Hydrogen pipeline versus power line

1
Small-Scale Wind for Hydrogen Production for
Rural Power SuppliesHyLink System at Totara
Valley
MUCER Energy Postgraduate Conference Wellington
3-5 June 2008

presented by Peter Sudol
(Massey University)
2
(No Transcript)
3
Totara Valley

• Demonstration site for Massey University
  and Industrial Research Limited on
  distributed generation Aims - design a
  renewable hybrid micro-power system at
  the end of 11 kV distribution line -
  provide network support

4
HyLink System
Demonstration on hydrogen as a means of
balancing and transporting the fluctuating
wind powerSystem implementationby IRLSystem
analysis by Massey UniversityMassey
Universitys 2.2 kWwind turbine incl. control
system will be used in conjunction with a larger
electrolyser currently being developed at IRL.

5
Characteristic HyLink System Power Line
Initial cost incl. labour NZ55,000 - current configuration - incl. pipeline mole ploughing NZ60,000 -NZ100,000 - underground wiring requires a trench - overhead wiring complicated due to difficult terrain
Cost of conversion devices NZ17,000 - electrolysis setup NZ16,000 - fuel cell system 2 x NZ2,500 - step up and step down transformers
Energy loss at conversion devices ?e/conv 60 - converter/electrolyser subsystem ?pemfc/inv 35 (electr.) - fuel cell/inverter subsystem 2 x 200 W power loss - power consumption at both transformers
Lifetime 50 years - MDPE gas pipeline 4,000 operational hrs - ReliOn PEM fuel cell 10,000 operational hrs - PEM electrolyser 60 years
Energy Storage Hydrogen pipeline/tank - easy to scale up Batteries - expensive for large-scale storage
6
Hylink in the IRL Laboratory
Hydrogen was stored in 150m MDPE pipeline located
in a container filled with sand, outside of the
lab.
Electrolysis setup
7
Alkaline Fuel Cell DCI 1200 Setup
The electricity produced was used to charge
batteries or was inverted to the grid.
Source IRL
8
Electrolyser Stack Connection
positive electrical potential
hydrogen pressure meter
water and oxygen outlet
hydrogen outlet
negative electrical potential
water inlet
Distilled water is pumped just through the anode
compartment (oxygen side) of the electrochemical
cells which is not pressurised.
9
Lynntech Electrolyser Stack
catalysed membrane
metal flow field
Source Lynntech Industries
active area of 33 cm2
The right stainless steel endplate (
electronics) was used for a previous application
and was replaced by a titanium endplate.
10
Electrolyser Stack - VI Curves
At higher stack temperature there is a higher
electr. current flow (?higher hydrogen
production) at the same voltage due to improved
reaction kinetics.
11
System Efficiency Estimation

• Electrolyser ? 65.3 ? electricity
  (heat)/hydrogen conversion efficiency
• (at a current flow of 23,5 A)
• - Not considered hydrogen pressure energy
  output, power consumption of the 12W water pump,
  heat transfer with circulating water

• Alk. FC DCI 1200 ? 41.1 ?
  hydrogen/electricity conversion efficiency
• (at 650 W electrical power output)
• - Not considered thermal energy output
  (approximately 20 ? combined heat
• and power efficiency over 60)

The above efficiencies were calculated using the
lower heating value (LHV) of hydrogen. Hydrogen
production/consumption was estimated by measuring
the pressure increase/drop in the pipe.
12
Proven Wind Turbine

• Rated power output 2.2 kW
• Zebedee furl ( cone) system
• allows for dynamic balance
• between the rpm and the pitch
• of the airfoil.
• ? During stormy winds turbine
• doesnt stop, instead, keeps
• generat. at nearly rated power.

Drawing Proven Energy Limited
13
Air-X 400 W
14
Wind Power Control and Electrolysis Container
3 x 48 W solar panels for additional battery
charging
Distilled water tank for the electrolysis
Hydrogen pipeline - the top riser
15
Source IRL
16
Electrolyser in the Container
Flash arrestor
Recombiner
Deionisation column
Electrolyser stack
Dehydration unit
Circulating water reservoir
Water pump
Source IRL
17
Hylink Transition to Totara Valley
Pipeline mole ploughing (60 cm deep)
Electrofusion joint between the pipeline sections
MDPE Internal diameter 16 mm Outer
diameter 21 mm Wall thickness 2.5
mm Length 2 km ?
Volume 402 L
Welder for electrofusion
Source IRL
18
Fuel Cell Connection in the Woolshed
Pressure sensor
PEM fuel cell ReliOn Independence 1000 (J48C)
IRL controller
IRL grid-connected inverter
48 V gel battery bank
The batteries power the control and data logging
equipment as well as provide a necessary buffer
for the fuel cell and inverter. Operating PEMFC
supplies the inverter and the controller as well
as charges the batteries.
Source IRL
19
Hydrogen Diffusion Rate Estimation

• The pipeline was pressurised with 4.1 barg
  hydrogen and
• then the pressure drop was recorded.
• Result
• Hydrogen loss 42.5 kPa/week ? 7.5mol/week ? 15
  g/week
• ? 0.5 kWh/week at LHV ? 3 W
• Currently, the fuel cell operates at pipeline
  pressure
• between 1 barg and 2 barg, so at average 2.5 bar
  abs.
• Using ? 1.5
  W mean power loss due
  
• 
  to H2 diffusion through pipe
• 
  walls during fuel cell operation
  
  

20
Hydrogen Permeability through PE - Comparison
Industry (at 23C)
Massey University (at 20C)
Totara Valley (at 10C)
General rule of thumb for Arrhenius equation
for every 10C increase the reaction rate
doubles.
Or EP and P0 can be estimated by measuring P at
different T and solving
21
Frictional Pressure Drop Estimation at Fuel
Cells max. Output

• According to the manufacturer, the fuel cell
  consumes at 1 kW
• 15 stdL H2 /min ? mean H2 velocity is 1.24 m/s
• Due to the gas flow is
  laminar, and hence, the
• friction factor f independent of roughness
  .
• Then the frictional pressure drop can be
  calculated using the Darcy-Weisbach equation as
  follows
• Considering that the fuel cell requires low H2
  pressure for operation, the calculated pressure
  drop can be neglected.

22
HOMER Simulation of the current HyLink System
Configuration
Selected Results
23
Data Inputs

• Batteries task is not to store energy to meet
  communitys load requirements. They cover the
  system internal electricity needs, and PV panels
  can be thought as the power source for that. For
  this reason batteries as well as PV panels were
  excluded from the simulation.
• Wind resource data was taken from the NASA
  website, however the four wind parameters
  (Weibull shape factor etc.) derive from the
  previous study at Massey University.
• The average of one of the eight monitored sites
  at Totara Valley was used as the primary load
  data.
• Furthermore, factual and not projected data was
  used e.g. for the ReliOn fuel cell the lifetime
  of 4,000 operational hours and not 40,000
  operational hours.

24
HyLink System Schematics used in HOMER
grid-connected
stand-alone
25
Providing Back-up Power for Peak Loads (May)
- Grid purchase capacity constrained at 2.3
kW - max. hourly peak load throughout a
year 3.3 kW - 1 kW fuel cell
26
Providing Back-up Power for Peak Loads (July)
- Grid purchase capacity constrained at 2.3
kW - max. hourly peak load throughout a
year 3.3 kW - 1 kW fuel cell
27
Providing Back-up Power for Peak Loads
Due to small system configuration, esp. wind
turbine/electrolyser, very dependent on the
prevailing wind conditions.
28
Daily Pipeline Filling Process Fluctuations due
to changing Wind
29
Scenario for HyLink with added Massey
Universitys 2.2 kW Wind Turbine and IRLs 1 kW
Electrolyser
The previous capacity shortage on 24th July is
compensated due to improved system response.
30
Scenario for HyLink with added Massey
Universitys 2.2 kW Wind Turbine and IRLs 1 kW
Electrolyser
31
Scenario for HyLink with added Massey
Universitys 2.2 kW Wind Turbine and IRLs 1 kW
Electrolyser
32
Outcomes

• Low durability and high replacement cost of
  electrochemical conversion devices represent the
  main barrier in introducing the HyLink system
• Small-sized system very dependent on the
  prevailing wind conditions low energy buffer
  capability
• The 36-efficient fuel cell/inverter subsystem
  consumes the full pipe content (3.3kWh at 3bar
  pressure difference) in ca. 1 hr at 1kW ac
  output.
• The wind turbine/electrolyser subsystem needs
  9hrs at its rated power (80 stdL/hr, 360 W) to
  provide this hydrogen content at optimal wind
  conditions
• Hence, the fuel cell/inverter efficiency (36
  electr.) constrains the overall system
  performance and the small wind turbine/electrolyse
  r size slows the systems response.

33
General Outcome

• Successful demonstration of a new energy concept
  in operation since May 2008
• The HyLink system reveals barriers and
  opportunities of hydrogen based energy systems.
• The HyLink system proves, that an energy carrier
  can be produced from a renewable resource high
  efficiently.
• The HyLink system proves that this energy carrier
  can be transported via cheap pipelines.
• The HyLink system proves that this energy carrier
  can be converted to electricity high efficiently
  (not Carnot Cycle constrained), carbon neutral
  and noiseless in fuel cells.

34
Acknowledgements

• Prof. Ralph Sims (Massey University)
• Attilio Pigneri (Massey University)
• Steve Broome (IRL)
• Edward Pilbrow and Eoin McPherson (IRL)
• Jim Hargreaves (Massey University), Phil Murray,
  Mark Carter
• Totara Valley residents
• and many others at Massey