Energy Storage #3 - The University of Texas at Austin

Download Report

Transcript Energy Storage #3 - The University of Texas at Austin

Energy Storage
• In the past 2 classes we have discussed battery technologies and how their
characteristics may or may not be suitable for microgrids.
• Batteries are suitable for applications where we need an energy delivery
profile. For example, to feed a load during the night when the only source is PV
modules.
• However, batteries are not suitable for applications with power delivery
profiles. For example, to assist a slow load-following fuel cell in delivering
power to a constantly and fast changing load.
• For this last application, two technologies seem to be more appropriate:
• Ultracapacitors (electric energy)
• Flywheels (mechanical energy)
• Other energy storage technologies not discussed in here are superconducting
magnetic energy storage (SMES – magnetic energy) and compressed air (or
some other gas - mechanical energy)
1
© Alexis Kwasinski, 2012
Power vs. energy delivery profile technologies
• Ragone chart:
• More information and charts can be found in Holm et. al., “A Comparison of
Energy Storage Technologies as Energy Buffer in Renewable Energy Sources
with respect to Power Capability.”
2
© Alexis Kwasinski, 2012
Power vs. energy delivery profile technologies
3
© Alexis Kwasinski, 2012
Electric vs. Magnetic energy storage
• Consider that we compare technologies based on energy density (J/m3)
[ Energy]  [Work ]  [ F ][d ]  Nm  J
[ Energy density ] 
J
Nm N

 2  Pa
3
3
m
m
m
• Plot of energy density vs. length scale (distance between plates or air gap):
University of Illinois at Urbana-Champaign
ECE 468 (Spring 2004)
• Hence, magnetic energy storage (e.g. SMES) is effective for large scale
systems (higher power)
4
© Alexis Kwasinski, 2012
Ultracapacitors
• Capacitors store energy in its electric field.
• In ideal capacitors, the magnitude that relates the charge generating the
electric field and the voltage difference between two opposing metallic plates
with an area A and at a distance d, is the capacitance:
C
• In ideal capacitors:
Q
V
A
d
• Equivalent model of real standard capacitors:
C 
ESR  Rw 
5
© Alexis Kwasinski, 2012
1
 2 Rl C 2
Ultracapacitors
• Ultracapacitors technology: construction
• Double-layer technology
http://www.ultracapacitors.org/img2/ultraca
pacitor-image.jpg
•Electrodes: Activated carbon (carbon cloth, carbon black, aerogel carbon,
particulate from SiC, particulate from TiC)
• Electrolyte: KOH, organic solutions, sulfuric acid.
6
© Alexis Kwasinski, 2012
Ultracapacitors
• Ultracapacitors technology: construction
Traditional standard
capacitor
The charge of ultracapacitors, IEEE
Spectrum Nov. 2007
Double layer
capacitor
(ultracapacitor)
Ultracapacitor with carbon
nano-tubes electrodes
A
d
• Key principle: area is increased and distance is
decreased
C 
• There are some similarities with batteries but there are
no reactions here.
7
© Alexis Kwasinski, 2012
Ultracapacitors
• Ultracapacitors technology: construction
www.ansoft.com/firstpass/pdf/CarbonCarbon_Ultracapacitor_Equivalent_Circuit_Model.pdf
8
© Alexis Kwasinski, 2012
Ultracapacitors
• Some typical Maxwell’s ultracapacitor packages:
www.ansoft.com/firstpass/pdf/CarbonCarbon_Ultracapacitor_Equivalent_Circuit_Model.pdf
• At 2.7 V, a BCAP2000 capacitor can store more than 7000 J in the volume of
a soda can.
• In comparison a 1.5 mF, 500 V electrolytic capacitor can store less than 200 J
in the same volume.
9
© Alexis Kwasinski, 2012
Ultracapacitors
• Comparison with other capacitor technologies
www.ansoft.com/firstpass/pdf/CarbonCarbon_Ultracapacitor_Equivalent_Circuit_Model.pdf
10
© Alexis Kwasinski, 2012
Ultracapacitors
• Charge and discharge:
• With constant current, voltage approximate a linear variation due to a very
large time constant:
• Temperature affects the output (discharge on a constant power load):
www.ansoft.com/firstpass/pdf/CarbonCarbon_Ultr
acapacitor_Equivalent_Circuit_Model.pdf
11
© Alexis Kwasinski, 2012
Ultracapacitors
• Aging process:
• Life not limited by cycles but by aging
• Aging influenced by temperature and cell voltage
• Overtime the materials degrade, specially the electrolyte
• Impurities reduce a cell’s life.
Linzen, et al., “Analysis and Evaluation of Charge-Balancing
Circuits on Performance, Reliability, and
Lifetime of Supercapacitor Systems”
12
© Alexis Kwasinski, 2012
Ultracapacitors
• Power electronic interface:
• It is not required but it is recommended
• It has 2 purposes:
• Keep the output voltage constant as the capacitor discharges (a
simple boost converter can be used)
• Equalize cell voltages (circuit examples are shown next)
13
© Alexis Kwasinski, 2012
Ultracapacitors
• Model (sometimes similar to batteries)
Mierlo et al., Journal of Power Sources 128
(2004) 76–89
http://www.ansoft.com/leadinginsight/pdf/High%20P
erformance%20Electromechanical%20Design/Ultrac
apacitor%20Distributed%20Model%20Equivalent%2
0Circuit%20For%20Power%20Electronic%20Circuit
%20Simulation.pdf
Ultracapacitors for Use in Power Quality and
Distributed Resource Applications, P. P. Barker
14
© Alexis Kwasinski, 2012
Flywheels
• Energy is stored mechanically (in a rotating disc)
Motor
Generator
Flywheels Energy
Systems
15
© Alexis Kwasinski, 2012
Flywheels
http://www.vyconenergy.com
http://www.pentadyne.com
16
© Alexis Kwasinski, 2012
Flywheels
• Kinetic energy:
1 2
Ek  I 
2
where I is the moment of inertia and ω is the angular velocity of a rotating disc.
I   r 2 dm
• For a cylinder the moment of inertia is
1
I  r 4 a 
2
• So the energy is increased if ω increases or if I increases.
• I can be increased by locating as much mass on the outside of the disc as
possible.
• But as the speed increases and more mass is located outside of the disc,
mechanical limitations are more important.
17
© Alexis Kwasinski, 2012
Flywheels
• Disc shape and material: the maximum energy density per mass and the
maximum tensile stress are related by:
em  K max / 
• Typically, tensile stress has 2 components: radial stress and hoop stress.
18
© Alexis Kwasinski, 2012
Flywheels
• Since
(1)
em  K max / 
and
1 2
Ek  I 
2
(2)
" I  r 2m"
(3)
and
then, from (2) and (3)
1
1
em  r 2 2  v 2
2
2
(4)
So, replacing (1) in (4) it yields
vmax 
19
2K max

© Alexis Kwasinski, 2012
Flywheels
• However, high speed is not the only mechanical constraint
• If instead of holding output voltage constant, output power is held constant,
then the torque needs to increase (because P = Tω) as the speed decreases.
Hence, there is also a minimum speed at which no more power can be
extracted
vmax
V

• If
r
vmin
and if an useful energy (Eu) proportional to the difference between the disk
energy at its maximum and minimum allowed speed is compared with the
maximum allowed energy (Emax) then
Eu/Emax
Eu
Vr2  1

Emax
Vr2
Vr
Vr
20
Bernard et al., Flywheel Energy
Storage Systems In Hybrid And
Distributed Electricity Generation
© Alexis Kwasinski, 2012
Flywheels
• In order to reduce the friction (hence, losses) the disc is usually in a vacuum
chamber and uses magnetic bearings.
Bernard et al., Flywheel Energy
Storage Systems In Hybrid And
Distributed Electricity Generation
• Motor / generators are typically permanent magnet machines. There are 2
types: axial flux and radial flux. AFPM can usually provide higher power and are
easier to cool.
Bernard et al., Flywheel Energy Storage Systems In Hybrid And
Distributed Electricity Generation
21
© Alexis Kwasinski, 2012
Flywheels
• Simplified dynamic model
• Typical outputs
Flywheels Energy
Systems
22
© Alexis Kwasinski, 2012