the future of electricity: a market with costs of zero ... › conference › iaee2017 › files ›...
Post on 28-Jun-2020
1 Views
Preview:
TRANSCRIPT
© Imperial College Business School
The Future of Electricity:
A Market with Costs of Zero?
1Richard Green, IAEE European Conference, 2017
Marginal
Supply
2
GW
£/MWh
Nuclear CCGT OCGT
Marginal Cost
Prices reflect Marginal Costs
Demand and Supply
3
GW
£/MWh
Nuclear CCGT OCGT
Marginal Cost
4
Annual Generation Costs
£/kW-yearOpen Cycle Gas Turbine
Combined Cycle
Gas Turbine
Nuclear
Hours per yearOCGT
cheapest
CCGT
cheapest
Nuclear
cheapest
5
Load-duration and Capacity
OCGT
cheapest
CCGT
cheapest
Nuclear
cheapest
Hours per year
GW
Nuclear
CCGT
OCGT
6
Generator Costs & Revenues
£/kW-yearOpen Cycle Gas Turbine
Combined Cycle
Gas Turbine
Nuclear
Hours per yearOCGT
cheapest
CCGT
cheapest
Nuclear
cheapest
0
10
20
30
40
50
2009 2010 2011 2012 2013 2014 2015 2016
Electricity Gas Coal
per MWh of electricity
© Imperial College Business School
British Energy Prices
7
£/MWh
Source: ElectricInsights.co.uk
0
10
20
30
40
50
2009 2010 2011 2012 2013 2014 2015 2016
Electricity Gas Coal
per MWh of electricity
© Imperial College Business School
British Energy Prices
8
£/MWh
Source: ElectricInsights.co.uk
© Imperial College Business School 9
German Energy Prices:
a market working well?
0%
20%
40%
60%
80%
100%
120%
140%
€0
€10
€20
€30
€40
€50
€60
€70
2002 2004 2006 2008 2010 2012 2014
Gas price / 45%
Electricity price
Prices
(€/MWh)
Installed capacity
(relative to peak load)
Source: Green and Staffell, Oxrep, 2016
Austrian
© Imperial College Business School 10
German Energy Prices:
The Merit Order Effect
0%
20%
40%
60%
80%
100%
120%
140%
€0
€10
€20
€30
€40
€50
€60
€70
2002 2004 2006 2008 2010 2012 2014
Gas price / 45%
Electricity price
Prices
(€/MWh)
Installed capacity
(relative to peak load)
Source: Green and Staffell, Oxrep, 2016
Austrian
© Imperial College Business School 11
Renewables in a Power Market
The long-run adjustment
Demand and Supply
13
GW
£/MWh
Nuclear CCGT OCGT
Marginal Cost
The merit order effect
14
Capacity and Load
OCGT CCGT Nuclear Hours per year
GW
Nuclear
CCGT
OCGT
After renewables
0
10
20
30
40
50
60
70
80
90
1999 2001 2003 2005 2007 2009 2011 2013 2015
Solar
Wind
Conventional
Peak Demand
Great Britain
© Imperial College Business School
Generating Capacity
15
GW
Source: Digest of
UK Energy Statistics
0%
10%
20%
30%
40%
50%
60%
70%
80%
1999 2001 2003 2005 2007 2009 2011 2013 2015
UK-wide, including Northern Ireland
© Imperial College Business School
Generators’ Load Factors
16Source: Digest of UK Energy Statistics
Coal
CCGT
0
10
20
30
40
50
60
70
80
90
1999 2001 2003 2005 2007 2009 2011 2013 2015
Solar
Wind
Conventional
Peak Demand
Great Britain
© Imperial College Business School
Capacity and Peak Demand
17
GW
Source: Digest of
UK Energy Statistics
-600
-500
-400
-300
-200
-100
0
100
200
0
10
20
30
40
50
60
70
80
Nuclear Biomass Gas Coal Hydro Peaking Imports Wind Solar D P R
A Low-Carbon Christmas
£/MWh
GW
Day-ahead Price
25 December 2016 26 December 2016 27 December 2016
Source: ElectricInsights.co.uk
-600
-500
-400
-300
-200
-100
0
100
200
0
10
20
30
40
50
60
70
80
Nuclear Biomass Gas Coal Hydro Peaking Imports Wind Solar D P R
A Low-Carbon Christmas
£/MWh
GW
Real-time Price
Day-ahead Price
25 December 2016 26 December 2016 27 December 2016
Source: ElectricInsights.co.uk
Supply and Demand
£/MWhDemand
GW
Marginal Cost
Price
-100
0
100
200
300
400
500
600
700
800
A volatile market
April May June July Aug Sept Oct Nov Dec Jan Feb Mar
£/MWhDay-ahead Electricity Price (half-hourly)
Great Britain, April 2016-March 2017
Source: ElectricInsights.co.uk
-100
0
100
200
300
400
500
600
700
800
© Imperial College Business School
Day-ahead Prices in 2016
22
£/MWh
Source: ElectricInsights.co.uk
hours0 1000 2000 3000 4000 5000 6000 7000 8000
-100
0
100
200
300
400
500
600
700
800
Surplus over annual average fuel cost
© Imperial College Business School
Day-ahead Prices in 2016
23
£/MWh
Source: ElectricInsights.co.uk
hours0 1000 2000 3000 4000 5000 6000 7000 8000
Black bars are for coal-fired plant, blue for gas
-50
-25
0
25
50
75
100
125
150
Surplus over annual average fuel cost
© Imperial College Business School
Day-ahead Prices in 2016
24
0 1000 2000 3000 4000 5000 6000 7000 8000
£/MWh
Source: ElectricInsights.co.uk
hours
Black bars are for coal-fired plant, blue for gas
-50
0
50
100
150
Surplus over annual average fuel cost
© Imperial College Business School
Day-ahead Prices in 2016
25
0 1000 2000 3000 4000 5000 6000 7000 8000
£/MWh
Source: ElectricInsights.co.uk
0%
50%
100%
Cumulative surplus
Gas
Coal
hours
Black bars are for coal-fired plant, blue for gas
0
10
20
30
40
50
0 2000 4000 6000 8000
2016 Demands
© Imperial College Business School
2016 Load-duration curves
More need for peaking plant?
26
40
42
44
46
48
50
52
0 100 200 300 400 500
(Detail)
Gross
Net
Gross
GW
Hours/yearSource: ElectricInsights.co.uk
0
10
20
30
40
50
0 2000 4000 6000 8000
2016 Demands
© Imperial College Business School
2016 Load-duration curves
More need for peaking plant?
27
40
42
44
46
48
50
52
0 100 200 300 400 500
(Detail)
Gross
Net
Gross
Net
Source: ElectricInsights.co.uk
GW GW
Hours/year Hours/year
0
50
100
150
200
250
300
350
400
450
0 1 2 3 4 5 6 7 8
Usage over 17 years of demand and weather data
More risk for peaking plants?
28
No Renewables
No Renewables
Current Renewables
(15 GW Wind, 12 GW PV)
Double Renewables
(30 GW Wind, 24 GW PV)
Current
Renewables
Double
Renewables
100-hour mean annual running time 20-hour meanGW
Hours/year
• Scheduling model with start costs and no-load costs
– Capacity assumed infinitely divisible
• Reserve requirement of 3GW in all periods
• Demands from GB load profiles, scaled to common base
• Demand reduction linear in price above £40/MWh
• Renewable profiles for wind and PV from Iain Staffell and
Stefan Pfenninger: renewables.ninja
• Assume 15 GW onshore wind
• Assume 50 GW offshore wind
• Assume 15 GW Solar PV
The Model With No Name
© Imperial College Business School
A market of the future…
29
-100
0
100
200
300
400
500
-10
0
10
20
30
40
50
Gas OCGT Wind Solar Spilled Demand Price (RHS)
Week 7 of “2010”
© Imperial College Business School
A simulated future
30
GW £/MWh
-100
0
100
200
300
400
500
-10
0
10
20
30
40
50
Gas OCGT Wind Solar Spilled Demand Price (RHS)
Week 7 of “2010”
© Imperial College Business School
A simulated future
31
GW £/MWh
-100
0
100
200
300
400
500
-10
0
10
20
30
40
50
Gas OCGT Wind Solar Spilled Demand Price (RHS)
Week 44 of “2010”
© Imperial College Business School
A simulated future
32
GW £/MWh
-100
0
100
200
300
400
500
-10
0
10
20
30
40
50
Gas OCGT Wind Solar Spilled Demand Price (RHS)
Week 44 of “2010”
© Imperial College Business School
A simulated future
33
GW £/MWh
-100
0
100
200
300
400
500
-10
0
10
20
30
40
50
Gas OCGT Wind Solar Spilled Demand Price (RHS)
Week 44 of “2010”
© Imperial College Business School
A simulated future
34
GW £/MWh
overall Gas OCGT Solar Wind
Relative revenues by type of plant
© Imperial College Business School
A barrier to renewables?
35
-100
0
100
200
300
400
500
600
700
800
A volatile market
April May June July Aug Sept Oct Nov Dec Jan Feb Mar
£/MWhDay-ahead Electricity Price (half-hourly)
Great Britain, April 2016-March 2017
Source: ElectricInsights.co.uk
-10
0
10
20
30
40
50
60
70
80
A less volatile market
April May June July Aug Sept Oct Nov Dec Jan Feb Mar
€/MWhDay-ahead Electricity Price (half-hourly)
Norway, April 2016-March 2017
Source: ElectricInsights.co.uk
© Imperial College Business School 38
Renewables in an Energy Market
© Imperial College Business School 39
Supply and Demand
£/MWhDemand
GW
Marginal Cost
Price
© Imperial College Business School 40
Supply and Demand
£/MWhDemand
GW
Marginal Cost
Remaining Resource
Future Demand
Opportunity Cost
( )
Price
Available
Resource
Finn’s bathtub, from Forsund (2007) Hydropower Economics
© Imperial College Business School 41
Reservoir Levels
GWh
Time
Energy Capacity
Low price minimises
the risk of throwing
away energy
High price minimises
the risk of running out
of energy
Low
price
again
-100
0
100
200
300
400
500
600
700
-10
0
10
20
30
40
50
60
70
Nuclear Wind Solar Storage Demand Price (RHS)
Week 7 of “2010”
© Imperial College Business School
A simulated future
42
GW £/MWh
-100
0
100
200
300
400
500
600
700
-10
0
10
20
30
40
50
60
70
Wind Solar Gas OCGT Spilled Demand Price (RHS)
Week 7 of “2010”
© Imperial College Business School
A simulated future
43
GW £/MWh
-100
0
100
200
300
400
500
600
700
-10
0
10
20
30
40
50
60
70
Nuclear Wind Solar Storage Demand Price (RHS)
Week 44 of “2010”
© Imperial College Business School
A simulated future
44
GW £/MWh
-100
0
100
200
300
400
500
600
700
-10
0
10
20
30
40
50
60
70
Wind Solar Gas OCGT Spilled Demand Price (RHS)
Week 44 of “2010”
© Imperial College Business School
A simulated future
45
GW £/MWh
Thermal Market Storage Market
Average Gas Peaking Solar Wind
Revenues by type of plant
© Imperial College Business School
A more level playing field?
46
Peaking/
Storage
0
2
4
6
8
10
12
14
16
18
0
2
4
6
8
10
12
14
16
18Seasonal Storage (LHS) Peaking Storage (RHS)
8760 hours of “2010”
© Imperial College Business School
Stored Energy Levels
47
TWh
Jan Mar SepMay Jul Nov DecOctAugJunFeb Apr
GWh
10
40
30
20
0
0
2
4
6
8
10
12
14
16
18
0
2
4
6
8
10
12
14
16
18Seasonal Storage (LHS) Peaking Storage (RHS)
8760 hours of “2010”
© Imperial College Business School
Stored Energy Levels
48
Jan Mar SepMay Jul Nov DecOctAugJunFeb Apr
TWh GWh
10
40
30
20
0
-75
-50
-25
0
25
50
75
100
125
-15
-10
-5
0
5
10
15
20
25
Peaking Seasonal Price (RHS)
Week 47 of “2010”
© Imperial College Business School
Storage flows and prices
GW £/MWh
0
20
40
60
80
100
120
0
5
10
15
20
25
30
Seasonal Storage Price
8760 hours in “2010”
© Imperial College Business School
A storage-renewable market
50
GWh £/MWh
Jan Mar SepMay Jul Nov DecOctAugJunFeb Apr
• Balancing
• Uncertainty
• Transmission
• Distribution
• Inertia
• What happens with intermediate amounts of storage
© Imperial College Business School 51
What have I left out?
• Markets based on power will have volatile prices in a
high-renewable world
• Storage can smooth these prices, creating markets based
on energy
• Prices would be set to meet the energy constraint over
long periods of time
• Would we value generators on their expected energy
output?
© Imperial College Business School 52
Conclusions
top related