The roles of electricity and hydrogen storage in a low ... - iea-etsap

Role of electricity and hydrogen storage in low carbon
energy system – Modelling in Temporal MARKAL model
Ramachandran Kannan
King’s College London
International Energy Workshop
3 July 2008
Outline
• UK MARKAL models
• DfT Hydrogen infrastructure project
• Flexible time slice in MARKAL
• Strengths and challenges
• Insights from temporal model
• Concluding remarks
2

The UK MARKAL model
• The first UK MARKAL model was developed for the 2003 Energy
White Paper
• Extensively updated during 2006-08 through the UK Energy
Research Centre (UKERC)
• Development and application of the model is maintained by a
consortium of UK research organizations
• King's College London
• AEA Energy and Environment
• Policy Studies Institute
• Imperial College
• University of Oxford
• Objectives of the consortium
• To maintain the model's transparency, peer review and open access
• To ensure that in iterative updates there is only one core model
structure and data-base of the UK MARKAL model family, thus
avoiding competing models
Application of the UK MARKAL model
• DTI: Energy White Paper 2007
• DfT: State-of-art modelling of hydrogen infrastructure
(with flexible time slice and GIS link)
• SuperGen: UKSHEC1 Hydrogen visions
• WWF 2050 Vision (80% CO2 reduction including international aviation)
• UK-Japan Low Carbon Scenarios
• UKERC Energy 2050 scenarios
• OFGEM: Electricity networks scearnio
• SuperGen: UKSHEC2 Plus
• SuperGen: Bioenergy
• TSEC (Towards sustainable energy economy) - Biosys
• EON

DfT Horizon project
• Five projects explored different aspects of the practicality &
timing of the introduction of the infrastructure required to
support hydrogen-fuelled vehicles funded by the
Department for Transport
• A state-of-the-art modelling of hydrogen infrastructure
development for the UK: Geographical, temporal and
technological optimisation modelling
• Methodology/tool: MARKAL energy system model
• A soft link to GIS database
• An improved temporal representation
• Full reports are available from
http://www.dft.gov.uk/pgr/scienceresearch/futures/horizons/june08
Temporal-MARKAL: Flexible time slice
• Electricity & heat are tracked seasonally and diurnally.
• By default six time slices in the standard MARKAL
• Electricity load by six day/night and seasonal splits
• Heat by three seasonal splits
• In flexible time-slice, user specify the number of time-slices
6

How many time slice can to be chosen?
• Depends on variations in electricity-driven energy service demands & energy resources
• UK experience:
GW
60
55
50
45
40
Actual electrical load demand in 2006
Domestic hot-water demand
35
30
25
Winter
Spring
Summer
Autumn
• Number of time slices - 20 Annual time periods from the original six
Diurnal
Seasonal
1. Morning: 6:00 – 9:00 (D1)
1. Winter: December – February (S1)
2. Daytime: 9:00 – 16:00 (D2)
2. Spring: March – May (S2)
3. Evening peak: 16:00 – 20:00 (D3)
3. Summer: June – August (S3)
4. Late evening: 20:00 – 23:00 (D4)
4. Autumn: September – November (S4)
5. Night storage: 23:00 – 6:00 (D5)
7
20
00:00 03:00 06:00 09:00 12:00 15:00 18:00 21:00
Strengths – Power sector
• Flexibility to model electricity demand profile thereby get a
better fit on electricity demand
60
Winter Electric Demand (2006)
60
Winter Electric Demand (2006)
50
50
GW
40
GW
40
30
Night time
demand
(22:30 - 6:30)
Day time demand
(6:30 – 22:30)
30
D5
Night Storage
(23 - 7)
D1
Mornin
g
(6 - 9)
D2
Day
(9 -16)
D3
Peak
(16 - 20)
D4
Evening
(20 - 23)
20
00:00 04:00 08:00 12:00 16:00 20:00
20
00:00 04:00 08:00 12:00 16:00 20:00
Actual
MARKAL
Actual
Temporal MARKAL
8

Strengths – Power sector
• Better convergence of actual versus MARKAL electric demand
margins, i.e. levelised electric peak and actual peak demand.
• More justifiable electric reserve margin
70
Winter Electric Demand
70
Winter Electric Demand (2006)
GW
Actual
MARKAL
60
reserve
reserve
Total capacity
capacity
installed
capacity
50
Night demand
Actual peak
(22:30 - 6:30)
demand
40
Levelised
Day demand
day demand
(6:30 – 22:30)
30
07:00 10:00 13:00 16:00 19:00 22:00 01:00 04:00
GW
60
50
40
30
Total
installed
capacity
Actual reserve
capacity
Actual peak
demand Levelised
peak demand
00:00 04:00 08:00 12:00 16:00 20:00
MARKAL
reserve
capacity
Actual Winter
MARKAL Winter
Actual
Temporal MARKAL
9
Strengths – Demand & Resources
• Detailed modelling of fluctuating energy service demands
• Seasonal representation of intermittence renewable energy
sources
Wind resources
Residential lights
10

Challenges
• Storage is still limited to ONE period (YNITE).
Therefore a period to period storage is not
possible to deal with intermittence renewable
energy sources
• Data Issues
• Diurnal/seasonal break-up of energy service
demands is not commonly available though their
electricity demand profile can be found
• Similar issues on availability of data for energy
resources
11
Electricity storage
Electricity storage: Base
160
140
120
PJ
100
80
60
40
20
0
2000 2010 2020 2030 2040 2050
Storage heaters Plug-in hybrid
Hydrogen storage Pumped hydro
PJ
Electricity storage: CO2
160
140
120
100
80
60
40
20
0
2000 2010 2020 2030 2040 2050
Storage heaters Plug-in hybrid
Hydrogen storage Pumped hydro
• On average, the system chooses about 7 - 10% of
electricity demand as storage
• Demand side storage is preferred, partly due to their
lower costs
12

Electricity storage in 2050
200
Electricity storage
150
100
50
0
Base
Base-1
Base-1-D
Base-1-N
Base1-H
Base-2
Base-3
Base-3-H
Base-3-W
Base-4
Base-4-H
CO2
CO2-SLT
CO2-1
CO2-1-D
CO2-1-N
CO2-1-H
CO2-2
CO2-3
CO2-3-H
PJ
CO2-4
CO2-4-H
Storage heaters Plug-in hybrid Hydrogen storage Pumped hydro
Base
A business as usual case
13
CO 2
60% CO 2
reduction from 2000 level by 2050
Base-** / CO 2
-**
1: No plug-in hybrid vehicles
2: No night storage heaters
3: No plug-in or night storage heaters
4: No storage
H: night storage hydrogen production enabled
W: alternate wind profile enabled
Generation by plant type
• Better utilization of base load plants
2000
Genertion mix by plant type: Base
2000
Genertion mix by plant type: CO2
1500
1500
PJ
1000
PJ
1000
500
500
0
2000 2010 2020 2030 2040 2050
0
2000 2010 2020 2030 2040 2050
Base non-base CHP Storage Demand
Base non-base CHP Storage Demand
14

Power system balancing
• Electricity storage is important as a power
system balancing mechanism (even though a 20
time period model still represent aggregation).
68
Seasonal electricity supply in 2050 - CO2 scenario
68.0
Seasonal electricity supply in 2050 - CO2-4
GW
64
60
56
52
.
48
44
40
6.00 9.00 12.00 15.00 18.00 21.00 24.00 3.00
Winter Autumn Spring summer
With electricity storage
GW
64.0
60.0
56.0
52.0
.
48.0
44.0
40.0
6.00 9.00 12.00 15.00 18.00 21.00 24.00 3.00
Winter Autumn Spring summer
Without electricity storage
15
Power generation mix in 2050
2000
Electricity generation mix by fuel and plant type
1500
PJ
1000
500
0
Base
Base-1
Base-1-D
Base-1-N
Base1-H
Base-2
Base-3
Base-3-H
Base-3-W
Base-4
Base-4-H
CO2
CO2-SLT
CO2-1
CO2-1-D
CO2-1-N
CO2-1-H
CO2-2
CO2-3
CO2-3-H
CO2-4
CO2-4-H
Coal Coal CCS Nuclear Gas Hydro Wind Bio & waste
Imports Marine Gas (CHP) Other (CHP) Storage
16

Concluding remarks
• Flexible time slice is successfully implemented in the
UK MARKAL model
• Temporal MARKAL enhances the depiction of
electricity demand profile and representation of
renewable energy resources
• Electricity storage is inevitable for system balancing.
However, with ONE time period storage option in
MARKAL, a detailed modelling of intermittence
renewable is inadequate
• Other models (e.g. electricity dispatch model) could be
used for intermittency analysis
17
Thanks!
For further information
• Contact:
• Ramachandran Kannan - r.kannan@kcl.ac.uk
• Neil Strachan - neil.strachan@kcl.ac.uk
• UK MARKAL documentation
http://ukerc.ac.uk/ResearchProgrammes/EnergySystemsandModelling/ESMMARKALDocs08.
aspx
• DfT hydrogen infrastructure reports
http://www.dft.gov.uk/pgr/scienceresearch/futures/horizons/june08
18