You are on page 1of 61

Pre-submission to the PC on Energy on 14 February 2017

Electricity Scenarios for South Africa


Presentation to the Portfolio Committee on Energy

CSIR

Cape Town, 21 February 2017

Dr Tobias Bischof-Niemz
Chief Engineer
CSIR delegation

Dr Rachel Chikwamba Group Executive: Strategic Alliances and Communication

Dr Tobias Bischof-Niemz Manager: Energy Centre

Crescent Mushwana Research Group Leader: Energy Systems

Jarrad Wright Principal Engineer: Energy Planning

Joanne Calitz Senior Engineer: Energy Planning

Mamahloko Senatla Researcher: Energy Planning

Tendani Tsedu Group Manager: Communications

Azeza Fredericks Parliamentary Liaison

2
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions

IRP Results and Least-cost Scenario

Summary

3
World:
In 2016, 124 GW of new wind and solar PV capacity installed globally

Solar PV
Global annual new capacity in GW/yr
Wind
124
120

91
57 70
76 73
71
40
56 31
30 38 Total South African
46
17 power system
7
33 (approx. 45 GW)
22 7 63
17 51 54
3 39 39 41 45
8 9 9 13 2 35
7 1 27
4 0 70 1 1 12 15 20
4 0 7 8 8

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

This is all very new: Roughly 80% of the globally existing


solar PV capacity was installed during the last five years
4
Sources: GWEC; EPIA; BNEF; CSIR analysis
World:
Significant cost reductions materialised in the last 5-8 years

Solar PV Solar PV cost


Global annual new capacity in GW/yr Wind Wind cost

120 124
100%

91
57 70
76 73
71 65%
40
56 31
30 38
46
17 Total South African
33 7
power system
7 22 7 63 (approx. 45 GW)
4 9 13 17 51 54
0 9 3 39 39 41 45 22%
0 8 1 2 35
7 1 1 27
4 0 15 20
7 8 8 12

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Subsidies Cost competitive


5
Sources: IEA; GWEC; EPIA; BNEF; CSIR analysis
South Africa:
From 2013 to 2016, 3.1 GW of wind, solar PV and CSP commissioned

Capacity
online in MW
(end of year)
+1 094
3 134

+520 Supply
1 474 Sources
+1 053 2 040
Solar PV
1 520
965 Wind
CSP
960
1 460
467 1 075
210 560
257 200

2013 2014 2015 2016 2017 2018 2019 2020

6 Notes: RSA = Republic of South Africa. Solar PV capacity = capacity at point of common coupling. Wind includes Eskoms Sere wind farm (100 MW)
Sources: Eskom; DoE IPP Office
South Africa:
In 2016, almost 7 TWh electricity produced from wind, solar PV & CSP

Annual energy
produced in TWh

6.9
Supply
2.6 Sources
4.7
Solar PV
2.2
Wind
CSP
2.2 3.7
1.1
2.5
0.01 0.1 0.05 1.1
0.5

2013 2014 2015 2016 2017 2018 2019 2020

7 Notes: Wind includes Eskoms Sere wind farm (100 MW)


Sources: Eskom; DoE IPP Office
2016: Wind, solar PV and CSP supplied 3% of the total RSA system load
Actuals captured in wholesale market for Jan-Dec 2016 (i.e. without self-consumption of embedded plants)

Annual
electricity
in TWh
3.7 (1.6%) 2.6 (1.1%) 0.5 (0.2%) 238.2
231.3

Residual Load Wind Solar PV CSP System Load


(domestic and
export load)
8 Notes: Wind includes Eskoms Sere wind farm (100 MW)
Sources: Eskom; DoE IPP Office
Actual tariffs: new wind/solar PV 40% cheaper than new coal in RSA
Results of Department of Energys RE IPP Procurement Programme (REIPPPP) and Coal IPP Proc. Programme

have made new solar PV & wind power 40%


Significant reductions in actual tariffs cheaper than new coal in South Africa today

Actual average tariffs Actual average tariffs


in R/kWh (Apr-2016-R) in R/kWh (Apr-2016-R)
5 -59%
-83%
4 3.65 -40%
Solar PV 1.03
3 Wind
2.18
0.62 0.62
2 1.51
1.19 1.17
0.87
1 0.62
0.87
0.69 0.62
0
Nov Mar Aug Aug Nov Solar PV IPP Wind IPP Baseload
2011 2012 2013 2014 2015 Coal IPP
Notes: Exchange rate of 14 USD/ZAR assumed Sources: http://www.energy.gov.za/files/renewable-energy-status-report/Market-Overview-and-Current-Levels-of-Renewable-Energy-
9 Deployment-NERSA.pdf; http://www.saippa.org.za/Portals/24/Documents/2016/Coal%20IPP%20factsheet.pdf; http://www.ee.co.za/wp-
content/uploads/2016/10/New_Power_Generators_RSA-CSIR-14Oct2016.pdf; StatsSA on CPI; CSIR analysis
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions

IRP Results and Least-cost Scenario

Summary

12
Integrated Resource Plan (IRP) aims for optimal electricity mix for RSA
In-principle process of IRP planning and implementation

Inputs Output
Planning / Demand forecast IRP Model Capacity exp. plan
Technology costs (PLEXOS) After policy
simulation assumptions adjustment: IRP
(techno-economical
world
Installed capacity

CO2 limits
Total installed
net capacity in GW

least-cost optimisation)
100

85.7
Solar PV
8.4
80
1.2 CSP
9.2
Wind


4.8
60 Hydro

Etc.
9.6
+1.8 Nuclear
42.2 7.3 Peaking
2.1 2.4
40 1.8 Gas
2.4

20 35.9
41.1 Coal

0
2010 2015 2020 2025 2030

Inputs Outcomes
Actuals / Procurement Preferred bidders
Ministerial
(competitive tender MW allocation
real world Determinations based
e.g. REIPPPP, coal IPPPP) Technology costs
on IRP capacities
actuals ( Tariffs)

13
Sources: CSIR analysis
IRP process as described in the Department of Energys Draft IRP 2016
document: least-cost Base Case is derived from technical planning facts

Scenario 1
Scenario 2

Constraint:
RE limits Case Cost
Constraint: 1) Public consultation
Forcing in Base Case Base on costed scenarios
Planning
Facts of nuclear, Scenario 1 Base + Rxx bn/yr
Least Cost
CSP, biogas, 2) Policy adjustment
Base Case Scenario 2 Base + Ryy bn/yr
hydro, others of Base Case
Scenario 3 Base + Rzz bn/yr
Constraint: 3) Final IRP
Advanced CO2
cap decline

Scenario 3

14
Sources: based on Department of Energys Draft IRP 2016, page 7; http://www.energy.gov.za/IRP/2016/Draft-IRP-2016-Assumptions-Base-Case-and-Observations-Revision1.pdf
The CSIR has embarked on power-system analyses to determine the
least-cost expansion path for the South African electricity system

The Integrated Resource Plan (IRP) is the expansion plan for the South African power system until 2050
Starting point of the IRP Base Case: pure techno-economic analysis to determine least-cost way to supply electricity
Later process steps: least-cost mix can be policy adjusted to cater for aspects not captured in techno-economic model

Draft IRP 2016 Base Case entails a limitation: Amount of wind and solar PV capacity that the model is allowed to build per
year is limited, which is neither technically nor economically justified/explained (no techno-economical reason provided)

The CSIR is therefore conducting a study to determine the Least Cost electricity mix in RSA until 2050
Majority of assumptions kept exactly as per the Draft IRP 2016 Base Case
First and most important deviation from IRP 2016: no new-build limits on renewables (wind/solar PV)
Second (smaller) deviation: costing for solar PV and wind until 2030 aligned with latest IPP tariff results
Scope of the CSIR study: purely techno-economical optimisation of the costs directly incurred in the power system

Two scenarios from the Draft IRP 2016 are compared with the Least Cost case
Draft IRP 2016 Base Case new coal, new nuclear
Draft IRP 2016 Carbon Budget significant new nuclear
Least Cost least-cost without constraints

An hourly capacity expansion and dispatch model (incl. unit commitment) using PLEXOS
15
is run for all scenarios to test for technical adequacy same software platform as by Eskom/DoE for the IRP
Sources: CSIR analysis
CSIR uses an industry standard software package for expansion
planning of the power system same package as used by DoE/Eskom

Commercial software used by DoE & CSIR covers all key cost drivers of a power system

Co-optimisation of long-term investment & operational Costs covered in the model include
decisions in hourly time resolution from today to 2050 All capacity-related costs of all power generators
What mix to build? CAPEX of new power plants (R/kW)
How to operate the mix once built? Fixed Operation and Maintenance (FOM)
Objective function: least cost, subject to an cost (R/kW/yr)
adequate (i.e. reliable) power system All energy-related costs of all power generators
Variable Operation and Maintenance (VOM)
cost (R/kWh)
Fuel cost (R/GJ)
Efficiency (heat rate) losses due to more flexible
operation
Reserves provision (included in capacity costs)
Key technical limitations of power generators covered
Maximum ramp rates (% of installed capacity/h) Costs not covered in the model currently used are
Minimum operating levels (% of installed capacity) Any grid-related costs (note: transmission-level
Minimum up & down times (h btw start/stop) grid costs typically ~10-15% of generation costs)
Start-up and shut-down profiles Costs related to add. system services (e.g. inertia
16 requirements, black-start and reactive power)
Sources: CSIR analysis
CSIR team has significant expertise from power system planning,
system operation and grid perspective

Dr Tobias Bischof-Niemz Joanne Calitz


Head of the CSIR Energy Centre Senior Engineer: Energy Planning
Member of the Ministerial Advisory Council (CSIR Energy Centre)
on Energy (MACE) Previously with Eskom Energy Planning
Member of IRP2010/2013 team at Eskom, Medium-Term Outlook and IRP for RSA
energy planning in Europe for large utilities

Robbie van Heerden Mamahloko Senatla


Senior Specialist: Energy Systems Researcher: Energy Planning
(CSIR Energy Centre) (CSIR Energy Centre)
Former General Manager and long-time Previously with the Energy Research
head of System Operations at Eskom Centre at University of Cape Town

Crescent Mushwana Jarrad Wright


Research Group Leader: Energy Systems Principal Engineer: Energy Planning
(CSIR Energy Centre) (CSIR Energy Centre)
Former Chief Engineer at Eskom strategic Commissioner: National Planning
transmission grid planning Commission (NPC)
17 Former Africa Manager of PLEXOS
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions

IRP Results and Least-cost Scenario

Summary

18
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions


Cost inputs for solar PV and wind
Limitations on build-out rates for solar PV and wind

IRP Results and Least-cost Scenario

Summary

19
IRP 2010 forecasted steep cost decline for solar PV from 2010 to 2030

Tariff in R/kWh
(Apr-2016-Rand)

4.0
Assumptions: IRP 2010 - high
3.5 Assumptions: IRP 2010 - low

3.0

2.5

2.0

1.5

1.0

0.5

0.0
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
22 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
Actual solar PV tariffs quickly moved below IRP 2010 cost assumptions

Tariff in R/kWh
(Apr-2016-Rand)

4.0
3.65 Assumptions: IRP 2010 - high
3.5 Assumptions: IRP 2010 - low
= 2.8 GW Actuals: REIPPPP (BW1-4Exp)
3.0

2.5 2.18
2.0

1.5 1.17
0.91
1.0

0.5
0.62
0.0
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
23 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
IRP 2016 increases cost assumptions for solar PV compared to IRP 2010

Tariff in R/kWh
(Apr-2016-Rand)

4.0
3.65 Assumptions: IRP 2010 - high
3.5 Assumptions: IRP 2010 - low
= 2.8 GW Assumptions: IRP 2016 - high
3.0
Assumptions: IRP 2016 - low
2.5 Actuals: REIPPPP (BW1-4Exp)
2.18
2.0

1.5 1.17
0.91
1.0

0.5
0.62
0.0
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
24 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
IRP 2010 forecasted small cost decline for wind from 2010 to 2030

Tariff in R/kWh
(Apr-2016-Rand)

1.75
Assumptions: IRP 2010
1.50

1.25

1.00

0.75

0.50

0.25

0.00
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
25 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
Actual wind tariffs quickly moved below IRP 2010 assumptions

Tariff in R/kWh
(Apr-2016-Rand)

1.75
Assumptions: IRP 2010
1.52
1.50 Actuals: REIPPPP (BW1-4Exp)
= 4.0 GW
1.25
1.19
1.00
0.87
0.75
0.69
0.50 0.62

0.25

0.00
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
26 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
IRP 2016 increases cost assumptions for wind compared to IRP 2010

Tariff in R/kWh
(Apr-2016-Rand)

1.75
Assumptions: IRP 2010
1.52
1.50 Assumptions: IRP 2016 - high
= 4.0 GW Assumptions: IRP 2016 - low
1.25 Actuals: REIPPPP (BW1-4Exp)
1.19
1.00
0.87
0.75
0.69
0.50 0.62

0.25

0.00
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
27 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions


Cost inputs for solar PV and wind
Limitations on build-out rates for solar PV and wind

IRP Results and Least-cost Scenario

Summary

29
Draft IRP 2016 limits the annual build-out rates for solar PV and wind

The imposed new-build limits for solar PV and wind mean that the IRP model is not allowed in any given
year to add more solar PV and wind capacity to the system than these limits

No such limits are applied for any other technology. No techno-economical reason/justification is provided
for these limits. No explanation given why the limits are constant until 2050 while the power system grows
Year System Peak New-build limit Relative new-build New-build limit Wind Relative new-build
Load in MW (as Solar PV in MW/yr limit Solar PV in MW/yr limit Wind
per Draft IRP) (as per Draft IRP) (derived from IRP) (as per Draft IRP) (derived from IRP)
2020 44 916 1 000 2.2% 1 600 3.6%

2025 51 015 1 000 2.0% 1 600 3.1%

2030 57 274 1 000 1.7% 1 600 2.8%

2035 64 169 1 000 1.6% 1 600 2.5%

2040 70 777 1 000 1.4% 1 600 2.3%

2045 78 263 1 000 1.3% 1 600 2.0%

2050 85 804 1 000 1.2% 1 600 1.9%


30 Note: Relative new-build limit = New-build limit / system peak load
Sources: IRP 2016 Draft; CSIR analysis
Today: Both leading and follower countries are installing more new
solar PV capacity per year than South Africas IRP limits for 2030/2050

Annual new solar PV Germany


Leader
capacity relative to Spain
17%
system peak load 10% Italy
9% 9%
UK
Australia Follower
Japan
RSA new-build limits 7%
7% in 2030 and 2050 China Follower
6% 2nd wave
6% India
South Africa
5%
4% 4%
4% 4% 4%

3% 3% 3% RSAs IRP relative


3% 3% 3% new-build limit
3% 3%
2% 2% 2% decreases
2% 2% 2%
2% 2% 2% over time
2%
1% 1% 1% 1% 2030 (1.7%)
1% 1% 1% 1% 1%
0%
1% 0% 0%
0% 1% 1%
0% 1% 0%
1% 1% 2050 (1.2%)
0% 0% 0% 1% 0%
0% 0% 0%
0% 0% 0% 0% 0% 0% 0% 0%

2007 2008 2009 2010 2011 2012 2013 2014 2015

31
Sources: SolarPowerEurope; CIGRE; websites of System Operators; IRP 2016 Draft; CSIR analysis
Today: Both leading and follower countries are installing more new
wind capacity per year than South Africas IRP limits for 2030/2050

Annual new wind


capacity relative to Germany
system peak load Spain Leader
8%
8% Ireland
RSA new-build limits 7% China
in 2030 and 2050 7% India Follower
6%
6% Brazil
6% 6%
5% South Africa
5% 5% 5%
5%
4% 4%
4% RSAs IRP relative
4% 4% new-build limit
4% 4%
3% 3% decreases
3% 3% 3% 3% 3%
3% 3% 3%
over time
2% 3%
2% 2% 2% 2% 2%
2030 (2.8%)
2% 2% 2% 2% 2%
2% 2% 2% 2% 2% 2% 2050 (1.9%)
1% 2% 1%
1% 1%
1% 1% 1%
1% 1% 1%
0% 1%
0% 0% 0%
0% 0% 0% 0%
0%

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

32
Sources: GWEC; CIGRE; websites of System Operators; IRP 2016 Draft; CSIR analysis
Solar PV penetration in leading countries today is 2.5 times that of
South Africas Draft IRP 2016 Base Case for the year 2050

70%
Germany
Spain Leader
60%
relative to system peak load

Italy
Total solar PV capacity

UK
50%
Australia Follower
Japan
40%
China Follower
India 2nd wave
30%
South Africa
South Africa IRP 2016 Base Case
20%

10%

0%
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
33 Year
Sources: SolarPowerEurope; CIGRE; websites of System Operators; IRP 2016 Draft; CSIR analysis
Wind penetration in leading countries today is 1.7-1.8 times that of
South Africas Draft IRP 2016 Base Case for the year 2050

70%
Germany
Spain Leader
60%
relative to system peak load

Ireland
China
50%
Total wind capacity

India Follower
Brazil
40%
South Africa
South Africa IRP 2016 Base Case
30%

20%

10%

0%
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
34 Year
Sources: GWEC; CIGRE; websites of System Operators; IRP 2016 Draft; CSIR analysis
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions

IRP Results and Least-cost Scenario

Summary

35
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions

IRP Results and Least-cost Scenario


Input Assumptions
Results

Summary

36
Input as per IRP 2016: Demand is forecasted to double by 2050
Forecasted demand for the South African electricity system from 2016 to 2050

Electricity
in TWh/yr
550 522
500
450 428 Note: by 2050 the electricity
Demand
demand per capita would still be
400 less than that of Australia today
344
350
300 280
244
250
200
150
100
50
0
2016 2020 2030 2040 2050

37
Sources: DoE (IRP 2016); Eskom MTSAO 2016-2021; StatsSA; World Bank; CSIR analysis
Input as per IRP 2016: Decommissioning schedule for existing plants
Decommissioning schedule for the South African electricity system from 2016 to 2050

Electricity
in TWh/yr
550
500
450 Demand All power plants considered for
existing fleet that are either:
400 Existing supply 1) Existing in 2016
350 Solar PV 2) Under construction
CSP 3) Procured (preferred bidder)
300
259 Wind
250
Other
200 174 Peaking
150 Gas (CCGT)
100 82 Hydro+PS
Decommissioning of Nuclear
50 Eskoms coal fleet Coal
0
2016 2020 2030 2040 2050

38
Sources: DoE (IRP 2016); Eskom MTSAO 2016-2021; StatsSA; World Bank; CSIR analysis
Demand grows, existing fleet phases out gap needs to be filled
Forecasted supply and demand balance for the South African electricity system from 2016 to 2050

Electricity
in TWh/yr
550 522
500
450 428 Demand The IRP model fills the supply
gap in the least-cost manner,
400 Supply gap subject to any constraints
344 Solar PV imposed on the model
350
CSP
300 84 254 439
Wind
250
Other
200 Peaking
150 Gas (CCGT)
100 Hydro+PS
Decommissioning of Nuclear
50 Eskoms coal fleet Coal
0
2016 2020 2030 2040 2050

39 Note: All power plants considered for existing fleet that are either Existing in 2016, Under construction, or Procured (preferred bidder)
Sources: DoE (IRP 2016); Eskom MTSAO 2016-2021; StatsSA; World Bank; CSIR analysis
Inputs as per IRP 2016:
Key resulting LCOE from cost assumptions for new supply technologies

Lifetime cost
3.69
per energy unit1
Same assumptions used Fixed
(LCOE) in R/kWh
(Apr-2016-R) as per IRP 2016 2.89 (Capital,
O&M)

1.41 1.41
1.09
Variable
1.00
(Fuel)
0.62 0.62

Solar PV Wind Baseload Nuclear Gas (CCGT) Mid-merit Coal Gas (OCGT) Diesel (OCGT)
Coal (PF)
CO2 in kg/MWh
0 0 1 000 0 400 1 000 600 600

Assumed capacity factor2 82% 90% 50% 50% 10% 10%


1 Lifetime cost per energy unit is only presented for brevity. The model inherently includes the specific cost structures of each technology i.e. capex, Fixed O&M, variable O&M, fuel costs etc.
2 Changing full-load hours for new-build options drastically changes the fixed cost components per kWh (lower full-load hours higher capital costs and fixed O&M costs per kWh);
41 Assumptions: Average efficiency for CCGT = 55%, OCGT = 35%; nuclear = 33%; IRP costs from Jan-2012 escalated to May-2016 with CPI; assumed EPC CAPEX inflated by 10% to convert EPC/LCOE
into tariff; Sources: IRP 2013 Update; Doe IPP Office; StatsSA for CPI; Eskom financial reports for coal/diesel fuel cost; EE Publishers for Medupi/Kusile; Rosatom for nuclear capex; CSIR analysis
IRP 2016 increases cost assumptions for solar PV compared to IRP 2010

Tariff in R/kWh
(Apr-2016-Rand)

4.0
3.65 Assumptions: IRP2010 - high
3.5 Assumptions: IRP2010 - low
= 2.8 GW Assumptions: IRP 2016 - high
3.0
Assumptions: IRP 2016 - low
2.5 Actuals: REIPPPP (BW1-4Exp)
2.18
2.0

1.5 1.17
0.91
1.0

0.5
0.62
0.0
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
43 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
CSIR study cost input assumptions for solar PV:
Future cost assumptions for solar PV aligned with IRP 2010

Tariff in R/kWh
(Apr-2016-Rand)

4.0
3.65 Assumptions: IRP2010 - high
3.5 Assumptions: IRP2010 - low
= 2.8 GW Assumptions: IRP 2016 - high
3.0
Assumptions: IRP 2016 - low
2.5 Assumptions for this study
2.18
Actuals: REIPPPP (BW1-4Exp)
2.0

1.5 1.17
0.91
1.0

0.5 0.62 0.56 0.52 0.49


0.0
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
44 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
IRP 2016 increases cost assumptions for wind compared to IRP 2010

Tariff in R/kWh
(Apr-2016-Rand)

1.75
Assumptions: IRP2010
1.52
1.50 Assumptions: IRP 2016 - high
= 4.0 GW Assumptions: IRP 2016 - low
1.25 Actuals: REIPPPP (BW1-4Exp)
1.19
1.00
0.87
0.75
0.69
0.50 0.62

0.25

0.00
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
45 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
CSIR study cost input assumptions for wind:
Future cost assumptions for wind aligned with results of Bid Window 4

Tariff in R/kWh
(Apr-2016-Rand)

1.75
Assumptions: IRP2010
1.52
1.50 Assumptions: IRP 2016 - high
= 4.0 GW Assumptions: IRP 2016 - low
1.25 Assumptions for this study
1.19
Actuals: REIPPPP (BW1-4Exp)
1.00
0.87
0.75
0.69
0.62
0.50 0.62 0.62 0.62

0.25

0.00
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
46 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
CSIR study cost input assumptions for CSP:
Todays latest tariff as starting point, same cost decline as per IRP 2010

Tariff in R/kWh
(Apr-2016-Rand) For bid window 3, 3.5 and 4 Exp,
weighted average tariff of base Assumptions: IRP2010 - high
4.0 and peak tariff calculated on the
Assumptions: IRP2010 - low
3.55 assumption of 64%/36%
3.32 base/peak tariff utilisation ratio Assumptions: IRP2016 - high
3.5
3.11 Assumptions: IRP2016 - low
2.90
3.0 Assumptions for this study
Actuals: REIPPPP (BW1-4Exp)
2.5
2.02
2.0

1.5
1.20 1.20 1.20
1.0

0.5

0.0
2010 2015 2020 2025 2030 2035 2040 2045 2050
Year
BW1 BW 4 (Expedited)
47 Notes: REIPPPP = Renewable Energy Independant Power Producer Programme; BW = Bid Window; bid submissions for the different BWs: BW1 = Nov 2011; BW2 = Mar 2012; BW 3 = Aug 2013;
BW 4 = Aug 2014; BW 4 (Expedited) = Nov 2015 Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis
CO2 emissions constrained by RSAs Peak-Plateau-Decline objective
PPD that constrains CO2 emission from electricity sector

CO2 Emissions Cap


(electricity sector)
[Mt/yr]

300 275 275


250
250
200
200

150

100

50

0
2010 2015 2020 2025 2030 2035 2040 2045 2050

48 PPD = Peak Plateau Decline


Sources: DoE (IRP 2010-2030 Update); StatsSA; CSIR analysis
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions

IRP Results and Least-cost Scenario


Input Assumptions
Results

Summary

49
Overview of scenarios available for comparison

Difference to
Scenario Source Draft IRP 2016 Base Case

Draft IRP 2016 Department of Energy N/A


Base Case Draft IRP 2016 as of November 2016

Draft IRP 2016 Department of Energy Tighter carbon reduction targets


Carbon Budget Draft IRP 2016 as of November 2016

Draft IRP 2016 Department of Energy No constraints on solar PV and wind


Unconstrained Scenario run by DoE/Eskom as per
Base Case request of the Ministerial Advisory
Council on Energy (MACE)

Least Cost CSIR No constraints on solar PV/wind


Draft Least Cost as of February 2017 Solar PV, wind and CSP costing
aligned with latest IPP results
50 Demand response by 2050 in
residential warm water provision
Common reporting layout applied to all scenarios by DoE and by CSIR

are analysed with respect to total installed


IRP scenarios as published by the DoE capacity (GW) and energy balance (TWh/yr)

Determine total operational capacity per year


Add existing fleet & its decommissioning schedule
Decommission new power plants at the end of
their economic life (wind = 20, solar PV = 25 years)

Determine energy balances with typical load factors for


different technologies and calibrate with IRP numbers

Total electricity
produced in TWh/yr

550 523
28 Solar PV
500 (5%)
93
CSP
450 431
22
(18%) Wind
400 39 Peaking
344 66 (7%)
350 13 33 Gas (CCGT)
33 (6%)
300 36 Hydro+PS
34
248 23 165 Nuclear
250 13 35
17 (32%)
15 Coal
200
Scenarios of the Draft IRP 2016 show the annual 150
235 229
new installed capacity per year per technology 100 207 159
(30%)
50
0
2016 2030 2040 2050

51
Sources: Draft IRP 2016, http://www.energy.gov.za/IRP/irp-presentaions/IRP-Update-Presentation-22-Nov-2016.pdf; CSIR analysis
Draft IRP 2016 Base Case is a mix of roughly 1/3 coal, nuclear, RE each

As per Draft IRP 2016

Draft IRP 2016 Base Case

Total electricity
produced in TWh/yr

550 523
28
500 (5%)
93
450 431
(18%)
22
400 39
344 66 (7%)
350 13 33
36 33 (6%)
300
34
248 23 165
250 13 35
17 (32%)
15
200
150
235 229
100 207 159
(30%)
50
0
More stringent
2016 2030 2040 2050 carbon limits

Solar PV Wind Gas (CCGT) Nuclear


52 CSP Peaking Hydro+PS Coal
Sources: DoE Draft IRP 2016; CSIR analysis
Draft IRP 2016 Carbon Budget case: 40% nuclear energy share by 2050

As per Draft IRP 2016

Draft IRP 2016 Base Case Draft IRP 2016 Carbon Budget

Total electricity Total electricity


produced in TWh/yr produced in TWh/yr

550 523 550 525


28 44
500 (5%) 500 (8%)
93 433
450 431 450 109
(18%)
22 39 (21%)
400 39 400
344 66 (7%) 345
350 33 350 63
13 23 103
33 (6%) (12%)
300 36 300 35
34 63
248 23 165 248 33 (7%)
250 13 35 250
17 (32%) 17 23 29
15 15
200 200 57 206
150 150 134 (39%)
235 229
100 207 159 100 207
161
(30%)
50 50 85 63
(12%)
0 0
More stringent
2016 2030 2040 2050 carbon limits 2016 2030 2040 2050
No RE limits, reduced wind/solar PV costing, warm water demand flexibility

Solar PV Wind Gas (CCGT) Nuclear


53 CSP Peaking Hydro+PS Coal
Sources: DoE Draft IRP 2016; CSIR analysis
Least Cost case is largely based on wind and solar PV

As per Draft IRP 2016

Draft IRP 2016 Base Case Draft IRP 2016 Carbon Budget Least Cost

Total electricity Total electricity Total electricity


produced in TWh/yr produced in TWh/yr produced in TWh/yr

550 523 550 525 550 527


28 44
500 (5%) 500 (8%) 500
130
93 433 433
450 431 450 109 450 (25%)
(18%)
22 39 (21%)
400 39 400 400 87
344 66 (7%) 345 346
350 33 350 63 350
13 23 103 35
33 (6%) (12%)
300 36 300 35 300
34 63 49
248 23 165 248 33 (7%) 248 282
250 13 35 250 250 22 189 (54%)
17 (32%) 17 23 29 17
15 15 15 15
200 200 57 200
206
(39%) 44
150 150 134 150 9
(8%)
100 207
235 229
100 207 100 207 212 22 16 20
159 15
161 (4%)
(30%)
50 50 85 63 50 83 36
(12%) (7%)
0 0 0
More stringent
2016 2030 2040 2050 carbon limits 2016 2030 2040 2050 2016 2030 2040 2050
No RE limits, reduced wind/solar PV costing, warm water demand flexibility

Solar PV Wind Gas (CCGT) Nuclear


54 CSP Peaking Hydro+PS Coal
Sources: DoE Draft IRP 2016; CSIR analysis
Least Cost means no new coal and no new nuclear until 2050,
instead 90 GW of wind and 70 GW of solar PV plus flexible capacities

As per Draft IRP 2016

Draft IRP 2016 Base Case Draft IRP 2016 Carbon Budget Least Cost

Total installed Total installed Total installed


net capacity in GW net capacity in GW net capacity in GW Technical
potential
250 250 250 237 in REDZ
only

200 200 200 73 1 782 GW


178

149
150 150 150 52
135
129 25
16
111 22 93 535 GW
12 30 98 36 100
100 85 100 100 60
13 34 19
7 21 13 10
11 12 20 16
51 22 51 8 51
8 17 10 33
50 8 50 19 50 22 31 37
5 3 2 6 5 3 8 5 3
5 8 20 7 8 8 2 5
37 39 37 34
17 26 37 34 5 7 2 18
33 25 5
19 10 19 10
0 0 0
More stringent
2016 2030 2040 2050 carbon limits 2016 2030 2040 2050 2016 2030 2040 2050
No RE limits, reduced wind/solar PV costing, warm water demand flexibility

Note: REDZ = Renewable Energy Development Zones Solar PV Wind Gas (CCGT) Nuclear Plus 25 GW demand
55 Current REDZ cover 7% of South Africas land mass CSP Peaking Hydro+PS Coal response from residential
Sources: DoE Draft IRP 2016; CSIR analysis warm water provision
On request by the Ministerial Advisory Council on Energy (MACE) the
DoE re-ran the IRP 2016 Base Case without constraining solar PV/wind

Source: MACEs presentation during the IRP public consultations on 7 December 2016 in Johannesburg

56
Sources: http://www.energy.gov.za/IRP/irp-presentaions/Comments-on-IRP-2016-Draft-KabiSolar.pdf
The DoEs Unconstrained Base Case is similar to the CSIRs Least Cost

Source: MACEs presentation during the IRP public consultations on 7 December 2016 in Johannesburg

57
Sources: http://www.energy.gov.za/IRP/irp-presentaions/Comments-on-IRP-2016-Draft-KabiSolar.pdf
Draft IRP 2016 Base Case:
Nuclear and coal dominate the supply mix in 2050

Demand and Exemplary Week under Draft IRP 2016 Base Case (2050)
Supply in GW
120
110
100
90
80
70
60
50
40
30
20
10
0
Monday Tuesday Wednesday Thursday Friday Saturday Sunday

Solar PV Peaking Gas (CCGT) Nuclear Demand


Wind Hydro Coal

58
Sources: CSIR analysis, based on DoEs Draft IRP 2016
Draft IRP 2016 Carbon Budget:
Nuclear dominates the supply mix in 2050, gas required for balancing

Demand and Exemplary Week under Draft IRP 2016 Carbon Budget (2050)
Supply in GW
120
110
100
90
80
70
60
50
40
30
20
10
0
Monday Tuesday Wednesday Thursday Friday Saturday Sunday

Solar PV Peaking Gas (CCGT) Nuclear Demand


Wind Hydro Coal

59
Sources: CSIR analysis, based on DoEs Draft IRP 2016
Least Cost:
Solar PV and wind dominate supply mix in 2050, with excess at times

Demand and Exemplary Week under Least Cost (2050)


Supply in GW
120
110
100
90
80
70
60
50
40
30
20
10
0
Monday Tuesday Wednesday Thursday Friday Saturday Sunday

Curtailed solar PV and wind Wind Hydro Coal Demand


Solar PV Peaking Gas (CCGT) Nuclear

60
Sources: CSIR analysis
Total cost of power generation: Draft IRP 2016 Base Case R86 bn/year
more expensive by 2050 than Least Cost (without cost of CO2)

Total cost of power


generation in bR/yr
(constant 2016 Rand)
522
550 518
Draft IRP 2016 Base Case
500 +86
Draft IRP 2016 Carbon Budget
413 (+20%)
450 Least Cost 398
400 436

350 286
300 359
300
250
262
200
150 132

100
Note: Medium-term from 2016 to 2030 not in the main focus of a long-term IRP study
50
and therefore only indicative. Will be investigated in more detail in a separate sub-study.
0
61 2010 2016 2020 2030 2040 2050
Sources: CSIR analysis
Average tariff (without cost of CO2):
Draft IRP Base Case tariff 17 cents/kWh higher than Least Cost by 2050

Average tariff in R/kWh


(constant 2016 Rand)
1.25 1.29
1.3 1.13
1.17 +0.17
1.2 1.22 1.13 (+15%)
1.1
1.13
1.0 1.06
0.9 0.83
0.8
0.7
0.6 Draft IRP 2016 Base Case
0.5 Draft IRP 2016 Carbon Budget
Least Cost
0.4
0.3
0.2
Note: Medium-term from 2016 to 2030 not in the main focus of a long-term IRP study
0.1 and therefore only indicative. Will be investigated in more detail in a separate sub-study.
0.0
62 2010 2016 2020 2030 2040 2050
Note: Average tariff projections include 0.30 R/kWh for transmission, distribution and customer service (todays average cost for these items) Sources: Eskom on Tx, Dx cost; CSIR analysis
Average tariff (with cost of CO2):
Draft IRP Base Case tariff 20 cents/kWh higher than Least Cost by 2050

Average tariff in R/kWh


(constant 2016 Rand)

1.4 1.34
1.3 +0.20
1.2 1.14 (+17%)
1.1
1.0 0.95
0.9
0.8
0.7
0.6 Draft IRP 2016 Base Case
0.5 Draft IRP 2016 Carbon Budget
0.4 Least Cost
0.3
0.2 Note: Medium-term from 2016 to 2030 not in the main focus of a long-term IRP study
0.1 and therefore only indicative. Will be investigated in more detail in a separate sub-study.
0.0
63 2010 2016 2020 2030 2040 2050
Note: Average tariff projections include 0.30 R/kWh for transmission, distribution and customer service (todays average cost for these items) Sources: Eskom on Tx, Dx cost; CSIR analysis
Least Cost without renewables limits is R82-86 billion/yr cheaper by
2050 than IRP 2016 Base Case and IRP 2016 Carbon Budget case

As per Draft IRP 2016

Draft IRP 2016 Base Case Draft IRP 2016 Carbon Budget Least Cost

~525 TWh/yr ~525 TWh/yr ~525 TWh/yr


5% 8% 7%
12% 4% 8%
18% 30% 25%
21% 1%
2%
1%
7% 1%
39%
6% 12%
32% 7% 54%

R522 billion/yr R518 billion/yr R436 billion/yr


tariff = 1.29 R/kWh tariff = 1.29 R/kWh tariff = 1.13 R/kWh

200 Mt/yr 100 Mt/yr 70 Mt/yr

38 bn l/yr 16 bn l/yr 9 bn l/yr

64 Coal Nuclear Hydro + Pumped Storage Gas (CCGT) Peaking Other Wind CSP Solar PV
Note: Average tariff projections include 0.30 R/kWh for transmission, distribution and customer service (todays average cost for these items) Sources: Eskom on Tx, Dx cost; CSIR analysis
Sensitivity on cost difference: Even if RE were 50% more expensive
than assumed, Least Cost is still cheaper than Draft IRP 2016 Base Case

1.1 Annual cost delta of -40


Draft IRP Least Cost
by 2050 in bR/yr -20
1.0
0
20
0.9
40

Relative 0.8 60
RE/nuclear cost
80
0.7 86
(Today:
RE = 0.62 R/kWh, 100
Nuclear = 1.09 R/kWh
0.57)
0.6 120
Today (2016)
140
0.5 Study assumptions (2020-2050) 160
180
0.4 Relative
RE/coal cost 200
0.3 (Today: 220
65 RE = 0.62 R/kWh,
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 Coal = 1.00 R/kWh
Sources: CSIR analysis
0.62)
Agenda

Background

CSIRs Approach and Project Team

Comments on IRP Assumptions

IRP Results and Least-cost Scenario

Summary

66
Summary:
A mix of solar PV, wind and flexible power generators is least cost

Solar PV, wind & flexible power generators (e.g. gas, CSP, hydro, biogas, demand res.) is the cheapest new-
build mix for the RSA power system. It is cost-optimal to aim for >70% renewable energy share by 2050

This Least Cost mix is > R80 billion per year cheaper by 2050 than the current Draft IRP 2016 Base Case

Additionally, Least Cost mix reduces CO2 emissions by 65% (-130 Mt/yr) over the Draft IRP 2016 Base Case

Therefore: Avoiding CO2 emissions and least-cost is not a trade-off anymore South Africa can de-
carbonise its electricity sector at negative carbon-avoidance cost

The IRP and this analysis factor in all first-order cost drivers within the boundaries of the electricity system,
but not external costs and benefits of certain electricity mixes that occur outside of the electricity system

Deviations from the Least Cost electricity mix can be quantified to inform policy adjustments
(e.g. forcing in of certain technologies not selected by the least-cost mix like
coal, nuclear, hydro, CSP, biogas, biomass, etc.)

67 Note: Wind and solar PV would have to be 50% more expensive than assumed before the IRP Base Case and the Least Cost case break even
Sources: CSIR analysis
Re a leboha
Ha Khensa

Enkosi
Siyathokoza

Thank you

Re a leboga

Ro livhuha
Dankie
Siyabonga

68
Note: Thank you in all official languages of the Republic of South Africa