
�•
38% efficiency
solar cells
now, 50% by 2010
�•
40% efficiency
for H2 production
now!
�•
2-3 Eurocent/kWh on
long term
Source: NREL
�•
Solid concentrator
38
Solar cells manufacturing
technologies
�•
Crystalline
silicon
cell
manufacturing
high
energy
demanding. Cost
is saturating
�•
Thin
film technologies
�–
higher
potential for cost
reduction
on
long-term
39
PV Inverters
Directly convert the dc power from solar panels to grid synchronized power
Typical requirements:
�•�“Very�”
high efficiency typ
> 95% (large variety of innovative topologies!)
�• �“Very accurate�”
Maximum Power Point Tracking MPPT (typ
>99% eff)
�• Grid connection standard requirements (apply to certain countries)
�• High performance grid monitoring and synchronization
�• Active Anti-islanding algorithms
�• Isolation,, leakage current monitoring, and dc current injection monitoring
�• High power quality (low current THD)
Typically IGBT/MOSFETS and DSP technologies are used
Due to increased complexity and smaller market the cost of PV inverter is significant higher than the inverters for drives. Typ.400-500 �€/kW
Source: Danfoss
Solar
40
Photovoltaic System Cost
System cost
is expected
to drop to 2.5�€/Wp by 2012 (optimistic!)
Due to the silicon
shortage
during
the last years
the cost
reduction
will
be
delayed
41
So
lar
En
erg
yMaximum
Power Point Tracking
dP/dV > 0
dP/dV < 0
�•
PV cells/panels exhibit a non-linear I-V characteristic �–
there is an optimum working point where the extracted power is the maximum (MPP) �•
The MPP depends on environmental conditions a Maximum Power
Point Tracking (MPPT) system is needed to follow the changes�•
Most of the actual MPP tracers are hill-climbing methods �–
no knowledge of
the PV string type or environmental conditions required
The most used technologies are:
�• Perturb & Observe�• Incremental Conductance�• Constant Voltage�• Parasitic Capacitance
Combinations of the above methods are often used
4242
PV System Configurations
High efficiency Mini-central PV inverters (8-15 kW) are also emerging for modular configuration in medium and high power PV systems
Central inverters�•
10 kW-250kW, three-
phase, several
strings
in
parallel�•
high
efficiency, low
cost, low
reliability, not
optimal MPPT�•Used
for power plants
Module
inverters
�•
50-180W, each
panel
has its
own
inverter
enabling
optimal MPPT
�•
lower
efficiency,
difficult
maintenance
�•highercost/kWp
String
(Multi)inverters
�•
1.5 -
5 kW, typical
residential
application�•
each
string
has its
own
inverter enabling
better
MPPT
�•
the strings
can
have different
orientations�•Three-phase
inverters
for
power < 5kW
43
Topologies for PV inverters
�•
The question of having a DC-DC converter or not is first of all related to the PV string configuration.
�•
Having more panels in series and lower grid voltage, like in US
and Japan, it is possible to avoid the boost function with a dc-dc converter. Thus a single stage PV inverter can be used leading to higher efficiencies.
PVInverters
with DC-DCconverter
without DC-DCconverter
with isolation
without isolation
on the LF side
on the HF side
with isolation
without isolation
So
lar
En
erg
y
4444
Both
technologies
are
on
the market! Efficiency
93-95%
DC
ACGridPV
Array
DC
DC
On
low
frequency
(LF) side
DC
ACGridPV
Array
DC
AC
AC
DC
On
high
frequency
(HF) side
PV inverters with boost converter and isolation
Boosting inverter with HF trafo based on FB boost converter [2] Boosting inverter with LF trafo based on boost converter
4545
DC
DC
DC
ACGridPV
Array
Transformerless PV inverters with
boost
�•High
efficiency
(>95%)�•Leakage
current
problem�•Safety
issue
�•Efficiency
> 96%�•Extra
diode to bypass
boost
when
Vpv
> Vg�•Boost
with
rectified
sinus reference
�•Time sharing
configuration[3]
�•FB inverter + boost�• Typical
configuration
[1]
4646
High efficiency topologies derived from H-bridgeHERIC (Sunways)[7] - max
= 98%
Two 0 output voltage states possible: S+ and D- = ON and S- and D+ = ONThe switching ripple in the current equals 1x switching frequency high filtering neededVoltage across filter is unipolar low core losses
VPE is sinusoidal has grid frequency component low leakage current and EMIHigh efficiency 98% due to no reactive power exchange as reported by Photon Magazinefor Sunways
AT series 2.7 �–
5 kW single-phase
4747
Control Structure Overview
PV PanelsString
dc-dcboost
LC LLow pass
filterC
VPV
IPV
+
-
L
N
Trafo&
G rid
Anti-IslandingProtections
Grid /PV plant M onitoring
Ig
Vg
dc-acPW M -V SI
VdcPWM PWM
M PPT
Active filtercontrol
Grid support(V ,f,Q )
Ancillary functions
PV specific functions
Basic functions (grid conencted converter)
CurrentControl
VdcControl
M icroGridControl
G ridSynchronization
Basic functions �–
common for all grid-
connected invertersGrid current control
THD limits imposed by standardsStability in case of grid impedance variationsRide-through grid voltage disturbances (not required yet!)
DC voltage controlAdaptation to grid voltage variationsRide-through grid voltage disturbances (optional yet)
Grid synchronizationRequired for grid connection or re-connection after trip.
PV specific functions �–
common for PV invertersMaximum Power Point Tracking �–
MPPT
Very high MPPT efficiency in steady state (typical > 99%)Fast tracking during rapid irradiation changes (dynamical MPPT efficiency)Stable operation at very low irradiation levels
Anti-Islanding �–
AI as required by standards (VDE0126, IEEE1574, etc)Grid Monitoring
Operation at unity power factor as required by standardsFast Voltage/frequency
detection
Plant MonitoringDiagnostic of PV panel arrayPartial
shading
detection
Ancillary
Support �–
(future?)Voltage
ControlFrequency
controlFault
Ride-throughQ compensationDVR
48
Ad
van
ced
In
terc
on
nect
ion
6. Conclusions
49
Co
ncl
usi
on
sConclusions
Renewable Energy Systems (RES) Solutions for the futureIncrease power production close to the consumption placeMaybe decrease the power volume in the transmission level and will make the central grid control very complex (long term).Should be able to run in on-grid and off-grid modesAncillary functions must be included to avoid grid instability and blackout. Wind Turbines the fastest growingPower Converters & Control
Include MPPT functionsProvide ride-through capabilitiesIntelligent grid connection
Grid impedance estimation Monitoring and advanced diagnosis
Power Electronics -> Key Enabling Technology for grid integration of RES
50
Ack
no
wle
dg
em
en
t
Acknowledgement
Prof. Remus Teodorescu, Aalborg University, Denmark
Prof. Pedro Rodriuguez, UPC, Spain/Aalborg University, Denmark
Prof. Marco Liserre, University of Bari, Italy
Supported with material to these slides !!!
51
Deadline : May 1, 2010
52