Formal comments on Integrated Resource Plan (IRP

Submitted to DoE on 25 October 2018

Formal comments on Integrated Resource Plan (IRP) 2018 CSIR Energy Centre Pretoria. 25 October 2018

1

Jarrad Wright

Joanne Calitz

Ntombifuthi Ntuli

[email protected]

[email protected]

[email protected]

Mpeli Rampokanyo

Pam Kamera

Ruan Fourie

[email protected]

[email protected]

[email protected]


2

Word Cloud (DoE, IRP 2010, 2011)

Word Cloud (DoE, IRP 2010 Update, 2013)

(Design: JG. Wright, using word_cloud, https://github.com/amueller/word_cloud)


Word Cloud (DoE, Draft IRP 2016, 2016)

Word Cloud (DoE, Draft IRP 2018, 2018)




Word Cloud (DoE, Draft IRP 2018) 3



Formal comments on Draft IRP 2018

1 Executive Summary 2 Background 3 Draft IRP 2018 scenarios 4 Draft IRP 2018 employment impacts 5 Draft IRP 2018 energy planning risks 6 System services and technical considerations

4




5


Key Messages

• Key Messages

1

6

•

Draft IRP 2018 is a very different plan to the Draft IRP 2016 and establishes solid principles

• •

VRE (PV and wind) with flexibility1 confirmed as least-cost energy mix as existing coal fleet decommissions; This energy mix also exhibits the least CO2 emissions and least water usage

•

Demand growth impacts the timing of new-build capacity but energy mix remains largely unchanged

•

New-build coal only post-2030 if CO2 emissions are not too restrictive and new-build VRE is constrained

•

New-build nuclear only post-2030 if CO2 emission are restrictive and new-build VRE is constrained

Natural gas fired peaking and mid-merit capacity considered as a proxy for this. NG - Natural gas; PV – Solar Photovoltaics; VRE – Variable renewable energy; EAF – Energy Availability Factor; RoCoF – Rate of Change of Frequency


Key Messages

• Key Messages

1

7

• •

To 2030 - outcomes similar across most scenarios but notable risks; Eskom coal fleet EAF (and decommissioning schedule), completion of new-build coal, stationary storage and DSR

•

To 2030, in the Recommended Plan, expect net employment increase (as system grows) but net decrease in coal

•

Post 2030 - key drivers include VRE new build limits, decommissioning, demand growth, stationary storage

•

Natural gas risk relatively small and can be replaced by appropriate domestic flexibility sources or stationary storage

•

No system integration issues foreseen pre-2030 but an informed and co-ordinated work program is necessary to sufficiently prepare for post-2030 relatively high VRE penetration levels

Natural gas fired peaking and mid-merit capacity considered as a proxy for this. NG - Natural gas; PV – Solar Photovoltaics; VRE – Variable renewable energy; EAF – Energy Availability Factor; RoCoF – Rate of Change of Frequency


At a glance Previous CSIR contributions have notably impacted and are mentioned throughout the Draft IRP 2018 • Demand forecast: CSIR (Built Environment) provide this as an input to DoE • Principles raised by CSIR comments on Draft IRP 2016 have been explicitly considered • These were: New-build limits removed (IRP1), RE costs aligned with REIPPPP, least-cost Base Case established

CSIR have engaged with key stakeholders in the 60 day public consultation period • Bilateral engagements: DoE, SALGA, EIUG, Nersa, Eskom • Attendance at IRP workshops: EE Publishers, FFF, NIASA (requested feedback) • Tentative: Parliament Portfolio Committee (Energy)

8


At a glance Draft IRP 2018 scenario summaries – this is a very different plan to the Draft IRP 2016 and establishes solid principles • Least-cost confirmed as combination of PV, wind and flexible capacity1 as coal decommissions – also exhibits lowest CO2 emissions and water usage by 2050 • Technology new-build limits (PV, wind) mean post-2030 deployment is constrained with new-build coal and gas replacing it (assuming less restrictive CO2 constraints)

• Higher NG price means less NG usage (notable capacity for system adequacy) and increased new-build coal • More strict CO2 emissions (Carbon Budget) with RE new-build limits means less new-build coal and deployment of nuclear instead • Higher NG price and more strict CO2 emissions (Carbon Budget) with RE new-build limits means less NG and coal, increased nuclear instead • Higher/Lower demand forecast means same mix of new-build of PV, wind and flexibility1 just earlier/later

9

1

Natural gas fired peaking and mid-merit capacity considered as a proxy for this; NG - Natural gas; PV – Solar Photovoltaics


Energy mix by 2030 similar across scenarios as coal dominates, IRP1 is ≈R10bn/yr cheaper than next best IRP3, IRP7 lowest CO2 emissions 2030 Energy Mix

Cost

in 2030

in 2030

202 (64%)

15 14 42 2

37

IRP3

197 (63%)

199 (63%)

158 13 44 2

15

14

199 (63%)

15

13

44 2

192 (61%)

15 4 26

44 2

206 (64%)

44 2

217

199

4

361

215

199

4

367

213

198

4

362

215

199

373

207

191

375

215

199

4 23

15 8 28 6 (2%) 3

10

360

22

3

Rec.

4

22

4 11

IRP7

Water usage (billion-litres/yr)

22

4 11

IRP6

CO2 emissions (Mt/yr)

25

9

IRP5

in 2030

Total system cost1 (R-billion/yr)

Demand: 313 TWh

IRP1

Environment

35 2

2

14

Coal

Nuclear

Gas - CCGT

Peaking

Wind

Solar PV

Other Storage

Coal (New)

Nuclear (new)

Gas - CCGE

Hydro+PS

CSP

Biomass/-gas

DR

Sources: DoE Draft IRP 2018; Eskom on Tx, Dx costs; CSIR analysis; flaticon.com


Least-cost mix confirmed as new-build solar PV, wind and flexible capacity (NG) - ≈R15-55 bn/yr cheaper than alternative scenarios 2040 Energy Mix

Cost

in 2040

in 2040

107 (30%)

in 2040


Demand: 353 TWh

IRP1

Environment

15 23 15

129

2

CO2 emissions (Mt/yr) 441

54

4

90

2 47

4

466

89

2 47

4

473

90

2 47

4

477

90

2 47

4

123

Water usage (billion-litres/yr) 59

5

IRP3

106 (30%)

151414 29

153

66

33 (9%)

IRP5

104 (29%)

15 11 29

168

70

51 (14%) 4

IRP6

97 (27%)

IRP7

99 (28%)

15 33

15 9 (2%)

25 12 30

43

12 29

353

494

116

56

119

58

5

Rec.

11

Coal

Nuclear

Gas - CCGT

Peaking

Wind

Solar PV

Other Storage

Coal (New)

Nuclear (new)

Gas - CCGE

Hydro+PS

CSP

Biomass/-gas

DR



By 2050 - Least-cost mix is 70% solar PV and wind, ≈R30-60 bn/yr cheaper than alternatives, least CO2 emissions and least water usage 2050 Energy Mix

Cost

in 2050

in 2050

68 (17%)

IRP3

66 (17%)

in 2050


Demand: 392 TWh

IRP1

Environment

29 17 31

164

78


3

82

31 8 31

110

63

3

560

13 32

107

64

3

564

106

67

3

106

64

3


160

54

80 (20%)

IRP5

64 (16%) 102 (26%)

IRP6 57 (15%)

55

47

178

59

5

12 31

580

92

36

90

38

13 (3%)

IRP7 58 (15%)

76

22 (6%)

17

31

591

13

Rec.

12

Coal

Nuclear

Gas - CCGT

Peaking

Wind

Solar PV

Other Storage

Coal (New)

Nuclear (new)

Gas - CCGE

Hydro+PS

CSP

Biomass/-gas

DR


Submitted to DoE on 25 October 2018 CO2 emissions trajectories for PPD Moderate never binding (only CB) while water use declines as expected as coal fleet decommissions

Scenarios from Draft IRP 2018

CO2 emissions

Water usage

Electricity sector CO2 emissions [Mt/yr]

Electricity sector Water usage [bl/yr]

300

275

300

275

250

250 210

200 150

178 IRP1

160

IRP3

200 150

IRP5

100

IRP3 92 82

IRP6 IRP7

50

IRP2 IRP4 PPD (Moderate)

13

PPD equiv.

100

2750 Mt

1800 Mt

920 Mt

> 2750 Mt

2720 Mt

2325 Mt

IRP5 IRP6

50

0 2015 2020 2025 2030 2035 2040 2045 2050 Carbon Budget

IRP1

IRP7 IRP2 IRP4

0 2015 2020 2025 2030 2035 2040 2045 2050

59

36


At a glance Jobs impact of Recommended Plan – reduced role of coal whilst growth in other sectors is expected • If the Recommended Plan is implemented, there is an employment reduction expected in coal pre-2030 • Overall jobs grow as the power system grows – in solar PV, wind and gas sectors • This transition needs sufficient preparation – our comments attempt to assist to quantify effects

14

EAF – Energy Availability Factor Sources: Eskom; CER


Net reduction of jobs in coal of ≈100k but net gain overall as gas grows to ≈55k jobs towards 2030, RE contributes up to ≈110k by 2030 Jobs (net) (construction + operations) [’000] 500

400 350 300

437

436

450 10 (3%) 365 3 (1%) 14 (4%) 44 (12%) 7 (2%)

398

394 14 3 25 7

359

3 4 3

7 44

44

342

7 4

7 44

362

39

22 10 4 12

20

44

33 44

250

41

42 4

4

60 42 44

434

64

403

43 5

5

44 5

62 61

49 44

62 47

44 44

200 150

287 (79%)

300

297

280

270

258

245

100

395 45 (11%) 5 (1%) 62 (16%) 56 (14%) 44 (11%)

233

217

202

183 (46%)

2027

2028

2029

2030

50 0

2020 Solar PV

15

2021 EG (PV)

2022 Wind

2023 Gas

2024

2025

Nuclear

2026 Coal

Sources: Draft IRP 2018; CSIR Energy Centre analysis Note: Job potential includes direct, indirect and induced jobs; Nuclear is estimated based on existing experience at Koeberg (KPMG, 2017)


At a glance Key energy planning risks – Existing fleet low EAF, stationary storage, DSR, further RE learning • • • • •

16

Existing fleet Low EAF: Earlier new-build (2023), outcomes return to IRP1 if low demand and low coal fleet performance Storage: Decreasing stationary storage costs (batteries) results in deployment pre-2030, less NG usage and increased solar PV DSR: A more responsive demand-side test via electric water heating and EVs delays some capacity investment and deploys slightly more solar PV Further RE learning: Increased solar PV and wind post-2030 with timing pre-2030 unchanged, no import hydro A risk adjusted scenario: Combining storage, DSR and further RE learning results in increased new-build wind, PV, storage and further lower NG use •

EAF – Energy Availability Factor; DSR – Demand Side Response; Evs – Electric Vehicles Sources: Eskom; Tesla; BNEF; CSIR

Submitted to DoE on 25 October 2018 Draft IRP 2018 (IRP1) - Least-cost deploys considerable wind, solar PV and NG capacity to 2030 and beyond as the coal fleet decommissions

Installed capacity and electricity supplied from 2016 to 2050 as planned in the Draft IRP 2018

Installed capacity

Energy mix

Total installed capacity (net) [GW]

Electricity production [TWh/yr]

160

148 1

140 119

120

3

350

31

309

1

1

245

50

38

10

60

58

52 2 1 3

5 2

40 20

2

3 3

5

0

38

42

200

1 5

5 8

10

23 3

32

17

0

2016

2020

2030

2040

25

10

Coal (new)

Other Storage

Wind

Gas - CCGT

Coal

Biomass/-gas

Hydro+PS

Nuclear (new)

Sources: Draft IRP 2018. CSIR Energy Centre analysis

100

78

53 2

39 13 15

6 3

127

164

203

214

202

14 23 5 15

31 1 17 29

107 68

0

2050

Gas - CCGE

Nuclear

15

2

11

CSP

Peaking

8

2 2 4 3 14

13

DR

Solar PV

262

25

4

8

12 4

2 14 3 3 15 4

78

80

4

300

21

100

17

392

400

RE CO2 free (RE + nuclear)

2016

2020

2030

2040

2050

8%

25%

70%

(20 TWh/yr)

(79 TWh/yr)

(274 TWh/yr)

14%

30%

70%

(35 TWh/yr)

(94 TWh/yr)

(274 TWh/yr)

to DoE on 25 October 2018 Draft IRP 2018 IRP1 with Submitted storage, DSR and lower RE costs results in increased new-build wind, solar PV, storage and less NG

Installed capacity and electricity supplied from 2016 to 2050 for IRP1 with storage, DSR and higher RE cost reductions

Installed capacity

Energy mix



196.8

200

7.2

177.3

180 160

20.7 0.6

400

4.4 11.8

63.0

0.6

87.3

80 51.6

40 20

60.5

5.1 13.1 6.3

107 (27%)

77

271

2 44 17 4 14

160 (40%)

122

18 4 18 5.1 12.8 1.9

13.6 9.9

16.9

Solar PV

Peaking

Coal (New)

2050

Nuclear

14

2045

Hydro+PS

2040

Biomass/-gas

0

2035

Nuclear (new)

17 (4%) 7 (2%) 38 (9%) 66 (17%)

103

2030

Wind

202

2025

Other Storage

Coal

209

2020

Gas

100

2016

CSP

2050

2045

2040

2035

2030

DR

Notes: NG = Natural gas Sources: Draft IRP 2018. CSIR Energy Centre analysis

(1%)

4

63.8

45.6

31.7

2025

2020

18

2016

0

1.5 57.8 1.5 5.1 3.4 1.9 37.4

399 4

31

200

4.8 0.6 17.1 0.5 15.3 5.1 8.7 1.2 1.9

341

320

249

41.7

100

298

300

112.8

120

60

361

147.2

140

383

Demand: Median First new-builds: PV (2027) Wind (2027) OCGT (2024) Storage (2027)

6.5 GW 2.1 GW 1.9 GW 1.1 GW

Submitted to DoE on 25 October 2018 Risk-adjusted scenario with Low EAF requires earlier new-build around 2023 too and increased absolute levels of new-build by 2030

Installed capacity and electricity supplied from 2016 to 2050 for Risk-adjusted scenario with low coal fleet EAF

Installed capacity

Energy mix

Total installed capacity (net) [GW] 198.3

200

160

153.6

140

11.6

400

61.0

0.6

298

300

271

80

69.4 57.8 51.6 1.5 1.5 5.1 3.4 1.9 37.4

40 20

5.1 9.4 1.2 1.9

5.1 13.7 6.3

31.7

1.9

12.1 9.9

16.9

Solar PV

Peaking

Coal (New)


2050

Nuclear

2045

Hydro+PS

14

2040

Biomass/-gas

2035

Nuclear (new)

0

2030

Wind

17 (4%) 7 (2%) 33 (8%) 66 (17%)

90

2025

Other Storage

Coal

180

2020

Gas

169 (42%)

135

17 4 18

208

2016

CSP

2050

2045

2040

2035

2030

2025

2020

2016 DR

100

77

17 4 15

67.9

5.1 13.2

104 (26%)

2

200 51.8

(1%)

4

61

97.7 0.6 20.1 0.5 21.4

340

399 4

36

249

41.4

382

361 320

119.3

100

19

21.1 0.7

4.4

120

0

7.2

180.8

180

60


Demand: Median First new-builds: PV (2023) Wind (2023) OCGT (2023)

0.4 GW 0.2 GW 1.9 GW


At a glance Descriptive inputs/comments – new-build limits, NG import risk and embedded generation • Technology new-build limits: Still unjustified, constant as power system grows, misaligned with international experience today • Coal capacity: Investigations reveal that older existing coal capacity could be decommissioned earlier, parts of under construction capacity could be replaced by alternatives and it is not economically optimal to build new coal • Natural gas import risk: Small role in energy mix (up to 5% pre-2030, 15% by 2050) - can be mitigated by range of domestic options (as well as stationary storage if costs decline) • Embedded generation: Planning for this is not yet explicit and will need to change (implicit as negative demand) • Demand profile will change further as different sectors grow, use energy differently and deploy EG for self-use

20


At a glance System services and technical considerations • System adequacy consistent across all scenarios i.e. reserve requirements included • System non-synchronous penetration: No barriers pre-2030. Only above 25% from 2028 and 37% by 2030 for 10% of the time but at above 80% by 2050 i.e. system integration focus becomes important post-2030 • A critical indicator (inertia): No barriers pre-2030. Post-2030 worst case mitigating solution cost is ≈1% of total system cost • Evolution of other system services need to be investigated for post-2030 transition expected (reactive power and voltage control, system strength) • Variable resource forecasting will become more important - SO should be equipped will relevant tools and skills

21


Going forward – Recommendations

22

•

Improvements for future IRPs

•

Need to have transparency in input assumptions, model and outcomes comprehensively and consistently published

• •

Investigate and establish the need for annual technology new-build limits; Remove annual new-build limits until further investigations establish the need for them

•

Optimise the existing coal fleet while remaining cognisant of opportunity cost of capital expenditure on older assets (retrofitting for improved reliability, efficiency and flexibility)

•

Inclusion of economic impacts of scenarios. At the very least – employment impacts

•

Improved approaches to better understanding demand (sector shifts, load profile shape, price elasticity of demand)

•

Better understanding the cost trajectory of all technologies for domestic application on a periodic basis

•

Develop and implement an integrated program of work on long-term system integration topics i.e. stability, system strength, reactive power/voltage control (CSIR already initiated CIGRE WG including Eskom and international SOs)

•

Focussed consideration and investigation into domestic flexibility options


Going forward – Recommendations

23

•

Long-term – structural and strategic

• •

Formally establishing a set of links/triggers between IRP and MTSAO processes (or equivalent) Periodic updating of the IRP should be prioritised to address dynamic planning environment

•

Further understanding just transition to address labour and socio-economic impacts in the energy sector

•

Integrating national and local level energy planning for improved co-ordination and leveraging of opportunities

•

Sector-coupling opportunities across the full energy sector (not just electricity)

•

Investigate approaches to include geospatial component of IRP – supply/network/demand (co-optimisation)

•

Further investigations into impacts/opportunities of new/emerging technologies e.g. stationary storage, EVs, DSR




24


Integrated Resource Plan (IRP): Process for power generation capacity expansion in South Africa Inputs

Planning / simulation world

1) Demand Forecast 2) Existing Supply Forecast: • Plants under construction • Preferred bidders • Decommissioning • Plant performance 3) New Supply Options: • Technology costs assumptions • Technology technical characteristics 4) Constraints: • CO2 limits • Security/adequacy of supply level

Inputs

Actuals / real world

1

25

2

LT = Long-term MT/ST = Medium-term/Short-term

• Ministerial Determinations for new technology specific generation capacity

Output IRP modelling framework (PLEXOS®) LT1 techno-economic least-cost optimisation MT/ST2 production cost testing system adequacy (security of supply)

Procurement (competitive tender e.g. REIPPPP, coal IPPPP)

Per scenario: • Total system costs • Capacity expansion (GW) • Energy share (TWh) • CO2 emissions • Water usage • Jobs in the electricity sector After policy adjustment: • Final “IRP” for promulgation • Key questions answered: o What to build (MW)? o When to build it (timing)?

Outcomes • Preferred bidders • MW allocation • Technology costs actuals (Ø IPP tariffs)


The IRP currently optimises for the dominant generation costs, system reserves (adequacy) and shallow grid cost components of total system cost

Optimised in PLEXOS model

Not optimised in PLEXOS modelling framework (Assumption consistent for all scenarios = 0.20 R/kWh)

[bR/yr]

• CAPEX (plant level) • FOM1 • VOM2 • Fuel • Shallow grid connection costs

Not considered

Not considered

Partially considered previously (quantified for system inertia)

Costs included in optimisation model:

Not considered

Costs excluded in optimisation model:

Generation (Gx)

Tx network (shallow)

System services (reserves)

Tx network (deep)

System services (other)

Other Distribution (metering, network billing, customer (Dx) services,

Total system cost

• Gx: Externalities e.g. CO2 emissions costs • Gx: Decommissioning costs • Gx: Waste management and/or rehabilitation • Gx: Major mid-life overhaul • Tx: Deep costs • Other system services • Dx: Distribution network costs • Other costs

overheads) 1

26

FOM = Fixed Operations and Maintenence costs; 2 VOM = Variable Operations and Maintenence costs; 3 Typically referred to as Ancillary Services includes services to ensure frequency stability, transient stability, provide reactive power/voltage control, ensure black start capability and system operator costs.


IRP process as described in the Department of Energy’s Draft IRP 2016 document: least-cost Base Case is derived from technical planning facts

Scenario 1 Scenario 2 Constraint: RE limits Planning Facts Least Cost Base Case

Constraint: e.g. forcing in of nuclear, CSP, biogas, hydro, others

Constraint: Advanced CO2 cap decline

Scenario 3 27

Sources: Based on Draft IRP 2018

Case

Cost

Base Case

Base

Scenario 1

Base + Rxx bn/yr

Scenario 2

Base + Ryy bn/yr

Scenario 3

Base + Rzz bn/yr

…

…

1. Public consultation on costed scenarios 2. Policy adjustment of Base Case 3. Final IRP for approval and gazetting


Reminder: IRP 2010 planned the electricity mix until 2030 Installed capacity and electricity supplied from 2010 to 2030 as planned in the IRP 2010

Installed capacity

Energy mix

Total installed net capacity [GW]

Electricity supplied [TWh]

150

600

125

500 441 392

400

100 86.6 8.4

73.6

75

9.2

62.7

50

25

0

28

43.0 0.0 2.4 2.1 0.0 1.8 0.0 35.9

2010

5.8

0.7 2.8

4.8

9.6

6.3

12 4

75

1.2

300

259

4.8 7.3

0.5 3.4 0.0 1.5

343

4 6

24 22

13

2.4

200

4 13 0 0

10

14 21

14

12 45

1.8 284 230

100

261

237

34.8

2020

2025

0

2030

Solar PV

Other

Gas

Coal (new)

CSP

Hydro

Nuclear (new)

Coal

Wind

Peaking

Nuclear

RE CO2 free

2010

2020

2025

2030

5%

14%

(12 TWh/yr)

(62 TWh/yr)

10%

34%

(25 TWh/yr)

(149 TWh/yr)

Note: Installed capacity and electricity supplied excludes pumped storage; Renewables include solar PV, CSP, wind, biomass, biogas, landfill and hydro (includes imports). Sources: DoE IRP 2010-2030; CSIR Energy Centre analysis


Draft IRP 2016 Base Case planned until 2050 Installed capacity and electricity supplied from 2016 to 2050 as planned in the Draft IRP 2016 Base Case

Installed capacity

Energy mix Electricity supplied [TWh]

Total installed net capacity [GW] 150

600

135.7

528

15.8

125

28

500

108.7 12.4 1.1

100 82.5 7.5

75

11.1

50

49.8

0.2 1.5 1.3 3.4 4.1 0.3 1.9 0.4

1.1 0.7 0.0

8.9 7.3 6.1 1.9 0.0 5.3

25 36.8

21.3 0.7 12.8 16.8 7.6 1.9 2.7 14.3

32.6 17.2

0

2016 Solar PV

29

2030

2040

Other storage

Gas

434 30.4

5

400

350

0.5 13.8

300 246

22.0

2 15 0

2

200

7.6

22

0

13 36 16 0

5 5 1 24 15

34

93 3

1 49

66 5 45

1

33

20

148

33 15

105 20.4

100

0.0 15.0

200

101

203 117

10.2

0

2050 Nuclear

CSP

Other

Hydro+PS

Coal (new)

Wind

Peaking

Nuclear (new)

Coal

RE CO2 free

2016

2030

2040

72

2050

10%

29%

(25 TWh/yr)

(152 TWh/yr)

10%

57%

(40 TWh/yr)

(300 TWh/yr)

Note: Installed capacity and electricity supplied includes pumped storage; Renewables include solar PV, CSP, wind, biomass, biogas, landfill and hydro (includes imports). Sources: DoE Draft IRP 2016; CSIR Energy Centre analysis




30


A range of scenarios have been assessed as part of the Draft IRP 2018 with key parameter changes

31

Submitted to DoE on 25 October 2018 Draft IRP 2018 (IRP1) - Least-cost deploys considerable wind, solar PV and NG capacity to 2030 and beyond as the coal fleet decommissions


Installed capacity

Energy mix



160

148 1

140 119

120

353

31

313

1

1

245

50

38

10

60

58

52 2 1 3

5 2

40 20

2

3 3

5

0

38

42

200

1 5

5 8

10

23 3

32

17

0

2016

2020

2030

2040

25

10

Coal (new)

Other Storage

Wind

Gas - CCGT

Coal

Biomass/-gas

Hydro+PS

Nuclear (new)


100

78

54 2

42 14 15

7 3

129

164

203

215

202

15 23 5 15

31 17 1 29

107 68

0

2050

Gas - CCGE

Nuclear

15

2

11

CSP

Peaking

8

2 2 4 3 15

13

DR

Solar PV

264

25

4

3

8

12 4

2 14 3 3 15 4

78

80

4

300

21

100

32

391

400


2016

2020

2030

2040

2050

8%

26%

70%

(20 TWh/yr)

(83 TWh/yr)

(274 TWh/yr)

14%

31%

70%

(35 TWh/yr)

(98 TWh/yr)

(274 TWh/yr)


Draft IRP 2018 (IRP3) – RE new-build limits mean post-2030 deployment of solar PV and wind is constrained with new-build coal and gas replacing it Installed capacity and electricity supplied from 2016 to 2050 as planned in the Draft IRP 2018

Installed capacity

Energy mix



140 125 1

120 103

353

25

313

1

100

300

18 1

77

80 52 2 1 3

5 2

40 20

38

2 5

3 3

9

1

58

13 5

4

8 7

1

14 6 5

32

17

0

2016

2020

4

2030

2040

12

10

100

44

13 15

203

214

Other Storage

Wind

Gas - CCGT

Coal

Biomass/-gas

Hydro+PS

Nuclear (new)

4 47 2 90

9 8 29 14 14 15 33

197

3 63

110

31 8 31 80

106 67

0

2050 Coal (new)


15

22

12

Gas - CCGE

Nuclear

8

2 2 4 3 15

17

CSP

Peaking

200

15

263

4 2

11

DR

Solar PV

245

26 8

9 2 42

30

1

1

60

33

393

400


2016

2020

2030

2040

2050

8%

26%

52%

(20 TWh/yr)

(80 TWh/yr)

(203 TWh/yr)

14%

30%

52%

(35 TWh/yr)

(95 TWh/yr)

(203 TWh/yr)

Submitted to DoE on 25 October 2018 Draft IRP 2018 (IRP5) – Market linked NG price means in less NG usage, notable capacity for system adequacy and increased new-build coal


Installed capacity

Energy mix



140 125 1

120 104

313 300

18 1

77

80

1

1

60

58

52 2 1 3

5 2

40

2 5

3 3

30

4 8

13 6

4 9 2

8 6

13

17

20 3

20

38

42

8

32

17

0

2016

2020

2030

2040

Coal (new)

Other Storage

Wind

Gas - CCGT

Coal

Biomass/-gas

Hydro+PS

Nuclear (new)


15

22 44

14

15

107

89

11 4 29 15

4 11

51

100

203

215

199

5

32 13

102

104

0

2050

Gas - CCGE

Nuclear

8

2 2 4 3 15

4 2

64

4 47 2

64 10

CSP

Peaking

200

15

264

16

DR

Solar PV

245

26

9

1

3

353

25

1

100

34

391

400


2016

2020

2030

2040

2050

8%

26%

51%

(20 TWh/yr)

(81 TWh/yr)

(202 TWh/yr)

14%

30%

51%

(35 TWh/yr)

(95 TWh/yr)

(202 TWh/yr)

Submitted to DoE on 25 October 2018 Draft IRP 2018 (IRP6) – Carbon Budget limits new-build coal capacity and deploys new-build nuclear capacity instead


Installed capacity

Energy mix



140 125 120

353

25

103

313

1

100

300

18

77

80 52 2 1 3

5 2

40 20

1

58

38

2 5

3 3

30

4

26

13 5

3 2

8 5

12

13

17

9

14

4

2

32

17

0

2016

2020

245

8

9 2 42

1 1 9

1

60

2030

2040

200

100

10

Gas - CCGE

Coal (new)

Other Storage

Wind

Gas - CCGT

Coal

Biomass/-gas

Hydro+PS

Nuclear (new) Nuclear


203

8 15

2 2 4 3 15

215

4 2

22 44

13

15

4 47 2 90

11 4

199

30 12 25 33 15 97

3 67

106

31 12 47 55

13 57

0

2050

CSP

Peaking

15

264

7

2

DR

Solar PV

35

391

400

1


2016

2020

2030

2040

2050

8%

26%

52%

(20 TWh/yr)

(81 TWh/yr)

(203 TWh/yr)

14%

30%

66%

(35 TWh/yr)

(96 TWh/yr)

(258 TWh/yr)


Draft IRP 2018 (IRP7) – Market linked NG price & Carbon Budget combines IRP 5&6 meaning less NG and coal capacity, increased nuclear new-build Installed capacity and electricity supplied from 2016 to 2050 as planned in the Draft IRP 2018

Installed capacity

Energy mix



140 125 1

120

353

25

104

300

18 1

77

80

1

60

52 2 1 3

5 2

40

3

38

8

8 5

13

8

15

18

3

42

4

2016

2020

2030

6 2

17

0

2040

10 4 10

CSP

Gas - CCGE

Coal (new)

Other Storage

Wind

Gas - CCGT

Coal

Biomass/-gas

Hydro+PS

Nuclear (new)

Peaking

Nuclear


100

15

8 15

2 2 4 3 15

4 2

23 44 26

15

90 3 4

29 5 12

64

106

31 13 17

203

215

192

9

14

76 22

99 58

0

2050

DR

Solar PV

200

264

3

43

6

2

32

245 4

8

2

20

30

26

13

2 5

3

9

1

58

1

4 47 2

313

1

100

36

391

400


2016

2020

2030

2040

2050

8%

30%

51%

(20 TWh/yr)

(95 TWh/yr)

(201 TWh/yr)

14%

35%

70%

(35 TWh/yr)

(110 TWh/yr)

(277 TWh/yr)

Submitted to DoE on 25 October 2018 Draft IRP 2018 (IRP2) – Upper Demand forecast with similar outcomes to IRP3 just with earlier first new-build and more new-build overall


Installed capacity

Energy mix



140

1

120

108

100

60

58

52 2 1 3

5 2

40 20

1

38

2 5

3 3

321

18 30

9

2016

2020

200

15

8

15

6

28 14

3 8

2030

217

31 9 39

103

198

2040

108 68

10

0

2050 Coal (new)

Other Storage

Wind

Gas - CCGT

Coal

Biomass/-gas

Hydro+PS

Nuclear (new)


100

203

15

Gas - CCGE

Nuclear

32 11 19 17

2

8

32

111 91

43

48 15

CSP

Peaking

4

2 4 2 3 15

64

4 46 2

21

18

DR

Solar PV

266

16

17

0

300

8

8 6

7 10 2

42

378

4 2

26

14 3

400

25

245

1 9

1

3

1

1

79

80

37

428

130


2016

2020

2030

2040

2050

8%

29%

48%

(20 TWh/yr)

(94 TWh/yr)

(207 TWh/yr)

14%

33%

48%

(35 TWh/yr)

(108 TWh/yr)

(207 TWh/yr)

Submitted to DoE on 25 October 2018 Draft IRP 2018 (IRP4) – Lower Demand forecast with similar outcomes to IRP3 just with later first new-build and less new-build overall


Installed capacity

Energy mix



140 121 1

120

333

1

60

58

52 2 1 3

5 2

40

7 2 5

3 3

10 3

38

42

23

5

8 7

6

2

20

1 1 1

17 2

2016

2020

2030

200

100

15

14

33 15 14

203

208

Coal (new)

Other Storage

Wind

Gas - CCGT

Coal

Biomass/-gas

Hydro+PS

Nuclear (new) Nuclear

4 41 2

63

81

112

29 15 8 15 33

30 11 33

2 8 5

193

54

105 65

0

2050

Gas - CCGE


38

2040

10

CSP

Peaking

8

3

11

DR

Solar PV

4

4

8 17

0

16

2 4 2 3 15

12

12 4 5

32

8

256

245

30

1

69

290

300

16

80

372

25

96

100

400


2016

2020

2030

2040

2050

8%

23%

55%

(20 TWh/yr)

(66 TWh/yr)

(204 TWh/yr)

14%

28%

55%

(35 TWh/yr)

(80 TWh/yr)

(204 TWh/yr)


Draft IRP 2018 (Recommended Plan) includes RE new-build limits and policy adjustment for new-build coal and imported hydro Installed capacity and electricity supplied from 2016 to 2030 as planned in the Draft IRP 2018

Installed capacity

Energy mix



100 400 80

60

74.1 61.8

57.8

7.6 3.1 8.1 1.9 1.0

1.6 5.1 3.4 1.9

20

292

300

264

245

200

100

37.6

231

217

203

31.7

1.0 GW 0.7 GW 0.2 GW 2.3 GW

2030

2029

Nuclear

First new-builds: Coal (2023) PV (2025) Wind (2025) CC-GE (2026)

2028

Peaking

2027

Solar PV

2026

Nuclear (new)

2025

Hydro+PS

2024

Biomass/-gas

2023

Coal

2022

Gas - CCGT

2021

Wind

2020

Other Storage

2019

Coal (New)

2018

Gas - CCGE

2017

CSP

2016

2030

2029

2028

2027

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

0

DR


39

0.3 8.0 0.6 11.5

51.91.5

40

0

14 (7%) 35 313(17%) 28 (14%) 8 (4%) 3 (1%) 15 (7%) 6 (3%) 199 (97%)

Submitted to DoE on 25 October 2018 CO2 emissions trajectories for PPD Moderate never binding (only CB) while water use declines as expected as coal fleet decommissions

Scenarios from Draft IRP 2018

CO2 emissions

Water usage



300

275

300

275

250

250 210

200 150

178 IRP1

160

IRP3

200 150

IRP5

100

IRP3 92 82

IRP6 IRP7

50

IRP2 IRP4 PPD (Moderate)

40

PPD equiv.

100

2750 Mt

1800 Mt

920 Mt

> 2750 Mt

2720 Mt

2325 Mt

IRP5 IRP6

50

0 2015 2020 2025 2030 2035 2040 2045 2050 Carbon Budget

IRP1

IRP7 IRP2 IRP4

0 2015 2020 2025 2030 2035 2040 2045 2050

59

36


Average tariff (without CO2 costs) across scenarios revealing how IRP1 (Least-cost) is 12% cheaper than the most expensive scenario (IRP7) Average tariff in R/kWh (Jan-2017 Rand) 1.60

1.51

1.40

1.35

+12%

1.20 1.00 0.80

0.84

IRP1 IRP3 IRP5 IRP6 IRP7 Recommended Plan

0.60 0.40 0.20 0.00 2015

2020

2025

Note: Shift from 2017 to 2018 based on immediate move to cost reflectivity Sources: Draft IRP 2018. CSIR Energy Centre analysis

41

2030

2035

2040

2045

2050


Total system cost increase as the power system grows (as expected) IRP1 is least-cost and ≈R60-bn/yr cheaper than IRP7 by 2050 Total system cost (Jan-2017 Rand) [bR/yr] 591

600

529

+63 (+12%)

500 400 300 200

IRP1 IRP3 IRP5 IRP6 IRP7 Recommended Plan

209

100 0 2015

2020

2025

Note: Shift from 2017 to 2018 based on immediate move to cost reflectivity Sources: Draft IRP 2018. CSIR Energy Centre analysis

42

2030

2035

2040

2045

2050


Energy mix by 2030 similar across scenarios as coal still dominates while IRP1 is ≈R10bn/yr cheaper than IRP7, IRP7 lowest CO2 emissions 2030 Energy Mix

Cost

in 2030

in 2030

202 (64%)

15 14 42 2

37

IRP3

197 (63%)

199 (63%)

158 13 44 2

15

14

199 (63%)

15

13

44 2

192 (61%)

15 4 26

44 2

206 (64%)

44 2

217

199

4

361

215

199

4

367

213

198

4

362

215

199

373

207

191

375

215

199

4 23

15 8 28 6 (2%) 3

43

360

22

3

Rec.

4

22

4 11

IRP7

Water usage (billion-litres/yr)

22

4 11

IRP6

CO2 emissions (Mt/yr)

25

9

IRP5

in 2030


Demand: 313 TWh

IRP1

Environment

35 2

2

14

Coal

Nuclear

Gas - CCGT

Peaking

Wind

Solar PV

Other Storage

Coal (New)

Nuclear (new)

Gas - CCGE

Hydro+PS

CSP

Biomass/-gas

DR



Least-cost mix confirmed as new-build solar PV, wind and flexible capacity (NG) - ≈R15-55 bn/yr cheaper than alternative scenarios 2040 Energy Mix

Cost

in 2040

in 2040

107 (30%)

in 2040


Demand: 353 TWh

IRP1

Environment

15 23 15

129

2


54

4

90

2 47

4

466

89

2 47

4

473

90

2 47

4

477

90

2 47

4

123


5

IRP3

106 (30%)

151414 29

153

66

33 (9%)

IRP5

104 (29%)

15 11 29

168

70

51 (14%) 4

IRP6

97 (27%)

IRP7

99 (28%)

15 33

15 9 (2%)

25 12 30

43

12 29

353

494

116

56

119

58

5

Rec.

44

Coal

Nuclear

Gas - CCGT

Peaking

Wind

Solar PV

Other Storage

Coal (New)

Nuclear (new)

Gas - CCGE

Hydro+PS

CSP

Biomass/-gas

DR



By 2050 - Least-cost mix is 70% solar PV and wind, ≈R30-60 bn/yr cheaper than alternatives, least CO2 emissions and least water usage 2050 Energy Mix

Cost

in 2050

in 2050

68 (17%)

IRP3

66 (17%)

in 2050


Demand: 392 TWh

IRP1

Environment

29 17 31

164

78


3

82

31 8 31

110

63

3

560

13 32

107

64

3

564

106

67

3

106

64

3


160

54

80 (20%)

IRP5

64 (16%) 102 (26%)

IRP6 57 (15%)

55

47

178

59

5

12 31

580

92

36

90

38

13 (3%)

IRP7 58 (15%)

76

22 (6%)

17

31

591

13

Rec.

45

Coal

Nuclear

Gas - CCGT

Peaking

Wind

Solar PV

Other Storage

Coal (New)

Nuclear (new)

Gas - CCGE

Hydro+PS

CSP

Biomass/-gas

DR





46


CSIR have analysed selected scenarios from Draft IRP 2018 using a transparent, well established and understood tool - iJEDI The International Jobs and Economic Development Impacts (I-JEDI) model is a freely available economic tool to understand economic changes (jobs focus at this stage) for energy technology choices I-JEDI has been customised for application to the South African environment by the CSIR team CSIR utilised the developed I-JEDI tool for South Africa to assess the Recommended Plan in the Draft IRP 2018 for a range of technologies including wind, solar, coal and natural gas (for now)… more in future High-level approach • I-JEDI estimates economic impacts by characterising construction and operation of energy projects in terms of expenditures and portion of these made within the country (localised) • These are then used in a country-specific input-output (I-O) model to estimate employment (amongst a range of other metrics) An indicative analysis of jobs in nuclear is also provided (but not based on I-JEDI for now) Analysis of other Draft IRP 2018 scenarios as well as those presented by CSIR to come1 1

Time available was not not sufficient to do this in the 60 day public consultation period. Sources: https://www.nrel.gov/analysis/jedi/about.html

47


Coal dominant in jobs (as expected) but declines to 2030 in Recommended Plan as gas grows, notable gap for wind and PV

Construction jobs [‘000]

Operations jobs (net) [‘000]

500

500

450

450

400

400

350

350

300

300

250

250

200

200

150

258

217

202

183

2030

233

2029

245

50


2027

Coal

2026

Nuclear

2025

Gas

265

2028

0

264

2024

Wind

24

265

2023

2028

EG (PV)

24

271

2022

2027

2025

2024

48

2023

2022

2021

Solar PV

100

278

2021

35

45

38 4 52

1 1 323 1 3221 2 7 3 7 3053 4 294 4 7 1 7 1 7 1 7 1 5 6 278 1 44 1 4 4 4 44 7 4 7 44 44 1 44 14 44 2 5 7 18 8 2 44 23 9 44 32 11 44 44 44

2020

57

38 4 52

2030

31 93 21 3 36 1 22 3 39 36 37 3 54 16 21 35 32 3 16 3 5 5 26 11 9 29

54

12

150

2029

4

82

2026

38 3

66

2020

0

131 134 141 117 117 38 38 4

100 50

3341


Net job decrease in coal of ≈100k but net gain overall as gas grows to ≈55k jobs towards 2030, RE contributes up to ≈110k by 2030 Jobs (net) (construction + operations) [’000] 500

400 350 300

437

436

450 10 (3%) 365 3 (1%) 14 (4%) 44 (12%) 7 (2%)

398

394 14 3 25 7

359

3 4 3

7 44

44

342

7 4

7 44

362

39

22 10 4 12

20

44

33 44

250

41

42 4

4

60 42 44

434

64

403

43 5

5

44 5

62 61

49 44

62 47

44 44

200 150

287 (79%)

300

297

280

270

258

245

100

395 45 (11%) 5 (1%) 62 (16%) 56 (14%) 44 (11%)

233

217

202

183 (46%)

2027

2028

2029

2030

50 0

2020 Solar PV

49

2021 EG (PV)

2022 Wind

2023 Gas

2024

2025

Nuclear

2026 Coal



Focus on coal: Emphasising impact of construction jobs via new-build (excl. Medupi/Kusile) and net decline in operations jobs to 2030

Construction jobs [‘000]

Operations jobs (net) [‘000]

500

500

450

450

400

400

350

350

300

300

250

250

200

200

150

87

100

107

107

107

245 104

99

217 94

183 88

83

80

76

73

74

68

63

57

79

77

76

75

76

74

70

67

62

58

52

2024

2025

2026

2027

2028

2029

2030

2030

Sources: Draft IRP 2018; CSIR Energy Centre analysis Note: Job potential includes direct, indirect and induced jobs

2029

Direct

2028

2027

2026

2025

2024

2023

2022

2021

Indirect

0

82

2023

5

83

2022

50

84

81

2021

29 32 16

Induced

50

109

265

2020

9 2020

0

112

265

150

100 50

278


Net job losses in coal overall of ≈100k, direct jobs in coal shifting from ≈80k in 2016 to ≈50k by 2030 Jobs (net) (construction + operations) [’000] 500 450 400 -104 (-36%)

350 300 250

287

300 120

115

297 118

280 112

200 150

94

90

93

88

270 108

84

100 50 0

82

86

2020

2021

Induced

51

85

2022 Indirect

Sources: Draft IRP 2018; CSIR Energy Centre analysis Note: Job potential includes direct, indirect and induced jobs

258 104

80

245 99

76

233 94

73

217 88

202 81

68

63

183 74

57

80

77

74

70

67

62

58

52

2023

2024

2025

2026

2027

2028

2029

2030

Direct



1 Executive Summary 2 Background 3 Draft IRP 2018 scenarios 4 Draft IRP 2018 employment impacts 5 Draft IRP 2018 energy planning risks 5.1 Scenarios 5.2 Descriptive comments 6 System services and technical considerations

52



1 Executive Summary 2 Background 3 Draft IRP 2018 scenarios 4 Draft IRP 2018 employment impacts 5 Draft IRP 2018 energy planning risks 5.1 Scenarios a

Impact of stationary storage

b

Impact of Demand Side Response (DSR)

c

Technology learning

d

Risk-adjusted scenario

e

Existing coal fleet performance

5.2 Descriptive comments 6 System services and technical considerations

53





b


c

Technology learning

d


e



54


Impact of stationary storage (scenario) In the Draft IRP 2018 – no cost reductions are considered for stationary storage What if stationary storage costs start to decline?

55


Stationary storage (excl. pumped storage): 200 $/kWh (2030), 150 $/kWh (2040), 100 $/kWh (2050) Overnight capital cost [R/kWh] (Jan-2017-Rand) 14 000

IRP2018 (Li-Ion 1hr)

This scenario (Li-Ion 3hr)

CSIRO - Low (2015)

IDC - 2016

This scenario (Li-Ion 1hr)

BNEF (2017)

Nykvuist - Low (2015)

IRENA - High (2017)

IRP2018 (Li-Ion 3hr)

CSIRO - High (2015)

Nykvuist - High (2015)

[USD/kWh] 1 000 900

-81%

12 000

-75%

800

10 000

700 600

8 000

500 6 000

400

300

4 000 2 800

200

2 100

2 000 0 2010 56

1 400

2015

2020

2025

2030

2035

USD:ZAR = 14.00; NOTE: Battery packs are assumed to make up 60% of total utility-scale stationary storage costs (from BNEF). Sources: StatsSA for CPI; Draft IRP 2018; CSIR; BNEF; CSIRO (2015); Nykvist (2015); IDC; IRENA (2017)

2040

2045

100

0 2050

to DoE on 25 October 2018 Draft IRP 2018 IRP1 with Submitted storage cost declines means notably less NG, with storage deployed from 2027, increased solar PV and wind


Installed capacity

Energy mix



100

400

90

350

81.7

80

3.2 0.6

70

14.3

60 50

60.2

57.8

0.5

1.5 5.1 3.4 1.9

5.1 8.6

37.4

31.7

10

150 209 (64%)


6.2 GW 1.2 GW 1.9 GW 1.3 GW

2030

2029

2028

2027

2026

Coal (New)

2025

Peaking

2024

Solar PV

2023

Nuclear

2022

Hydro+PS

2021

Biomass/-gas

2020

Nuclear (new)

2019

Wind

2018

Other Storage

Coal

2017

Gas

2016

CSP

249

227

209

50 0

2030

2029

2028

2027

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

DR


57

200

100

20

271 249

1.9

30

0

250

(8%) 38 (12%) 18 (5%) 7 (2%) 14 (4%)

298

300

13.3

51.61.5

40

32626

to DoE on 25 October 2018 Draft IRP 2018 IRP1 with Submitted storage cost declines means less NG, increased solar PV and wind with considerable deployment post-2030

Installed capacity and electricity supplied from 2016 to 2050 for IRP1 with Demand Side Response

Installed capacity

Energy mix



200

429

180

400

169.1

160

151.2

140

129.1

120

11.0

103.8

57.8 51.6 1.5

60.2

1.5 5.1 3.4 1.9 37.4

40 20

298

48.9

271 249

200

3.2 0.6 14.3 0.5 13.3 5.1 8.6 2.6 1.9

7.6 11.9 7.7

7.6 10.4 1.9

15.8 9.9

16.9

60 2

38 18 7 14

110

Solar PV

Peaking

Coal (New)

Notes: Assumed capacity factors for wind and solar PV differ slightly to IRP1 Sources: Draft IRP 2018. CSIR Energy Centre analysis

0

2050

Nuclear

111

2045

Hydro+PS

209

2040

Biomass/-gas

23 14

2035

Nuclear (new)

209

2030

Wind

100

2025

Other Storage

Coal

33 (8%) 5 (1%) 48 (12%) 71 (18%)

3

2020

Gas

153 (39%)

33

2016

CSP

2050

2045

2040

2035

2030

2025

DR

4

26

56.2

39.4

31.7

2020

2016

58

326 300

81.7

80

0

352

19.7 0.6

0.6

4 (1%) 83 (21%)

375

32.2

100

60

404


6.2 GW 1.2 GW 1.9 GW 1.3 GW

Submitted to DoE on 25 October 2018 Difference in installed capacity and energy mix with storage cost declines relative to IRP1, less NG with increased solar PV

Installed capacity and electricity supplied from 2016 to 2050 for IRP 1 with storage cost declines

Installed capacity

Energy mix

Difference - Total installed capacity (net) [GW]

Difference - Electricity production [TWh/yr]

50

50

40

40

30

30 20

20 10 3 2 -2

-10

-3

-20

-12

-10

-8

-20

Solar PV

Peaking

Coal (New)

-6 -1

2050

Nuclear

-6

2045

Hydro+PS

2

2040

Biomass/-gas

3

-2

2035

Nuclear (new)

2030

Wind

2025

Other Storage

Coal

2020

Gas

-4 -1

2016

CSP

2050

2045

2040

2035

-50

2030

-50

2025

-40

2020

-40

DR

6 -2

-17

-30

2016

59

-2

0

-30


4 1

12

3 1 -9

-2 -1

20

10

11

0

20

to DoE on 25 October 2018 CO2 emissions trajectoriesSubmitted for PPD Moderate never binding while water use declines as expected as coal fleet decommissions

IRP 1 with storage cost declines

CO2 emissions

Water usage



300

275

300

275

250

250 210

200

200

150

150

100 50

82

IRP1

78

IRP1 with storage cost declines PPD (Moderate)

0 2015 2020 2025 2030 2035 2040 2045 2050 Carbon Budget 60

PPD equiv.

2750 Mt

1800 Mt

920 Mt

> 2750 Mt

2720 Mt

2325 Mt

100 50

IRP1 IRP1 with storage cost declines

0 2015 2020 2025 2030 2035 2040 2045 2050

36 35


Total system cost: IRP1 with storage cost declines ≈R8 bn/year less expensive by 2050 than IRP1, marginal difference before 2030 Net Present Value System Cost

Total System Cost

(2016 - 2050)

Total system cost in bR/yr (Jan-2017 Rand)

bR

600 529 500

441

400

358

437

4 018

4 005

IRP1

Storage cost declines

299

300 200

521

-12 (-0.3%)

199

100 0 2015

61

2020

2025

2030

2035

2040

IRP1

IRP1 with storage cost declines

2045

2050

Note: Average tariff projections (and resulting total system cost) consider an offset representative of Tx/Dx/Other costs to align with starting point of 0.84 ZAR/kWh (0.20 ZAR/kWh). From 2017 to 2018, immediate cost reflectivity is considered too (as in Draft IRP 2018) i.e. 0.21 ZAR/kWh offset. Sources: Draft IRP 2018. CSIR Energy Centre analysis.





b


c

Technology learning

d


e



62


Impact of demand response - EVs, warm water heating (scenario) What if the demand side became more flexible and responsive i.e. Demand Side Response (DSR) Similar to previous comments in Draft IRP 2016 with some updated analysis on DSR options Warm-water heating (geysers) Electric vehicles

63


Majority of EV owners charge vehicles overnight in off-peak – this will have a relatively small impact on demand profile (beyond 2030) Impact of typical EV charging on RSA demand profile

Typical EV charging profiles 0.12 Hourly electricity demand (GW)

60

Charging Fraction (%)

0.10 0.08 0.06 0.04 0.02 0.00

50 40 30 20 10 0

0

2

4

6

8 10 12 14 16 18 20 22 24

+2.2 (+5%)

0

2

4

6

8 10 12 14 16 18 20 22 24

Hour

Hour EPRI (2007)

Base Case: No EVs

California Energy Commission and NREL (2018)

3 million electric vehicles (9-mln vehicle parc)

U.S Department of Energy (2014)

64

NOTE: EVs assumed only light passenger vehicles for now. Sources: Calitz (2018); Bedir, N. et. Al. (2018); U.S. Department of Energy (2014); EPRI (2007)


Electric vehicle usage for demand side flexibility i.e. Vehicle-to-Grid (V2G)

Inclusion of a demand side flexibility resource in the form of mobile storage (electric motor vehicles) demonstrates impact on the power system as adoption increases Considered with similar functionality as that of Electric Water Heating (EWH) demand shaping - a resource with intra-day controllability (can be dispatched as needed on any given day) based on power system needs i.e. vehicle-to-grid (V2G) Key input parameters to estimate potential demand shaping via electric motor vehicles: • • • • • • • •

65

Current population Expected population growth to 2050 Current number of motor vehicles Expected motor vehicles per capita Adoption rate of electric vehicles to 2050 Electric vehicle fleet capacity (MW) Electric vehicle energy requirement (GWh/d) Proportion of electric vehicle fleet connected simultaneously


Electric vehicle demand shaping can provide ~96 GW/4.2 GW (demand increase/decrease) with ~40 GWh/d of dispatchable energy by 2050

Property

Unit

2016-2019

2020

2031

2040

2050

Population

[mln]

0-0

58.0

61.7

64.9

68.2

Number of motor vehicles

[mln]

7 - 8.2

8.5

12.3

16.2

20.5

[%]

0-0

1.5

8.1

28.5

48.9

[mln]

0-0

0.1

1.0

4.6

10.0

EVs adoption Number of EVs

66

EVs energy requirement

[TWh/a]

-

0.5

3.7

17.1

37.0

EVs energy requirement EVs (demand increase) EVs (demand decrease)

[GWh/d] [MW] [MW]

-

1.3 -

10.1 4 600 400

46.8 44 300 2 000

101.4 95 800 4 200

Sources: CSIR estimates; StatsSA; eNaTis


Demand shaping as a demand side resource - domestic electric heaters (EWHs)

Many opportunities for demand shaping in a number of end-use sectors (domestic, commercial, industrial) In the scenarios assessed by CSIR - the intention of including one particular demand shaping opportunity (domestic electric water heating) is to demonstrate the significant impact this can have on the power system. Modelled as a resource with intra-day controllability (can be dispatched as needed on any given day) based on power system needs Key input parameters to estimate potential demand shaping via EWH: • • • • • • • •

67

South African population (to 2050) Number of households (current) Number of persons per household (future) EWHs (current) EWHs per household (future) Adoption rate of demand shaping via EWHs (future) Calibration for power (MW) and energy (TWh) used for electric water heating (existing) Movement to EWH technologies i.e. heat pumps vs electric geysers (future)


Demand shaping can provide ~24 GW/3 GW (demand increase/decrease) with ~70 GWh/d of dispatchable energy by 2050

Property

Unit

2016-2019

2020

2030

2040

2050

Population

[mln]

55.7 - 57.5

58.0

61.7

64.9

68.2

Number of HHs

[mln]

16.9 - 18.1

18.5

22.4

26.0

27.3

[ppl/HH]

3.29 - 3.17

3.13

2.75

2.50

2.50

HHs with EWH

[%]

28 - 33

34

50

75

100

HHs with EWH

[mln]

4.7 - 5.9

6.3

11.2

19.5

27.3

Residents per HH

Demand shaping adoption

68

[%]

-

2

25

100

100

Demand shaping

[TWh/a]

-

0.4

5.4

28.3

26.4

Demand shaping

[GWh/d]

-

1.1

14.9

77.4

72.3

Demand shaping (demand increase)

[MW]

-

371

4 991

25 970

24 265

Demand shaping (demand decrease)

[MW]

-

46

620

3 226

3 015

Sources: CSIR estimates; StatsSA; AMPS survey; Stastista; Eskom; Draft IRP 2016

to DoE on 25 October 2018 Draft IRP 2018 IRP1 with Submitted DSR impact marginal, shift in timing of import hydro, wind & PV (2030-2040)


Installed capacity

Energy mix



200 180

167.0

160

149.1

140

4.4 0.6

81.3

80 57.8 51.6 1.5

40 20

13.4 10.7

271

200

0.5 5.1

7.6

4.1 1.9

20.2

10.7

2

39 18 10 14

104

209

9.9

209

Solar PV

Peaking

Coal (New)

71 (18%)

2050

Nuclear

34 (9%) 3 (1%) 65 (16%)

2045

Hydro+PS

2040

Biomass/-gas

2035

Nuclear (new)

2030

Wind

2025

Other Storage

Coal

0

2020

Gas

158 (40%)

114

2016

CSP

2050

2045

2040

2035

2030

2025

DR

(1%) 64 (16%)

35 3 30 14

23.8

1.9

16.9


100

398 4

54

7.6 22.2

379

4

23

37.3

31.7

2020

69

2016

0

60.5

1.5 5.1 3.4 1.9 37.4

12.8

298

300 249

58.2 0.6

339

320

29.3

94.3

100

360

7.2 0.7 37.4

128.7

120

60

400


5.2 GW 1.1 GW 1.9 GW

Submitted to DoE on 25 October 2018 Difference in installed capacity and energy mix with DSR relative to IRP1 marginal, shift in timing of import hydro, wind & PV (2030-2040)


Installed capacity Difference - Total installed capacity (net) [GW]


50

50

40

40

30

30

20

20

10

10

0

-1 1

1 -1

7

4 -1

Nuclear

Solar PV

Peaking

Coal (New)

1 -1

2050

Hydro+PS

1

2045

Biomass/-gas

2040

Nuclear (new)

2035

Wind

2030

Other Storage

Coal

2025

Gas

2020

CSP

-3 1

-4

2016

-50

2050

-50

2045

-40

2040

-40

2035

-30

2030

-30

2025

-20

2020

-20

2016

-10

DR

1 2 2

0 -10


70

Energy mix


IRP 1 with Demand Side Response

CO2 emissions

Water usage

Electricity sector CO2 emissions [Mt/yr] 300

Electricity sector Water usage [bl/yr] 275

300

275

250

250 210

200

200

150

150

100 50

82 82

IRP1

50

IRP1 with DSR PPD (Moderate)

0 2015 2020 2025 2030 2035 2040 2045 2050 Carbon Budget 71

PPD equiv.

100

2750 Mt

1800 Mt

920 Mt

> 2750 Mt

2720 Mt

2325 Mt

IRP1 IRP1 with DSR

0 2015 2020 2025 2030 2035 2040 2045 2050

36 36


Total system cost: IRP1 with DSR marginal difference in total system cost relative to IRP1 Net Present Value System Cost

Total System Cost

(2016 - 2050)


bR

600

529

500

441 360

400

4 018

4 015

IRP1

DSR

-2 (0.1%)

299

300 200

440

199

100 0 2015

2020

2025 IRP1

72

2030

2035

2040

2045

2050

IRP1 with DSR

Note: Average tariff projections (and resulting total system cost) consider an offset representative of Tx/Dx/Other costs to align with starting point of 0.84 ZAR/kWh (0.20 ZAR/kWh). From 2017 to 2018, immediate cost reflectivity is considered too (as in Draft IRP 2018) i.e. 0.21 ZAR/kWh offset. Sources: Draft IRP 2018. CSIR Energy Centre analysis. Eskom on Tx, Dx costs


Generally – additional EV demand met by least-cost mix of wind, solar PV and gas whilst V2G shifts this mix more towards solar PV and less gas Impact on least-cost capacity mix

EVs typical configuration (G2V i.e. additional demand)

EVs as a demand shaping resource (V2G i.e. implicit in demand)

• For 1-mln EVs, increase of ≈3 TWh/yr

• Increase in proportion of new solar PV vs. wind

• For 1-mln EVs, increase ≈1 GW peak demand

• Less gas capacity i.e. lower gas energy share

• New build capacity to meet charging demand is

• Relative reduction in total system cost

mostly wind, solar PV and gas • Most charging demand met by wind and gas

2030 G2V - Grid to vehicle 73

V2G - Vehicle to grid





b


c

Technology learning

d


e



74


Solar PV learning assumptions in Draft IRP 2018 Actual solar PV tariffs and forecasted tariff trajectory

Equivalent tariff, Jan 2017, [ZAR/kWh] 4.00

3.85 Actuals: REIPPPP (BW1-4Exp) Assumption for further RE cost reductions Assumptions: IRP 2016 (Upper) Assumptions: IRP 2016 (Lower)

3.00

Assumptions: IRP 2018 (Upper) Assumptions: IRP 2018 (Lower)

2.29

2.00

-66% -40% 1.29

1.00

1.23 0.96

1.06

0.89

0.72

0.65

0.00 2010

2015

0.79

0.39

2020

2025

2030

0.54

0.22

2035

2040

0.22

2045

2050

BW1  BW 4 (Expedited) 75

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 . Learning rate - Bloomberg New Energy Outlook 2017


Wind cost learning assumptions in Draft IRP 2018 Actual wind tariffs and forecasted tariff trajectory


Actuals: REIPPPP (BW1-4Exp) Assumption for further RE reductions

1.60

Assumptions: IRP 2016 (Upper)

1.60

Assumptions: IRP 2016 (Lower) Assumptions: IRP 2018 (Upper)

1.20

1.04

0.92

0.80

Assumptions: IRP 2018 (Lower)

1.16

1.25

0.87

0.81 -28%

0.73

0.60

0.47

0.65

0.35

0.40

0.00 2010

0.78

-47%

2015

2020

2025

2030

2035

2040

0.35

2045

2050

BW1  BW 4 (Expedited) 76

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 . Learning rate - Bloomberg New Energy Outlook 2017


Input assumptions for CSP from Draft IRP 2018 and further cost declines Today’s latest tariff as starting point, same cost decline as per IRP 2010


3.00

3.74 3.50 3.28 3.07

3.30

2.89

Assumptions: IRP2010 - high

Assumptions for further RE reductions

Assumptions: IRP2010 - low

Actuals: REIPPPP (BW1-4Exp)





2.65

2.65 2.49

-36% 2.12

2.00

1.75

1.75

1.32

1.36

1.36 1.14

1.00

0.00 2010

2015

BW1  BW 4 (Expedited)

77

2020

2025

2030

2035

2040

2045

2050

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; For CSP bid window 3, 3.5 and 4 Exp, weighted average tariff of base and peak tariff calculated on the assumption of 64%/36% base/peak tariff utilisation ratio; Sources: StatsSA for CPI; IRP 2010; South African Department of Energy (DoE); DoE IPP Office; CSIR analysis

to DoE on 25 October 2018 Draft IRP 2018 IRP1 with Submitted further RE cost reductions, increased solar PV and wind from 2030 onwards, timing unchanged, no import hydro

Installed capacity and electricity supplied from 2016 to 2050 for IRP1 with higher RE cost reductions

Installed capacity

Energy mix



200

184.2

180

400

164.8

160

361

49.2

84.0

80 51.6

40 20

60.2

27.6

100

209

14

204 103

Solar PV

Peaking

Coal (New)

2050

Nuclear

2045

Hydro+PS

2040

Biomass/-gas

2035

Nuclear (new)

2030

Wind

2025

Other Storage

Coal

2020

Gas

0

2016

CSP

2050

2045

2040

2035

2030

2025

2020

DR

17 (4%) 5 (1%) 51 (13%) 66 (17%)

18 5 22

20.2

31.7

172 (43%)

128

7.6

15.8 5.1 12.0 2.6


78

200

0.6 13.9

(1%) 84 (21%)

4

2

46 18 6 14

69.1

9.9

2016

0

1.5 57.8 1.5 5.1 3.4 1.9 37.4

271

399 4

66 25

249

107.2

100

60

298

300

120

342

321

142.8 140

381


5.9 GW 2.8 GW 1.9 GW

Submitted to DoE on 25 October 2018 Difference in installed capacity and energy mix with higher RE cost reductions relative to IRP1, higher RE & peaking gas, less mid-merit gas

Installed capacity and electricity supplied from 2016 to 2050 for IRP1 with higher RE cost reductions

Installed capacity Difference - Total installed capacity (net) [GW]


50

50

40

40

30

30

20

12 21

12

20

11

10

6

10

10 1 2

0

5 -3

2 -3 -2

-1

21 3 3

0

Nuclear

Biomass/-gas

Wind

Gas

Coal (New)

-7 -1 -10

-13 -4

2050

Peaking

-17

2045

CSP

-16

2040

Other Storage

Coal

2035

Nuclear (new)

2030

Hydro+PS

2025

Solar PV

2020

DR

2016

-50

2050

-50

2045

-40

2040

-40

2035

-30

2030

-30

2025

-20

2020

-20

2016

-10

12 2

2

-2 -4

-4

-10


79

Energy mix


IRP 1 with higher RE cost reductions

CO2 emissions

Water usage



300

275

250

250 210

200

200

150

150

100 50

82

IRP1

76

IRP1 with further RE cost reductions PPD (Moderate)

0 2015 2020 2025 2030 2035 2040 2045 2050 Carbon Budget 80

PPD equiv.

2750 Mt

1800 Mt

920 Mt

> 2750 Mt

2720 Mt

2325 Mt

100 50

IRP1 IRP1 with further RE cost reductions

0 2015 2020 2025 2030 2035 2040 2045 2050

36 33


Total system cost: IRP1 with higher RE cost reductions would result in lower system cost Net Present Value System Cost

Total System Cost

(2016 - 2050)


bR

600 529 500

441 360

400

299

300 200

487

418

4 018

3 951

IRP1

RE cost reductions

-67 (-2%)

356

199

100 0 2015

2020

2025 IRP1

81

2030

2035

2040

2045

IRP1 with further RE cost reductions

2050






b


c

Technology learning

d


e



82

to DoE on 25 October 2018 Draft IRP 2018 IRP1 with Submitted storage, DSR and lower RE costs results in increased new-build wind, solar PV, storage and less NG


Installed capacity

Energy mix



196.8

200

7.2

177.3

180 160

20.7 0.6

400

4.4 11.8

63.0

0.6

87.3

80 51.6

40 20

60.5

5.1 13.1 6.3

107 (27%)

77

271

2 44 17 4 14

160 (40%)

122

18 4 18 5.1 12.8 1.9

13.6 9.9

16.9

Solar PV

Peaking

Coal (New)

2050

Nuclear

14

2045

Hydro+PS

2040

Biomass/-gas

0

2035

Nuclear (new)

17 (4%) 7 (2%) 38 (9%) 66 (17%)

103

2030

Wind

202

2025

Other Storage

Coal

209

2020

Gas

100

2016

CSP

2050

2045

2040

2035

2030

DR


(1%)

4

63.8

45.6

31.7

2025

2020

83

2016

0

1.5 57.8 1.5 5.1 3.4 1.9 37.4

399 4

31

200

4.8 0.6 17.1 0.5 15.3 5.1 8.7 1.2 1.9

341

320

249

41.7

100

298

300

112.8

120

60

361

147.2

140

383


6.5 GW 2.1 GW 1.9 GW 1.1 GW

Submitted to DoE on 25 October 2018 Difference in installed capacity and energy mix Risk-adjusted scenario relative to IRP1, higher wind, PV & storage, less NG


Installed capacity

Energy mix Difference - Electricity production [TWh/yr]

Difference - Total installed capacity (net) [GW] 60

7

40

60

21

40

26

20

4 23

12

20 13 5 5 1 -2

0

5

7 -3 -8 -4

-3 -9 -10

-20

0

9 1

15

-4 -5

-17

44

4

-20

1

-17

-11 -27

-10

-40

-40

-4

Solar PV

Peaking

Coal (New)

2050

Nuclear

2045

Hydro+PS

2040

Biomass/-gas

2035

Nuclear (new)

2030

Wind

2025

Other Storage

Coal

2020

Gas

-60

2016

CSP

2050

2045

DR


84

2040

2035

2030

2025

2020

2016

-60


IRP 1 with storage, DSR and higher RE cost reductions

CO2 emissions

Water usage



300

275

250

250 210

200

200

150

150

100 50

82

IRP1

72

IRP1 Risk-adjusted scenario PPD (Moderate)

0 2015 2020 2025 2030 2035 2040 2045 2050 Carbon Budget 85

PPD equiv.

2750 Mt

1800 Mt

920 Mt

> 2750 Mt

2720 Mt

2325 Mt

100 50

IRP1 IRP1 Risk-adjusted scenario

0 2015 2020 2025 2030 2035 2040 2045 2050

36 32


Total system cost: Risk-adjusted scenario results in lower system cost than IRP1 Net Present Value System Cost

Total System Cost

(2016 - 2050)


bR

600 529 500

441 360

400

299

300 200

477

413

4 018

3 937

IRP1

RiskAdjusted

-81 (-2%)

354

199

100 0 2015

2020

2025

2030 IRP1

86

2035

Risk-Adjusted

2040

2045

2050



Draft IRP 2018 IRP1: 2030 Exemplary Week under Draft IRP 2018 IRP1

Demand and Supply in GW

60 50

40 30 20 10 0 Monday

87

Tuesday

Wednesday

Thursday

Curtailed solar PV and wind

Biomass/-gas

Wind

Gas

DR

Solar PV

Hydro+PS

Coal

Other Storage

CSP

Peaking

Nuclear

Sources: CSIR analysis

Friday

Saturday

Electricity demand

Sunday


Risk-Adjusted Scenario: 2030 Exemplary Week under Risk-Adjusted Scenario


60 50

40 30 20 10 0 Monday

88

Tuesday

Wednesday

Thursday


Biomass/-gas

Wind

Gas

DR

Solar PV

Hydro+PS

Coal

Other Storage

CSP

Peaking

Nuclear


Friday

Saturday

Electricity demand

Sunday




80 70 60 50 40 30 20 10 0 Monday

89

Tuesday

Wednesday

Thursday


Biomass/-gas

Wind

Gas

DR

Solar PV

Hydro+PS

Coal

Other Storage

CSP

Peaking

Nuclear


Friday

Saturday

Electricity demand

Sunday


Risk-Adjusted: 2040 Exemplary Week under Risk-Adjusted Scenario


80 70 60 50 40 30 20 10 0 Monday

90

Tuesday

Wednesday

Thursday


Biomass/-gas

Wind

Gas

DR

Solar PV

Hydro+PS

Coal

Other Storage

CSP

Peaking

Nuclear


Friday

Saturday

Electricity demand

Sunday




90 80 70

60 50 40 30 20 10 0 Monday

91

Tuesday

Wednesday

Thursday


Biomass/-gas

Wind

Gas

DR

Solar PV

Hydro+PS

Coal

Other Storage

CSP

Peaking

Nuclear


Friday

Saturday

Electricity demand

Sunday


Risk-Adjusted: 2050 Exemplary Week under Risk-Adjusted Scenario


90 80 70

60 50 40 30 20 10 0 Monday

92

Tuesday

Wednesday

Thursday


Biomass/-gas

Wind

Gas

DR

Solar PV

Hydro+PS

Coal

Other Storage

CSP

Peaking

Nuclear


Friday

Saturday

Electricity demand

Sunday





b


c

Technology learning

d


e

Existing coal fleet performance e.a IRP1 with low coal fleet performance e.b Risk-adjusted with low coal fleet performance


93





b


c

Technology learning

d


e



94


Eskom existing fleet performance – EAF (scenario) If the existing coal fleet does not recover to the expected “Moderate“ EAF used in the Draft IRP 2018 “Low“ EAF of existing coal fleet is considered to test Energy Availability Factor (EAF) [%]

Low

Moderate

High

90 88

86 84 82

81

83

83

80

80

84 82 80

80 80

78

75

76 74

74

72

72 70

0 2016

95

2018

2020

2022

2024

2026

2028

2030

2032

2034

2036

2038

2040

2042

2044

2046

2048

2050

to DoE on 25 October 2018 Draft IRP 2018 IRP1 with Submitted Low EAF requires earlier new-build around 2023 and increased absolute levels of new-build by 2030

Installed capacity and electricity supplied from 2016 to 2050 for IRP1 with low coal fleet EAF

Installed capacity

Energy mix



200 180

400 158.6

160

144.1

140

0.6

120

80

68.3 57.8 51.6 1.5 1.5 5.1 3.4 1.9 37.4

40 20

0.6 18.3 0.5 16.6 5.1 11.1 6.3

299

300

271 249

24.4

18.5

1.9

12.1

100

18 13 15

208

186

20.9

1.9

16.9

9.9

Solar PV

Peaking

Coal (New)

102

72 (18%)

2050

Nuclear

34 (8%) 4 (1%) 59 (15%)

2045

Hydro+PS

170 (43%)

35 3 35 15

2040

Biomass/-gas

2035

Nuclear (new)

2030

Wind

2025

Other Storage

Coal

2020

Gas

2016

CSP

2050

2045

2040

2035

2030

2025

DR

0

4

109

7.6

7.6

(1%) 56 (14%)

2

48

200

398 4

57 33

39.1


96

341

320

30.5 62.5

31.7

2020

2016

0

101.7

92.0

100

60

0.6 32.7

127.1

360

380


0.2 GW 3.6 GW 2.0 GW

to DoE on 25 October 2018 Difference in capacity andSubmitted energy mix with low EAF relative to IRP1, increased level of new-build, built earlier to cater for lower coal supply

Installed capacity and electricity supplied from 2016 to 2050 for IRP1 with low coal fleet EAF

Installed capacity

Energy mix



50

50

40

40

30

30

20

20

10

6 3

0

2 -3

1

4 2

1

-4 -3

-10

-10

Peaking

Coal (New)

2050

Solar PV

-7

2045

Nuclear

-11

2040

Hydro+PS

10

-5

2035

Biomass/-gas

2030

Nuclear (new)

2025

Wind

2020

Other Storage

Coal

2 6

-21

2016

Gas

2050

2045

2040

-50

2035

-50

2030

-40

2025

-40

2020

-30

2016

-30

CSP

5 5

0

-20


97

10

-20

DR

11


IRP 1 with low coal fleet EAF

CO2 emissions

Water usage



300

275

250

250 210

200

200

150

150

100 50

82 82

IRP1

50

IRP1 with low EAF PPD (Moderate)

0 2015 2020 2025 2030 2035 2040 2045 2050 Carbon Budget 98

PPD equiv.

100

2750 Mt

1800 Mt

920 Mt

> 2750 Mt

2720 Mt

2325 Mt

IRP1 IRP1 with low EAF

0 2015 2020 2025 2030 2035 2040 2045 2050

36 36


Total system cost: IRP1 with low EAF higher system cost than IRP1 due to additional capacity required Net Present Value System Cost

Total System Cost

(2016 - 2050)


bR

600 529 500

441

299

300 200

445

360

400

4 018

4 067

IRP1

IRP1 Low EAF

+49 (+1.2%)

371

199

100 0 2015

2020

2025

2030 IRP1

99

2035 IRP1 Low EAF

2040

2045

2050






b


c

Technology learning

d


e



100

Submitted to DoE on 25 October 2018 Risk-adjusted scenario with Low EAF requires earlier new-build around 2023 too and increased absolute levels of new-build by 2030


Installed capacity

Energy mix



200

191.0 174.9

180

400

361

0.7

160

149.2

140

0.6

116.9

120

80 60

0.6 20.1 0.5 21.4

69.1 57.8 51.6 1.5 1.5 5.1 3.4 1.9 37.4

40 20

5.1 13.7 6.3

5.1 13.2 1.9

12.1 9.9

16.9

2

Solar PV

Peaking

Coal (New)

2050

Nuclear

2045

Hydro+PS

14

2040

Biomass/-gas

0

2035

Nuclear (new)

17 (4%) 7 (2%) 33 (8%) 66 (17%)

90

2030

Wind

180

2025

Other Storage

Coal

208

169 (42%)

135

17 4 18

2020

Gas

104 (26%)

77

17 4 15

2016

CSP

2050

2045

2040

2035

2030

2025

DR

100

(1%)

4

61

67.9

51.8

31.7


101

271 249

41.4

399 4

36

200

5.1 9.4 1.2 1.9

2020

2016

0

298

300

97.1

100

340

320 61.0

382


0.4 GW 0.2 GW 1.9 GW

to DoE on 25 October 2018 Difference in capacity andSubmitted energy mix with low EAF relative to IRP1, increased level of new-build, built earlier to cater for lower coal supply


Installed capacity

Energy mix



70

70

60

7

50

50 40

21

4 12

30 20 0

24

14

7

9

-3 -7 -4

-2

-10

-3 -9 -12

-20

41

10

Peaking

Coal (New)

2050

Solar PV

2045

Nuclear

2040

Hydro+PS

2035

Biomass/-gas

2030

Nuclear (new)

-4

2025

Wind

-11 -32

2020

Other Storage

Coal

-17

-18

-23

2016

Gas

2050

2045

2040

-70

2035

-70

2030

-60

2025

-60

2020

-50

2016

-50

9 4

1

-4 -27

-20 -40

CSP

1

-10

-40

28

19

0

-30

DR

14

20

-30


23

40 30

12

5 8

10

102

60


Risk-adjusted scenario with low coal fleet EAF

CO2 emissions

Water usage



300

275

250

250 210

200

200

150

150

100 50

82

IRP1

71

Risk-adjusted with low EAF PPD (Moderate)

0 2015 2020 2025 2030 2035 2040 2045 2050 Carbon Budget 103

PPD equiv.

2750 Mt

1800 Mt

920 Mt

> 2750 Mt

2720 Mt

2325 Mt

100 50

IRP1 Risk-adjusted with low EAF

0 2015 2020 2025 2030 2035 2040 2045 2050

36 31


Total system cost: Risk-Adjusted with low EAF results in higher system cost than mod EAF due to additional capacity required pre-2035 Net Present Value System Cost

Total System Cost

(2016 - 2050)


bR

600 500

441

466

360

400

299

300 200

529 477 3 937

3 974

413

+37 (+1%)

354

199

100 0 2015

2020 IRP1

104

2025

2030

2035

Risk-Adjusted Mod EAF

2040

2045

Risk-Adjusted Low EAF

2050

Risk-Adj Risk-Adj Mod Low EAF EAF




1 Executive Summary 2 Background 3 Draft IRP 2018 scenarios 4 Draft IRP 2018 employment impacts 5 Draft IRP 2018 energy planning risks 5.1 Scenarios 5.2 Descriptive comments a

Technology new-build limits

b

New-build and under-construction coal capacity

c

Natural gas

d

Planning for embedded generation

e

Demand profile

6 System services and technical considerations

105





b


c

Natural gas

d


e

Demand profile


106


Draft IRP 2018 investigated implications of unconstrained least-cost (IRP1) but RE new-build limits are maintained for all other scenarios “The scenario without new-build limits provides the least-cost option by 2030“ [DoE, Draft IRP 2018, pp. 34 of 75]

“Imposing new-build limits on RE will not affect the total installed capacity and the energy mix for the period up to 2030“ [DoE, Draft IRP 2018, pp. 34 of 75]

“The scenario without RE annual build limits provides the least-cost option by 2050“ [DoE, Draft IRP 2018, pp. 35 of 75]

“The scenario without RE annual build limits provides the least-cost electricity path to 2050“ [DoE, Draft IRP 2018, pp. 35 of 75]

Why new-build limits (on any technology – needs justification)? Could be various reasons • Import/transport link limitations (infrastructure – ports, roads) • Industry ability to deliver (skills, development, construction) • Available Tx/Dx networks to evacuate power • System security/stability 107


Draft IRP 2018 RE annual new-build limits have thusfar not been justified

New-build limits on technologies means no more than these limits are allowed to be built in any given year Limits have been applied to two technologies (others unlimited): Solar PV Wind Limits are constant as power system grows No justification provided for these limits

108

Sources: Scatec; Hopefield


Solar PV is limited to 1000 MW annually resulting in a move from 2.5% of peak demand in 2020 to 1.7% of peak demand by 2050 Absolute

Relative

Max. new-build capacity, annually [MW]

Peak demand [MW]

Max. new-build capacity, annually [% of peak demand]

4 000

80 000

8

80 000

3 000

60 000

6

60 000

2 000

40 000

4

40 000

1 000

20 000

2

20 000 2.7

0

2016

2020

2025

2030

2035

2040

2045

2050

IRP_2018_Median Maximum allowed new-build solar PV (annually)

109

Peak demand [MW]

Sources: Eskom; Draft IRP 2018; CSIR analysis

0

0

2016

2.5

2.2

2.1

2.0

1.8

1.7

1.7

2020

2025

2030

2035

2040

2045

2050

0


Wind is limited to 1600 MW annually resulting in a move from 4.0% of peak demand in 2020 to 2.7% of peak demand by 2050 Absolute

Relative

Max. new-build capacity, annually [MW]

Peak demand [MW]

Peak demand [MW]

4 000

80 000

8

80 000

3 000

60 000

6

60 000

2 000

40 000

4

40 000

1 000

20 000

2 4.3

0

2016

2020

2025

2030

2035

2040

2045

IRP_2018_Median Maximum allowed new-build wind (annually)

110

Max. new-build capacity, annually [% of peak demand]

Sources: Eskom; Draft IRP 2018; CSIR analysis

2050

0

0

2016

4.0

2020

3.6

2025

20 000

3.3

3.1

3.0

2.8

2.7

2030

2035

2040

2045

2050

0


Already happening: Both leader, follower and 2nd wave countries installing more new solar PV per year than South Africa’s IRP limits for 2030/2050

New-build solar PV capacity, annually [% of peak demand]

Germany Spain

17% 10%

9%

9%

Italy UK Australia 8%

RSA new-build limits in 2030 and 2050

India

6%

6%

6%

4%

3% 2% 1%

1%

1%

1%

2007

111

2008

2009

1% 1%

2010

3% 3% 2%

2% 2%

1%

4% 4%

4%

4% 3%

2%

Follower 2nd wave

South Africa

5% 5% 5%

5%

4%

Follower

Japan China

7%

7%

4%

Leader

3% 3% 3% 2%

2%

2%

1%

1%

2011

1%

1% 1% 0%

1% 1%

2012

3%

2013

Sources: SolarPowerEurope; CIGRE; websites of System Operators; IRP 2016 Draft; CSIR analysis

2014

3% 2%

2% 1%

2%

2% 1%

2015

2%

2%

1%

2016

2017

RSA’s IRP relative new-build limit decreases over time 2030 (2.1%) 2050 (1.7%)


Already happening: Both leader and follower countries are installing more new wind capacity per year than South Africa’s IRP limits for 2030/2050

New-build wind capacity, annually [% of peak demand]

Germany Spain Ireland

9%

China

8%

South Africa

7%

5%

4% 4%

3%

3%

3% 3% 3%

3%

2%

2006

2% 2% 2%

2% 1%

1%

2007

2% 2% 2%

2%

2%

2% 2%

112

4%

4%

3%

3%

2009

2011

Sources: GWEC; CIGRE; websites of System Operators; IRP 2016 Draft; CSIR analysis

5% 5% 4%

4%

RSA’s IRP relative new-build limit decreases over time

5%

3% 3%

3% 2%

1%

1%

2010

2% 2%

1% 0%

2008

6%

6%

5% 4%

2012

Follower

Brazil

7%

6% 6%

5%

India

8%

RSA new-build limits in 2030 and 2050

1%

Leader

2% 2% 1%

2013

2%

2% 1%

1%

1%

1%

2014

3% 3% 3%

2015

2016

2017

2030 (3.3%) 2050 (2.7%)


Solar PV penetration in leading countries already up to 1.3x levels expected in Draft IRP 2018 (constrained scenarios) by 2050 Total solar PV capacity relative to system peak demand 100% Germany Leader Spain Italy UK Australia Follower Japan China Follower 2nd wave India South Africa South Africa IRP 2016 Base Case Draft IRP 2018 (IRP1) Draft IRP 2018 (Others) Draft IRP 2018 (Recommended)

90% 80% 70% 54

60% 50%

36

40%

31 23

30%

22 19

20%

12

10%

4

0% 2005

2010

2015

2020

2025

2030

Year 113

Sources: SolarPowerEurope; CIGRE; websites of System Operators; IRP 2016 Draft; CSIR analysis

2035

2040

2045

2050


Wind penetration in leading countries is already at levels up to 1.4x Draft IRP 2018 (constrained scenarios) by 2050 Total wind capacity relative to system peak load 100% Germany Leader Spain Ireland China Follower India Brazil South Africa South Africa IRP 2016 Base Case Draft IRP 2018 (IRP1) Draft IRP 2018 (Others) Draft IRP 2018 (Recommended)

90% 80%

70 67

70%

60

60% 50% 40%

27

30%

21 20

20% 6

10% 0% 2005

2010

2015

2020

2025

2030

Year 114

Sources: GWEC; CIGRE; websites of System Operators; IRP 2016 Draft; CSIR analysis

2035

2040

2045

2050


Some historical perspective - installed capacity reveals the considerable coal build-out South Africa pursued previously Installed capacity [GW] 59

60

58

55

50

49

47 44

45 41 40

42

41

37

35 30

Wind

27

CSP

24

25

22

Solar PV

20

16 15

Biomass/-gas Nuclear

5 0

Coal 1965

1975

Sources: Draft IRP 2018; CSIR analysis

115

Gas Peaking

9

10

Hydro+PS

1985

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050


Initial smaller coal, followed by large 6-pack build-out, more recently Medupi & Kusile – most decommissions in Draft IRP 2018 time horizon Installed capacity [GW]

Arnot

60

Camden Duvha Grootvlei

50

Hendrina Kendal 42

40 35

35

35

Komati

40

Kriel

36

Kusile 32

32 30

Lethabo MajubaDry

26

MajubaWet Matimba

20

18

Matla

17

Medupi

13 10

10

Tutuka Coal IPPs

7

Sasol Other 0

1965

1975


116

1985

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050


South Africa embarked on a significant new-build capacity programme previously… in coal – why not now in any other technologies? New-build coal capacity, annually [MW] 3 048

2 433 2 433

1 895 1 793

1 725

1 481

1 411

1 303

1 178 1 178 1 135 1 150 1 150 1 150 1 050 950 847

931 555 555 555

117

746 640

575

559

722 711 706

610 670 670 670 610 610

187

100 100

1965

744

1 148

1970

Sources: Draft IRP 2018

1975

1980

1985

1990

1995

2000

2005

2010

2015

2020





b


c

Natural gas

d


e

Demand profile


118


Even currently under-construction and committed capacity will be part of planned decommissioning and will require new-build Installed capacity [GW] 60

57

58

Wind

55

CSP

50 45

Solar PV

49

47

Hydro+PS

44

Gas

42

Peaking

40

Biomass/-gas 35

Nuclear

30

Coal

27

25

22

20

16 15 10 5 0

2010

2015


119

2020

2025

2030

2035

2040

2045

2050


New-build largely driven by the planned decommissioning of the existing coal fleet - mostly post 2030 Installed capacity [GW]

Arnot

60

Camden Duvha Grootvlei Hendrina

50

-9.9 (-24%)

42 40 35

-24.7 (-59%)

-31.7 (-76%)

Kendal Komati

40

Kriel

-14.8 (-47%)

36

Kusile

32 30

Lethabo MajubaDry

26

MajubaWet -7.0 (-41%)

20

Matimba Matla

17

Medupi

13 10

10

Tutuka Coal IPPs

Sasol Other 0

2010

2015


120

2020

2025

2030

2035

2040

2045

2050


Should we freely optimise the decommissioning of the existing coal fleet or finish the last 2 units at Kusile? Should we build new coal capacity in South Africa? Should a freely optimised existing coal fleet be decommissioned any earlier/later? Should the last 2 units at Kusile be completed in light of alternatives? This has been studied in previous work Steyn, G., Burton, J. & Steenkamp, M. Eskom’s financial crisis and the viability of coal-fired power in South Africa. (2017). Wright, J. G., Calitz, J., Bischof-Niemz, T. & Mushwana, C. The long-term viability of coal for power generation in South Africa (Technical Report as part of "Eskom’s financial crisis and the viability of coal-fired power in South Africa). (2017). Summary of outcomes are presented here for reference

121


System Alternate Value (SAV) approach attempts to obtain the present value of a power station over the full time horizon PS included

PS excluded

Equivalent value [R/kWh]

Total discounted system costs [ZAR-billion]

-

Total (with PS costs incl.)

PS costs

Total (with PS costs excl.)

Total (with PS costs incl.)

PS costs

Total (with PS costs excl.)

Value difference [ZAR-billion]

./.

Total energy (discounted) [TWh]

Total energy (PS)

122

Total energy (PS)

Coal

Peaking

CSP

Pumped storage

Nuclear

Hydro

Solar PV

Li-ion

Gas

Wind

Biomass/-gas

DR

Energy difference [TWh]


Early decommissioning of oldest selected coal-fired power stations would result in significant savings and a cheaper power system

[R/kWh]

SAV

Will cost 0.11-0.17 R/kWh more from Grootvlei if not decommissioned early

High demand SAV

0.61 Max. allowed equivalent capex cost of remaining 2 Kusile units

0.58

0.57

0.57 0.53

0.50

0.50

0.49

0.47

0.48

0.48

0.47 Low-demand SAV

0.41 0.34

0.36 0.31

0.31

0.33

Power station(s)

0.23

0.22

Levelised costs Env. Retrofit Capex Env. Retrofit O&M Fuel Kusile

Arnot

Complete 6 units vs. 4 units

Decom. Early Off from 1Apr20

Source: Wright (2017), Steyn (2017)

123

Camden Decom. Early Off from 1Apr18

Grootvlei Decom. Early Off from 1Apr18

Hendrina Decom. Early Off from 1Apr19

Komati

GrHeKo

Decom. all 3 early

Decom. Early Off from 1Apr20

Water cost VOM FOM





b


c

Natural gas

d


e

Demand profile


124


Need for natural gas: Understanding domestic natural gas options Natural gas source not critical in IRP at this stage but... whether it is domestic/imported is important What is the energy security risk of importing all of the natural gas required in Draft IRP 2018 scenarios?

Sources: PetroSA; Eskom; Excellerate

125


The role of natural gas in the energy mix (likely via imported LNG at this stage) is relatively small to 2030 (2-5%) and up to 15% by 2050

2030 IRP1

2040

2050

313 10 (3%)

353 28 (8%)

IRP3

47 (12%)

313 15 (5%)

353 29 (8%)

IRP5

355 14 (4%)

IRP6 15 (5%)

352 37 (10%) 352 17 (5%)

Rec.

391 59 (15%)

311 7 (2%)

388 19 (5%)

313

IRP7

393 40 (10%)

313 15 (5%)

392

390 31 (8%)

313 12 (4%) Energy [TWh]

Energy [TWh] NG

Energy [TWh]

Other

Domestic sources of flexibility and/or storage could replace some/all of these relatively small volumes of (imported) natural gas in the electricity sector e.g. UCG, CBM, shale gas, offshore, hydrogen, DSR, biomass/-gas, CSP, storage (pumped, batteries)s 126

NG – Natural gas; Sources: Draft IRP 2018


Total system cost contribution of NG fuel requirements (likely via imported LNG at this stage) is 2-4% to 2030 and up to 10% by 2050

2030 IRP1

350 (97%)

2040 360

IRP1

10 (3%) IRP3

361 367

473 14 (3%)

347 (96%)

362

477 37 (8%)

365 (98%)

373

495 17 (3%)

364 (97%)

375

580 60 (10%)

IRP7

8 (2%)

564 20 (3%)

IRP6

15 (4%)

560 40 (7%)

IRP5

15 (4%)

Rec.

466 29 (6%)

351 (96%)

IRP7

529 47 (9%)

IRP3

15 (4%)

IRP6

442 28 (6%)

346 (96%)

IRP5

2050

591 31 (5%)

Rec.

12 (3%) Total system cost [bR/yr]

Total system cost [bR/yr]

NG (fuel)

NG – Natural gas; Sources: Draft IRP 2018

127

Other

Total system cost [bR/yr]





b


c

Natural gas

d


e

Demand profile


128


Just 5 years ago – power generation capacity was concentrated around Mpumalanga (coal) with some hydro, peaking and nuclear

129

Sources: Eskom; CSIR analysis


By 2018 – generation capacity (albeit still small) has already started to distribute across the country - not only in Mpumalanga anymore

130

Sources: Eskom; CSIR analysis


IRP meets energy demand for RSA – expressed as equivalent demand (assumes EG as reduced demand) 1

1

2

Eskom1

IPPs1

(Gx)

(Gx)

3

Imports

Exports

1

Net sent out

2

Imports

3

Exports

Available for = 1 + + 2+ - 3 distribution in RSA

Eskom (Tx2)

1

Eskom

Munics.

(Dx2)

(Dx2)

Customers

EG

(Dom/Com/Ind)

(Gx)

Other (Gx)

Customers

EG

(Dom/Com/Ind)

(Gx)

1

Large customers

Other (Gx)

EG = Embedded Generation; Gx = Generation; Tx = Transmission; Dx = Distribution 1 Power generated less power station load; Minus pumping load (Eskom owned pumped storage); 2 Transmission/distribution networks incur losses before delivery to customers

131

Submitted to DoE on 25 October 2018 From Jan-Dec 2016, 234 TWh of net electricity was sent out in SA with imports of almost 11 TWh means system load was 245 TWh

Actuals captured in wholesale market for Jan-Dec 2016 (i.e. without embedded plants)

Annual electricity in TWh 234.2 18.2

10.6

244.8

16.5

216.0

228.2 Net Sent Out Other Eskom + IPPs

Imports

System Load (domestic and export load)

Exports

Available for distribution in RSA

Notes: “Net Sent Out“ = Total domestic generation (less auxillary load) minus pumping load of Eskom pumped storage stations (not shown seperately) Sources: Eskom; Statistics South Africa

132


IRP 2018 expects demand growth to be more certain, slower - average growth from 2016 of 1.2-2.0%/yr to 2030, 1.2-1.7%/yr to 2050

Sources: StatsSA; Draft IRP 2018

133


Lower demand forecast implies 3 things: Increased EE, fuel switching and... Embedded Generation – but how much? ”Due to the limited data at present and for the purpose of this IRP Update, these developments were not modelled as standalone scenarios, but considered to be covered in the low-demand scenario. The assumption was that the impact of these would be lower demand in relation to the median forecast demand projection.” [DoE, Draft IRP 2018, pp. 21 of 75]

In the Draft IRP 2018, it is clearly stated that the relative to the Median demand forecast, the Lower demand forecast is representative of a combination of: Embedded Generation (likely mostly solar PV) Energy efficiency (EE) Fuel switching Growth of embedded generation (EG) market being implicitly included can then be calculated based on: Share of EG in the difference between Median and Lower demand forecasts Almost all EG will likely be solar PV (with associated capacity factor)

134


Between 1.1 - 5.8 GW of embedded generation (assumed to be PV dominated) is implicitly considered in the Draft IRP 2018 by 2030 2016

2030

2030

(TWh)

(TWh)

(MW)

313 5

290

1100-1500 MW (80-100 MW/yr) 245

0

0

245

20% is EG 313 9

290

3300 – 4400 MW (230-310 MW/yr)

Median demand forecast EE/fuel switching

60% is EG

EG Lower demand forecast

313 19

290

4400 - 5800 MW (310-420 MW/yr) Sources: StatsSA; Draft IRP 2018; CSIR analysis

135

80% is EG





b


c

Natural gas

d


e

Demand profile


136


Demand profile is assumed unchanged in Draft IRP 2018 – updated aproached to demand profile will need to be pursued in future

Demand profile will change as constituent components of the demand forecast change

Not just purely an energy demand forecast and fitted peak load (assuming similar demand profile)

Updated approaches will be required linking with EG to ensure sufficiently accurate capacity expansion

137


Change in demand profile will result in very different capacity expansion options Future

Today Weekday

EG

138

Agri

Weekend

Dom

ComMan

Min

Weekday

Trans

Weekend



1 Executive Summary 2 Background 3 Draft IRP 2018 scenarios 4 Draft IRP 2018 employment impacts 5 Draft IRP 2018 energy planning risks 6 System services and technical considerations 6.1 Networks 6.2 System services

139



1 Executive Summary 2 Background 3 Draft IRP 2018 scenarios 4 Draft IRP 2018 employment impacts 5 Draft IRP 2018 energy planning risks 6 System services and technical considerations 6.1 Networks 6.2 System services

140


Good to include shallow grid connection costs explicitly in Draft IRP 2018 based on extensive grid planning experience at Eskom Draft IRP 2018 includes collector network costs in the various Eskom Customer Load Networks (CLNs) as well as shallow grid connection costs for VRE (solar PV and wind) and all other technologies This is a welcome inclusion and is a good starting point to start to incorporate technology specific network costs previously not considered Good objectives: • Avoid premature congestion at Eskom Main Transmission Substations (MTSs) • Minimising absolute number of MTSs • Connect more smaller VRE plants in a specific area • Allow more orderly network development and increased utilisation i.e. more efficient integration Network costs based on unitised costing of individual equipment Transmission substations, Satellite stations, Transformers, Transmission/Distribution bays, Overhead lines, Static VAr Compensators (SVCs) 141


Deep network costs are not included in Irp but covered as part of periodic and well established TDP and SGP developed by Eskom Deep network costs (backbone strengthening) not included in Draft IRP 2018 (or any previous IRP iteration) but instead established based on periodic TDPs and SGPs developed by Eskom Grid Planning What is most important is how these deep network costs change on a relative basis across scenarios Generally, these costs do not change significantly across scenarios and thus would not materially impact least-cost outcomes In future iterations of the IRP, there should be a continued pursuit to include geospatial components of supply, networks and demand side options (co-optimisation) where feasible and tractable

TDP – Transmission Development Plan; SGP – Strategic Grid Plan

142



1 Executive Summary 2 Background 3 Draft IRP 2018 scenarios 4 Draft IRP 2018 employment impacts 5 Draft IRP 2018 energy planning risks 6 System services and technical considerations 6.1 Networks 6.2 System services a

143

RoCoF and frequency stability



1 Executive Summary 2 Background 3 Draft IRP 2018 scenarios 4 Draft IRP 2018 employment impacts 5 Draft IRP 2018 energy planning risks 6 System services and technical considerations 6.1 Networks 6.2 System services a

144

RoCoF and frequency stability


System services and ensuring security of supply to enable any range of future energy mixes System services and security of supply Frequency stability/control (system inertia and RoCoF) – particular focus in these comments Transient stability and fault level (system strength) Voltage and reactive power control Variable resource forecasting

All of these should inform a programme of research and effort required by all key stakeholders to ensure any future energy mix is secure

145


System operators are managing high non-synchronous penetration – Ireland SNSP limits and services to manage low inertia power systems

SNSP [%] = System Non-Synchronous Penetration = (Wind/PV + Imports)/(Demand + Exports)

146

Sources: David Cashman (EirGrid), DS3 and beyond, 2017, IEA Working Group


Averaging window is important – for frequency stability typically a 500 ms averaging window for RoCoF is considered

The RocoF should not exceed a particular threshold within the pre-defined averaging window e.g. 500 ms 147

Sources: EirGrid, SONI


The demand for system inertia is driven by two assumptions: the maximum allowable RoCoF & the largest assumed system contingency

Key assumptions: Maximum allowed 𝑅𝑜𝐶𝑜𝐹: 0.5 Hz/s Largest contingency (𝑃𝑐𝑜𝑛𝑡 ): 2 400 MW Kinetic energy lost in contingency event 𝐸𝑘𝑖𝑛(𝑐𝑜𝑛𝑡.) : 5 000 MWs 𝐸𝑘𝑖𝑛.(𝑚𝑖𝑛)

𝑓𝑛 = 𝑃𝑐𝑜𝑛𝑡. + 𝐸𝑘𝑖𝑛(𝑐𝑜𝑛𝑡.) 2(𝑅𝑜𝐶𝑜𝐹)

Term “inertia” is used a bit loosely to describe the amount of kinetic energy that is stored in the rotating masses of all synchronously connected power generators (and loads to be precise) 𝑓𝑛 = System frequency = 50 Hz

148

Sources: P. Kundur, Power System Stability and Control, 1994

Demand for inertia

124 800 MW.s of system inertia are required at any given point in time in order for RoCoF to stay below 0.5 Hz/s in the first 500 ms after the largest system contingency occurred


As a starting point – we have assessed system inertia on an hourly basis via UCED in PLEXOS and some high level assumptions

Technology Coal (old) Coal (new) OCGT/ICE CCGT/CC-GE Biomass Hydro/PS Imports Nuclear Wind PV CSP DR

149

Inertia constant [MWs/MVA] 4.0 2.0 6.0 6.0 2.0 3.0 0.0 5.0 0.0 0.0 2.0 0.0

Sources: P. Kundur, Power System Stability and Control, 1994

Supply of inertia Depending on what mix of power stations is operational at any given point in time, the total actual system inertia will be different For example, if 20 GW of old coal, 10 GW of new coal and 2 GW of nuclear are online, system inertia is: ≈20 GW * 4 MWs/MVA + 10 GW * 2 MWs/MVA + 2 GW * 5 MWs/MVA = 110 000 MWs If wind, PV and 5 GW of CCGTs are online, system inertia is only 47 000 MW.s


SNSP levels in IRP1 of Draft IRP 2018 only above 25% from 2028 and 37% by 2030 for 10% of the time but above 80% by 2050

150


System synchronous energy for IRP1 – confirmation that the power system really does only start to change until after 2030

151


Minimal cost of ensuring acceptable RoCoF levels even at very high non-synchronous generation penetration levels

2016

2020

2025

2028

2030

2040

2050

104 482

104 482

104 482

104 482

104 482

104 482

104 482

104 482

116 485

116 609

122 260

104 992

55 832

31 907

M in im u m in e rtia n e e d e d

[ M W .s ]

M in im u m in e rtia ( a ctu a l)

[ M W .s ]

A d d itio n a l in e rtia n e e d e d

[ M W .s ]

N u m b e r o f h o u rs

[ h rs ]

S h a re o f h o u rs

[% ]

R o ta tin g s ta b ilis e rs n e e d e d

[M W ]

-

-

-

-

A n n u a l co s t f o r ro ta tin g s ta b ilis e rs

[ b R / y r]

-

-

-

-

(% o f sy ste m co sts)

[% ]

24 0.3%

0.0%

-

-

-

-

48 650

72 575

-

-

-

-

1 799

2 282

20.5%

26.1%

-

1 220

1 810

-

3.7

5.6

0.9%

1.2%

0.0%

0.0%

0.0%

0.0%

0.0%

0.0%

0.0%

0.0%

The worst-case cost to ensure RoCoF levels are acceptable post2030 is at most ≈1% of total system costs

152


Thank you

153