Setting up for the 2020s - africaleadership.net · South Africa had the worst year of loadshedding on record in 2019 (1352 GWh, 530 hours) with up to ...

1

Setting up for the 2020sAddressing South Africa’s electricity crisis and getting ready for the next decade

CSIR Energy CentrePretoria. January 2020v1.0

DR JARRAD WRIGHT JOANNE CALITZ

2

45Mitigation options/solutions5

39What if?4

30Current plans (IRP 2019)3

18The burning platform2

2Executive Summary1

Agenda

3

What is being addressed in this contribution?

1 As announced by the President of South Africa (March 2019); 2 Crises are (a) sources of uncertainty, disruption, and change, (b) harmful/threatening for organisations and

stakeholders, (c) behavioural phenomena (socially constructed by actors involved) (d) parts of larger processes (rather than discrete events); 2 As an individual household, as a

business, as a regulator, as a policymaker/decision-maker; 3 It is not intended to address other topics in the electricity industry in South Africa e.g. financial sustainability & operational

efficiency of institutions, political economy of stakeholders interacting on the basis of relevant interests.

Sources: Bundy

Focus is to analyse, interpret and present an evidence-base that can assist in supporting custodians & stakeholders to ensure short-term system adequacy3

Is South Africa in an electricity crisis1,2?

How long will loadshedding last?

What options are available to solve theelectricity crisis3?

What is happening now and what willhappen in the next 3-5 years?

Assessing whether this is an electricity crisis (andthe extent thereof)

Present two scenarios to understand theexpected intensity & duration of loadshedding

Proposing feasible options/solutions to alleviatethe crisis and testing their impact

What is CSIR addressing?

https://www.fin24.com/Economy/Eskom/ramaphosa-we-will-overcome-the-electricity-crisis-20190321-2

4

Why is CSIR making this contribution?

1 Public domain information; 2 As with previous analyses in this domain e.g. Draft IRP 2016, Draft IRP 2018, IRP 2019 (amongst others); 3 As announced by the President of South Africa (March 2019); 4 DMRE RMPPPP RfI is in the public domain (closing date 31 January 2020) with RfP to follow (publication for all custodians/stakeholders will inform responses, reduce risk and allow for more informed procurement)

• CSIR is an independent scientific research institution with no vested interests in outcomes

Independence

• CSIR has (over the years) developed & collated necessary data, analytics and modelling frameworks1

• CSIR publish all analysis in this domain and intend to continue (public interest)2

• Providing an evidence-base to inform and aid planning for all custodians/stakeholders

Transparency

• CSIR is highly proficient/competent in this domain with well respected capabilities• CSIR is well positioned to undertake this suite of analysis as a publicly trusted independent institution

Experience/expertise

• South Africa is in an electricity crisis3

• CSIR is already engaging with key custodians/stakeholders & intends to formally engage further to assistin driving expedited responses/actions by various custodians/stakeholders4

Expedience

https://www.fin24.com/Economy/Eskom/ramaphosa-we-will-overcome-the-electricity-crisis-20190321-2

5

The burning platform & need for an urgent response requires a widely understood evidence-base for all custodians/stakeholders to take action

1 Even if Medupi, Kusile, REIPPPP capacity under construction comes online as expected; EAF – Energy Availability Factor

South Africa had the worst year of loadshedding on record in 2019 (1352 GWh, 530 hours) with up toStage 6 load shedding being implemented having significant impacts on the economy (≈R 60-120 bln)

Loadshedding is expected to continue for 2-3 years depending on key decisions/actions

An urgent response is necessary to ensure short-term adequacy and set South Africa on a path towardslong-term adequacy in the 2020s

Systemic changes in Eskom fleet performance (EAF) expectation and demand forecast requires anupdated understanding of capacity & energy gap relative to IRP 2019 assumptions1 (as will be shown)

6

Notes: Load shedding assumed to have taken place for the full hours in which it was implemented. Practically, load shedding (and the Stage) may occassionally change/ end during a

particular hour; Total GWh calculated assuming Stage 1 = 1 000 MW, Stage 2 = 2 000 MW, Stage 3 = 3 000 MW, Stage 4 = 4 000 MW, Stage 5 = 5 000 MW, Stage 6 = 6 000 MW;

Cost to the economy of load shedding is estimated using COUE (cost of unserved energy) = 87.50 R/kWh

Sources: Eskom Twitter account; Eskom se Push (mobile app); Nersa; CSIR analysis

400

0

200

600

1 600

800

1 000

1 200

1 400

Load shed [GWh]

2007 2008

568

45

123

2009 2010 2011 20202012 2013 2014

874

406

2015 2016

203

2017

80

1 325

130

62

618

43

30

476

93

20192018

192

1 352

143

17126

176

Unknown Stage 4Stage 5Stage 6 Stage 3 Stage 2 Stage 1

Year

2007

2008

….

2014

2015

2016

2017

2018

2019

2020 (YTD)

Duration

of outages

(hours)

-

-

.…

121

852

-

-

127

530

80

Energy

shed

(GWh)

176

476

.…

203

1325

-

-

192

1352

143

2019 was the most intensive year of loadshedding to date in South Africa with Stage 6 being implemented in December 2019

Est. econ.

Impact

(ZAR-bln)

8-15

21-42

.…

9-18

58-116

-

-

8-17

59-118

6-12

7

System inadequacy has been highlighted recently by DMRE and Eskom – but no specific solutions/interventions provided (yet)

Integrated Resource Plan 2019 (IRP 2019)

• Released: October 2019

• Highlighted short-term supply gap between 2019-2022

• Primarily due to lead-time of first new-build capacity in2023

Medium-Term System Adequacy Outlook 2019 (MTSAO 2019)

• Released: November 2019

• Also highlighted lack of system adequacy if Eskom fleetEAF is below 72% for time horizon 2019-2024

• No new investments made in time-horizon

NOTE: Different assumptions made in a few dimensions for IRP 2019 and MTSAO 2019 (demand forecast, EAF, supply) but same conclusions reached.

Sources: DMRE; Eskom

8

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030

70

94

0

68

62

64

86

80

82

72

90

98

78

100

88

66

84

74

96

76

92

67.0%

83.1

EAF[%]

73.6

80.0

75.5

Historical fleet EAF decline seems irreversable, IRP 2018 EAF has not materialised... risk of IRP 2019 or MTSAO 2019 EAF (High) not materialising?

Actual

MTSAO 2019 (MES 1, High)

Low (IRP 2018)

High (IRP 2018)

Moderate (IRP 2018)

IRP 2019

MTSAO 2019 (MES 1, Low)


Planned

EAF – Energy Availability Factor; MTSAO – Medium-term System Adequacy Outlook

NOTE: 2019 EAF actual is YTD

Sources: Draft IRP 2018; IRP2019; Eskom; CSIR Energy Centre analysis

9

Not explored

Worst scenario

IRP 2019

(DMRE)

Updated

(CSIR)

Not explored

Best scenario

What if the demand forecast is lower than in IRP 2019 (very likely) and what if the Eskom fleet EAF is lower than in IRP 2019 (also very likely)?

HighEAF [%]

Sources: CSIR Energy Centre analysis

Demand forecast [TWh]

Low

Low

High

10

Would an updated EAF expectation and demand forecast make any difference to capacity and energy shortages?

NOTE: “Updated” scenario is a test scenario developed by CSIR; Demand forecast is based on Eskom MTSAO demand forecast (until 2024) and IRP 2019 growth rates thereafter;

Updated EAF based on MTSAO MES 1 (Low); EAF – Energy Availability Factor

Sources: IRP 2019; MTSAO; CSIR

Demand forecastEAF

2000 2005 2010 2015 2020 2025 2030

80

60

85

0

65

70

90

95

75

100

64.7

Energy Availability Factor (EAF)[%]

67.0

75.5

2000 2005 2010 2015 2020 2025 2030

150

200

50

250

300

350

0

400

100

246

267

306

Electrical energy demand [TWh]

298

284

Updated

IRP 2019

Actual

IRP 2019

Updated

Actual

11

4

1

0

3

2

5

6

7

8

9

102

02

1

[GW]

20

18

20

19

20

20

20

22

20

23

20

24

20

25

Shortage from IRP 2019 indicating a dominant short-term capacity gap & small energy gap until planned new-build capacity comes online

* Estimated Eskom Demand Response (DR) capability (mostly industrial & energy limited); NOTES: Energy & capacity shortage is demand that cannot be served due to a lack of

capacity (including OCGTs, pumped storage & Eskom DR); Outcomes shown are from deterministic simulations - thus indicative; 99th percentile of capacity & energy shortage is

reported; All IRP 2019 capacity is assumed to come online as planned (Step 3 is always considered implemented); Cost of load shedding is estimated using COUE (cost of unserved

energy) = 87.50 R/kWh; Sources: CSIR Energy Centre analysis

Capacity (shortage) Energy (shortage)

2 000

0

500

1 000

1 500

2 500

3 000

3 500

4 000

4 500

5 000

20

25

20

20

20

19

[GWh/yr]

20

18

20

21

20

22

20

23

20

24

Supply gap (from IRP 2019)


IRP 2019 EAF & IRP 2019 demand forecast(EAF recovery from ≈67% in 2019 to 75.5% by 2024) (Demand forecast immediately growing to 284 TWh by 2025)

Stage 1

Stage 6

Stage 5

Stage 4

Stage 3

Stage 2

Eskom DR*

Loadshedding (actual)

2019: 1352 GWh

(R60-120bln cost to the economy)

NOTE: Understatement of energy shortages

in simulation (conservative estimates)

Stage 7

Stage 8

12

500

3 000

0

1 000

1 500

2 000

2 500

3 500

4 000

5 000

4 500

20

21

20

24

20

23

[GWh/yr]

20

18

20

25

20

19

20

20

20

22

Updated EAF & Updated demand forecast(EAF from ≈67% in 2019 to ≈64% by 2024) (Demand forecast initially flat & growth to 267 TWh by 2025)

6

0

1

2

3

4

5

7

8

9

10

20

23

20

19

20

21

20

18

[GW]

20

20

20

22

20

24

20

25

Updated EAF & demand forecast indicates further shortage relative to IRP 2019 requiring capacity and significantly more energy









2019: 1352 GWh




Stage 1

Stage 6

Stage 5

Stage 4

Stage 3

Stage 2

Eskom DR*

Stage 7

Stage 8

13

Criteria to decide on options to meet the short-term gap should be informed and applied consistently to ensure reasonable cost & timeous delivery

Criteria for choice of short-term options available can ensure a portfolio of options:

Can be supply-side, demand-side and/or storage

Can be delivered in 1-2 years

Will not require extensive procurement process (lead-time)

Can meet capacity (MW) and/or energy needs (MWh)

Can be contracted for 1-3 years (or more if aligned with long-term energy mix)

Ease of implementation (does not require extensive regulatory reform/change)

Aligned with long-term energy mix pathways (technology choices)

Does not require extensive network expansion or augmentation for interconnection

1

2

3

4

5

6

7

8


14

01/2020 01/202401/2021 01/2022 01/2023 01/2025

Critical decisions/actions needed now along with accelerated processes to ensure timeous implementation allowing RSA to ramp into 2020s successfully

RfP (response)PBs (COD)

IPPPP1 (constr.)

RfP (release)

SSEG2 & DG regulations3

REIPPPP PPA neg.REIPPPP PPA (operations)

Comms/incentives (SSEG)

RfI (response)

PBs (FC)

Min. Determinations

IPPPP1 (release)

PBs

IPPPP1 (response)

IPPPP1 (PBs)

IPPPP1 (constr.)

PBs (constr.)

IPPPP1(FC)

PBs (operations)

IPPPP1 (operations)

Cu

sto

me

r re

spo

nse

at

sca

leSt

ep 1

DM

RE

RM

PP

PP

Step

2

Imp

lem

ent

IRP

20

19

4

Step

3

Capacity1

Decision/action

Stakeholders/custodians

Capacity1

Decision/action

Stakeholder/custodian

Capacity1

Decision/action

Stakeholder/custodian

NOTES: Timelines are estimated and in no way prescriptive; PBs – Preferred bidders, PPA – Power Purchase Agreement; RfI – Request for Information; RfP – Request for Proposal; FC – Financial

Close, COD – Commercial Operations Date; 1 Total additional installed capacity; 2 Requires adjusted SSEG regulations (proposed lifting licensing requirement for SSEG & only requiring registration

with NERSA - from 1 MW to 10 MW (or more)); 2 SSEG (res.) does not require regulatory changes (just communications rollout but could be further incentivised by Eskom/municipalities); 3 Will require

Ministerial Determination - generators expected >10 MW (technologies aligned with IRP 2019); 4 Unlikely to get capacity online before 2023-2024 (risk of misalignment with IRP 2019)

1.2 GW (supply/storage) 2.6 GW (supply/storage) 4.9 GW (supply/storage)

1.3-2.8 GW (supply) 0.1-1.9 GW (supply) 0.1-1.6 GW (supply) 0.1-1.6 GW (supply) 1.2-3.9 GW (supply)

1.6 GW (wind), 1.0 GW (PV), 0.5 GW (storage)

3.2 GW (wind), 2.0 GW (PV), 0.5 GW (storage),

0.5 GW (DG/EG), 0.75 GW (coal)





Custodians: IPPO/DBSA, DMRE, NERSA, Eskom (DPE), NT Stakeholders: Customers (businesses, households), municipalities, IPPs

4.9 GW (supply/storage) 4.9 GW (supply/storage) 4.9 GW (supply/storage)

Custodians: IPPO/DBSA, DMRE, NERSA, Eskom (DPE), NT Stakeholders: Customers (businesses, households), municipalities, IPPs, financial institutions

Custodians: IPPO/DBSA, DMRE, NERSA, Eskom (DPE), NT Stakeholders: Customers (businesses, households), municipalities, IPPs

This timeline is ambitous (but achievable) - means mostcapacity will only come online after it is needed (in 2021) –accelerated processes are necessary

15

Step 1 is the only immediate feasible response as Step 2 & 3 are expected from 2022/2023 only (best case) but still critical to ensure system adequacyKey recommendations

SSEG – Small-Scale Embedded Generation; DG – Distributed Generation; EG – Embedded Generation; RMPPPP – Risk Mitigation Power Purchase Procurement Programme

Immediate focus on customer response at scale(self-supply) in all customer segments via enablingregulations (easy to implement)Driven by SSEG (residential), EG (commercial/agricultural), EG/DG(industrial/mining), municipalities & storage, REIPPPP ‘power-up’

Accelerate DMRE RMPPPP process to addressremaining capacity & energy gap and ensure capacitycan come online timeouslyAn accelerated process necessary due to immediate shortages &should be based on estimated required capacity (complementingStep 1)

Immediate focus on Ministerial Determinations forall technologies in IRP 2019 followed byprocurement process to ensure timeousimplementationDue to procurement processes & technology specific lead-times thisaction/decision is required now

NOTE: Even with short-term interventions, if EAF does not recover to IRP 2019 levels, shortage is expected in 2020-2021 depending on capacity thatcan feasibly come online, structural load shedding may still need to be considered for 2-3 years (shortages exaggerated in Updated scenario)

• Immediate reduced load shedding as capacity cancome online in 2020 already

• Further assistance from 2021 onwards as morecapacity comes online

• Capacity online from 2022 only (mid-2021 withaccelerated DMRE RMPPPP process)

• Further reduction in load shedding once capacitycomes online

• First capacity online from 2023 only (best case)but required in 2022 (as per IRP 2019)

• Adequate power system into mid-2020s asexisting capacity is decommissioned if IRP 2019planned new-build capacity comes online

DECISION/ACTION IMPACT

Cu

sto

me

r re

spo

nse

at

sca

leSt

ep 1

DM

RE

RM

PP

PP

Step

2

Imp

lem

ent

IRP

20

19

4

Step

3

16

This is a crisis - get capacity under construction online, recover Eskom plant performance AND implement all steps with urgencyKey recommendations

Ensure capacity under construction is delivered as planned (Medupi/Kusile, REIPPPP)

Recover Eskom fleet EAF to realistic levels whilst ensuring value for money relative to alternatives

Implement/enable 3 steps concurrently and with urgency:

• Step 1: Intentionally drive a customer response at scale (with enabling regulations) driven by SSEG(residential), EG (commercial/agricultural), EG/DG (industrial/mining) & storage1

• Step 2: Address remaining capacity/energy gap via an accelerated DMRE RMPPPP process to ensurecapacity is online when required

• Step 3: Continued implementation of IRP 2019 as an immediate focus to ensure sufficient lead-timefor procurement processes and technology specific construction lead-times

1 As shown, these are the only options/solutions that would assist in mitigating load shedding in the next 2-3 years (other options/solutions would take further time to implement).

SSEG – Small-Scale Embedded Generation; DG – Distributed Generation; EG – Embedded Generation; RMPPPP – Risk Mitigation Power Purchase Procurement Programme;

NOTE: For further details – please refer to the remainder of this presentation; SSEG could include a range of technologies but would be dominated by solar PV.

17

Urgency required on actions/decisions now to ensure capacity comes online timeouslyKey recommendations

In order to enable and implement recommended steps, first immediate decisions/actions are needed:

• DMRE/Nersa to update regulations to enable streamlined & expedited self-supply options1

• DMRE/Nersa to publish Ministerial Determinations and/or update EG/DG regulations toenable DMRE RMPPPP additional capacity

• DMRE/Nersa to publish Ministerial Determinations aligned with IRP 2019

• IPPO engage & implement feasible existing REIPPPP PPA negotiations (REIPPPP ‘power-up’)

• Various stakeholders to drive intentional communications program and/or incentives forSSEG deployment (res./com./agri. focus)2

• IPPO to run bid windows for new-build procurement (technologies aligned with IRP 2019)and undertake annually going forward

IPPO – Independent Power Producers Office; DMRE – Department of Mineral Resources and Energy; res. – residential; com. – commercial.1 Adjusted/updated SSEG regulations (proposed lifting licensing requirement for SSEG & only requiring registration with NERSA - from 1 MW to 10 MW (or more)); 2 SSEG (res./com.)

does not require regulatory changes (just communications rollout but could be further incentivised by Eskom/municipalities with appropriate tariffs).

Q1-2020

18


39What if?4



2Executive Summary1

Agenda

19





400

0

200

600

1 600

800

1 000

1 200

1 400

Load shed [GWh]

2007 2008

568

45

123

2009 2010 2011 20202012 2013 2014

874

406

2015 2016

203

2017

80

1 325

130

62

618

43

30

476

93

20192018

192

1 352

143

17126

176

Unknown Stage 4Stage 5Stage 6 Stage 3 Stage 2 Stage 1

Year

2007

2008

….

2014

2015

2016

2017

2018

2019

2020 (YTD)

Duration

of outages

(hours)

-

-

.…

121

852

-

-

127

530

80

Energy

shed

(GWh)

176

476

.…

203

1325

-

-

192

1352

143

2019 has been the most intensive year of loadshedding to date with Stage 6 being implemented in December 2019

Est. econ.

Impact

(ZAR-bln)

8-15

21-42

.…

9-18

58-116

-

-

8-17

59-118

6-12

20

Hourly distribution of actual load shedding January to June 2019

January February AprilMarch May June

Day 1

Day 31

0h00 24h00 0h00 24h00 0h00 24h00 0h00 24h00 0h00 24h00 0h00 24h00

No load shedding





21

Hourly distribution of actual load shedding July to December 2019

July August OctoberSeptember November December

Day 1

Day 31

0h00 24h00 0h00 24h00 0h00 24h00 0h00 24h00 0h00 24h00 0h00 24h00

No load shedding





22

0

5

20

15

10

25

30

35

40

GW

5

3

0

1

4

2

6

GW

Severely constrained system from 4 December 2019, compounded by further breakdowns up to 9 December 2019 resulting in Stage 6 loadsheddingActual hourly production from all power supply sources in RSA for December 2019

NOTES: Pumping load excluded; UCLF – Unplanned Capacity Loss Factor; OCGTs – Open-Cycle Gas TurbinesSources: Eskom; CSIR Energy Centre analysis

Day

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Coal

Solar PV

Load Shedding

Wind

Hydro + PS

CSP

Imports, Other

Diesel

Nuclear

Stage 6

Stage 5

Stage 4

Stage 3

Stage 2

Stage 1

High UCLF (conveyer supply, mine flooding and unitoutages) combined with depletion of emergency reserves(diesel OCGTs and pumped storage)

23

Other(Gx)

Other(Gx)

Munics.(Dx2)

South African supply-demand balance is maintained by the System Operator (Eskom) with some supply/demand responsive customers

Eskom(Tx2)

Eskom(Dx2)

Eskom1

(Gx)Imports ExportsIPPs1

(Gx)

EG = Embedded Generation; Gx = Generation; Tx = Transmission; Dx = Distribution1 Power generated less power station load; Minus pumping load (Eskom owned pumped storage); 2 Transmission/distribution networks incur losses before delivery to customers

Largecustomers

(Com/Ind)

EG(Gx)

Customers(Dom/Com/Ind)

EG(Gx)

Customers(Dom/Com/Ind)

1 1Production1

2 Imports

3 Exports

32

1 1

1 1

24

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030

66

74

100

78

84

86

80

72

90

88

82

76

98

0

94

70

96

68

92

EAF (annual)[%]

67.0%

71.5%

Historical fleet EAF decline since 2001 with apparent recovery in 2016-2017 but trend continued thereafter and is at ≈67% for 2019

Actual

Planned

EAF – Energy Availability Factor


Sources: IRP2019; Eskom; CSIR Energy Centre analysis

25

70

1-Jan-16 1-Jan-17

75

0

1-Jan-18

90

1-Jan-19 1-Jan-20

60

95

80

65

1-Jan-21

85

100

EAF (weekly)[%]

61.4%

Recent Eskom weekly fleet EAF has been declining with unfortunate consequences of a highly constrained power system

NOTES: EAF - Energy Availability FactorSources: Eskom; CSIR Energy Centre analysis

2016: 76.4% 2017: 78.7% 2018: 71.9% 2019: 67.0%

26

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030

72

70

66

82

80

84

86

88

76

64

90

74

78

94

62

0

96

98

100

68

92

73.6

EAF (annual)[%]

67.0%

80.0

83.1

75.5

Historical fleet EAF decline seems irreversable... IRP 2018 EAF has also not materialised, risk of IRP 2019 or MTSAO 2019 EAF (High) not materialising?

IRP 2019

High (IRP 2018)

Low (IRP 2018)

Moderate (IRP 2018)

Actual

MTSAO 2019 (MES 1, Low)



Planned

EAF – Energy Availability Factor; MTSAO – Medium-term System Adequacy Outlook


Sources: Draft IRP 2018; IRP2019; Eskom; CSIR Energy Centre analysis

27

The supply-demand balance must be maintained for every hour of every day utilising available supply/demand options (dispatchable & self-dispatched)

60

10

0

20

30

40

50

[GW]

20

18

53.6

Other Storage

Biomass/-gas

DG

Solar PV

CSP

Wind

Hydro

NuclearPS

Peaking

Gas

Nuclear (new)

Coal

Supply(Installed capacity)

Demand (typical week)

0

50

10

40

20

30

60

[GW]

Peak demand (35 GW)

Min. demand (20 GW)

Mo

nd

ay

Tuesd

ay

Wed

nesd

ay

Thu

rsday

Friday

Saturd

ay

Sun

day

Summer

Winter

28

Example – Lower EAF (lower coal fleet availability) would mean a more constrained power system and high-risk of loadshedding

20

0

10

30

40

50

60

[GW]

20

18

37.1

CSP

Wind

Solar PV

Other Storage

DG

Biomass/-gas

Hydro

PS

Peaking

Gas

Nuclear (new)

Nuclear

Coal

Supply(Available capacity – high outages)


Mo

nd

ay

Tuesd

ay

Wed

nesd

ay

Thu

rsday

Friday

Saturd

ay

Sun

day

EAF

40

0

60

10

20

30

50

Min. demand (20 GW)

[GW]

Peak demand (35 GW)

Summer

Winter

29

Example – Higher EAF (higher coal fleet availability) would mean a less constrained power system and low-risk of loadshedding

0

10

20

30

40

50

60

20

18

[GW]

43.6

CSPOther Storage

Biomass/-gas

Solar PV

DG Hydro

Wind

PS

Peaking

Gas

Nuclear (new)

Nuclear

Coal

Supply(Available capacity – low outages)


Mo

nd

ay

Tuesd

ay

Wed

nesd

ay

Thu

rsday

Friday

Saturd

ay

Sun

day

EAF

60

10

30

0

40

20

50

[GW]

Peak demand (35 GW)

Min. demand (20 GW)

Summer

Winter

30


39What if?4



2Executive Summary1

Agenda

31

System inadequacy has been highlighted recently by DMRE and Eskom – but no specific solutions/interventions provided

Integrated Resource Plan 2019 (IRP 2019)

• Released: October 2019

• Highlighted short-term supply gap between 2019-2022

• Primarily due to lead-time of first new-build capacity in2023

Medium-Term System Adequacy Outlook 2019 (MTSAO 2019)

• Released: November 2019

• Also highlighted lack of system adequacy if Eskom fleetEAF is below 72% for time horizon 2019-2024

• No new investments made in time-horizon

NOTE: Different assumptions made in a few dimensions for IRP 2019 and MTSAO 2019 (demand forecast, EAF, supply) but same conclusions reached.

Sources: DMRE; Eskom

32

CSIR simulated South African power system model benchmarked against actuals shows good alignment (2018 as a reference - annual)

200

0

60

160

180

20

80

220

40

140

240

100

260

280

120

6(2.3%)

3(1.4%)6(2.7%)

2018 (Simulated)

11(4.6%)

1(0.4%)

2(0.8%)

6(2.4%)

10(4.1%)

11(4.8%)

3(1.4%)

1(0.4%)

6(2.7%)2(0.8%)

11(4.4%)

Annual electricity production [TWh]

2018 (Actual)

199(82.5%)

241 241

203(84.3%)

Imports, Other

Wind

Hydro + PS

Solar PV

CSP

Diesel

Coal

Nuclear

Supply Sources

PS – Pumped Storage; HVDC – High Voltage Direct Current;

NOTES: Includes generation for pumping load; Simulation in PLEXOS: production cost model with hourly temporal resolution and public input data; “Imports, Other” includes mostly

imported hydro generation (via HVDC imports from Mozambique);

Sources: Eskom; CSIR Energy Centre analysis

33

2

0

10

14

4

8

6

12

16

18

20

JulJan May Jun

Monthly electricity production from Eskom coal in TWh

OctFeb Mar Apr Aug Sep Nov Dec

2018 (Actual) 2018 (Simulated)

CSIR simulated South African power system model benchmarked against actuals shows good alignment – coal focus (2018 as a reference - monthly)

NOTES: Simulation in PLEXOS: production cost model with hourly temporal resolution and public input data;

Sources: Eskom; CSIR Energy Centre analysis

0

60

40

80

120

100

140

160

180

200

20

220

Annual electricity production in TWh

2018 (Actual)

2018 (Simulated)

203 199 -4(-2.1%)

34

350

0

50

400

300

100

200

450

250

150

20

26

20

22

20

20

202.8169.1

20

30

29.653.0

Electricity production [TWh/yr]

3.36.514.5

3.41.9

13.4

8.2

20

18

14.92.9

20

24

13.40.92.2

20

28

0.5

11.8

20

32

20

34

20

36

20

38

20

40

20

42

20

44

20

46

20

48

20

50

248.0

311.3

IRP 2019 has been gazetted indicating an immediately inclining demand forecast, decommissioning coal fleet but with improving EAFInstalled capacity and electricity supplied from 2018 to 2050

1 Projection based on optimisation of 2030-2050 energy mix utilising input assumptions from DMRE IRP 2019 (not unconstrained least-cost);

DSR – Demand Side Response; DG = Distributed Generation; VRE – variable renewable energy;

NOTE: Energy share is a best estimate based on available data.

Sources: IRP 2019. CSIR Energy Centre analysis

Installed capacity Energy mix

40

0

20

200

140

60

180

120

80

160

100

6.2

0.6

2.1

20

20

3.4

1.5

5.1

1.9

2.1

Total installed capacity (net) [GW]

20

32

4.68.3

17.8

38.0

20

18

30.9

20

30

20

34

20

36

20

38

20

40

20

42

20

44

20

46

20

24

0.3

1.0

20

26

0.512.0

20

28

6.3

7.6

0.7

0.4

32.0

20

48

7.6

1.9

14.4

1.5

6.5

19.2

20

22

8.5

20

50

53.6

82.2

108.8

Other Storage

Biomass/-gas

DG Hydro+PSCSP

WindSolar PV Peaking

Gas

Nuclear (new)

Nuclear

Coal (New)

CoalFirst new-builds:

Wind (2022) 1.6 GWPV (2022) 1.0 GWStorage (2022) 0.5 GWCoal (2023) 0.75 GW Gas (2024) 1.0 GW

Gazetted IRP 2019 Projection1Gazetted IRP 2019Projection1

35

50

300

0

400

450

100

350

150

200

250 6.514.5

3.4

3.3

20

28

20

20

202.8

20

18

20

22

169.1

20

24

20

26

13.4

53.0

0.5

8.21.9

14.9

11.8

2.9

2.2

29.60.9

Electricity production [TWh/yr]

13.4

20

30

20

32

20

34

20

36

20

38

20

40

20

42

20

44

20

46

20

48

20

50

248.0

311.3

0

140

80

40

20

160

60

100

120

180

200

38.030.9

3.41.9

0.7

5.1

8.3

20

48

20

22

20

50

0.5

Total installed capacity (net) [GW]

4.6

1.52.1

6.2

2.1

0.617.8

1.9

20

26

0.4

7.6

20

30

20

32

20

34

20

36

20

40

20

38

20

42

20

44

20

46

20

28

1.0

20

24

53.6

12.06.3

1.5

32.0

20

18

7.6

19.2

0.3

14.4

20

20

6.5

82.2

8.5

108.8

IRP 2019 has also indicated an expected gap between 2018-2022 following which expected new-build options can come online (lead time dependant)Installed capacity and electricity supplied from 2018 to 2050

1 Projection based on optimisation of 2030-2050 energy mix utilising input assumptions from DMRE IRP 2019 (not unconstrained least-cost);

DSR – Demand Side Response; DG = Distributed Generation; VRE – variable renewable energy;

NOTE: Energy share is a best estimate based on available data.

Sources: IRP 2019. CSIR Energy Centre analysis

Installed capacity Energy mix

Biomass/-gas

Other Storage CoalCSPDG

Solar PV Wind

Hydro+PS

Peaking

Gas

Nuclear (new) Coal (New)

NuclearFirst new-builds:

Wind (2022) 1.6 GWPV (2022) 1.0 GWStorage (2022) 0.5 GWCoal (2023) 0.75 GW Gas (2024) 1.0 GW

Gazetted IRP 2019 Projection1 Projection1

Supply gap

Supply gap

Gazetted IRP 2019

36

4

1

0

3

2

5

6

7

8

9

102

02

1

[GW]

20

18

20

19

20

20

20

22

20

23

20

24

20

25

Shortage from IRP 2019 indicating a dominant short-term capacity gap & small energy gap until planned new-build capacity comes online






2 000

0

500

1 000

1 500

2 500

3 000

3 500

4 000

4 500

5 000

20

25

20

20

20

19

[GWh/yr]

20

18

20

21

20

22

20

23

20

24



IRP 2019 EAF & IRP 2019 demand forecast(EAF recovery from ≈67% in 2019 to 75.5% by 2024) (Demand forecast immediately growing to 284 TWh by 2025)

Stage 1

Stage 6

Stage 5

Stage 4

Stage 3

Stage 2

Eskom DR*


2019: 1352 GWh




Stage 7

Stage 8

37

IRP 2019 daily shortage profile shows capacity need in the morning/evening peak combined with daytime energy needs

0

3 000

6 000

9 000

12 000

15 000

18 000

21 000

24 000

27 000

30 000

33 000

36 000

0

1 000

2 000

3 000

4 000

5 000

6 000

7 000

8 000

9 000

10 000

11 000

12 000

13 000

14 000

15 000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Hour of day

Hourly shortage profile[MW]

Typical Demand profile[MW]

Summer weekend (RHS)

2019

2018

2020 2024

2022

2021

2023

2025

Winter weekday (RHS)

Winter weekend (RHS)

Summer weekday (RHS)

NOTE: Energy & capacity shortage is demand that cannot be served due to a lack of capacity (including OCGTs, pumped storage & Eskom DR); Outcomes shown are from

deterministic simulations - thus indicative; 99th percentile of capacity & energy shortage is reported.


38

0

40

10

30

20

50

60

[GW]

Week 1 Week 2

Representative 2 weeks - constrained/unconstrained showing capacity needs along with daytime energy needs during some weekdaysSimulated hourly generation stack of the total power supply in RSA for 2 weeks in April 2021

Eskom fleet EAF (week) = 64% Eskom fleet EAF (week) = 79%

System Demand

Available Capacity

Unserved Energy

DG

GasStorage

Solar PV

CSP

Wind

Biomass/gas

PS

Hydro

Diesel

Coal

Nuclear

40

0

10

20

30

50

60

20

21

Installed Capacity [GW]

57.0

Energy shortage (daytime)

Capacity shortage (morning/evening)

Pumped storage (pumping mode)

NOTE: Representative week(s) shown but entire time horizon is simulated at hourly resolution in production cost model applying unit commitment and economic dispatch principles;


39


39What if?4



2Executive Summary1

Agenda

40

Not explored

Worst scenario

IRP 2019

(DMRE)

Updated

(CSIR)

Not explored

Best scenario

What if the demand forecast is lower than in IRP 2019 (very likely) and what if the Eskom fleet EAF is lower than in IRP 2019 (also very likely)?

HighEAF [%]


Demand forecast [TWh]

Low

Low

High

41

Would an updated EAF expectation and demand forecast make any difference to capacity and energy shortages?

NOTE: “Updated” scenario is a test scenario developed by CSIR; Demand forecast is based on Eskom MTSAO demand forecast (until 2024) and IRP 2019 growth rates thereafter;

Updated EAF based on MTSAO MES 1 (Low); EAF – Energy Availability Factor

Sources: IRP 2019; MTSAO; CSIR

Demand forecastEAF

2000 2005 2010 2015 2020 2025 2030

80

60

85

0

65

70

90

95

75

100

64.7

Energy Availability Factor (EAF)[%]

67.0

75.5

2000 2005 2010 2015 2020 2025 2030

150

200

50

250

300

350

0

400

100

246

267

306

Electrical energy demand [TWh]

298

284

Updated

IRP 2019

Actual

IRP 2019

Updated

Actual

42

500

3 000

0

1 000

1 500

2 000

2 500

3 500

4 000

5 000

4 500

20

21

20

24

20

23

[GWh/yr]

20

18

20

25

20

19

20

20

20

22

Updated EAF & Updated demand forecast(EAF from ≈67% in 2019 to ≈64% by 2024) (Demand forecast initially flat & growth to 267 TWh by 2025)

6

0

1

2

3

4

5

7

8

9

10

20

23

20

19

20

21

20

18

[GW]

20

20

20

22

20

24

20

25

Updated EAF & demand forecast indicates further shortage relative to IRP 2019 requiring capacity and significantly more energy









2019: 1352 GWh




Stage 1

Stage 6

Stage 5

Stage 4

Stage 3

Stage 2

Eskom DR*

Stage 7

Stage 8

43

Updated scenario shows initial morning/evening capacity need combined with all-day energy, shifting to morning/evening need only in later years

0

3 000

6 000

9 000

12 000

15 000

18 000

21 000

24 000

27 000

30 000

33 000

0

1 000

2 000

3 000

4 000

5 000

6 000

7 000

8 000

9 000

10 000

11 000

12 000

13 000

14 000

15 000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Hour of day

Hourly shortage profile[MW]

Typical Demand profile[MW]

2018

2019

2022

2021

2020

Winter weekend (RHS)2023

2024

Summer weekend (RHS)2025

Winter weekday (RHS)

Summer weekday (RHS)

NOTE: Energy & capacity shortage is demand that cannot be served due to a lack of capacity (including OCGTs, pumped storage & Eskom DR); Outcomes shown are from

deterministic simulations - thus indicative; 99th percentile of capacity & energy shortage is reported.


44

0

10

20

30

60

50

40

[GW]

Week 1 Week 2


System Demand

Available CapacityGas

Unserved Energy Solar PV

Hydro

Storage

Biomass/gas

CSP

DG Wind

PS

Diesel

Coal

Nuclear

0

20

10

40

30

60

50


20

21

57.0

Example – Updated scenario showing significant capacity & energy shortages during representative 2 weeksSimulated hourly generation of the total power supply in RSA for 2 weeks in May 2021

Extensive daily energy shortages

Capacity shortage (morning/evening)Minimal pumped storage

generating/pumping



45


39What if?4



2Executive Summary1

Agenda

46

Criteria to decide on options to meet the short-term gap should be informed and applied consistently to ensure reasonable cost and timeous delivery

Criteria for choice of short-term options available can ensure a portfolio of options:

Can be supply-side, demand-side and/or storage

Can be delivered in 1-2 years

Will not require extensive procurement process (lead-time)

Can meet capacity (MW) and/or energy needs (MWh)

Can be contracted for 1-3 years (or more if aligned with long-term energy mix)

Ease of implementation (does not require extensive regulatory reform/change)

Aligned with long-term energy mix pathways (technology choices)

Does not require extensive network expansion or augmentation for interconnection

1

2

3

4

5

6

7

8


47

Choice of the suite of solutions/interventions to fill the gap is highly dependant on lead-times and limits resulting options available

Supply-side options

Demand-side options

NOTES: Storage deployment is considered as a demand-side option for simplicity and ease of understanding (is a net user of energy); SSEG could include a range of technologies but

would be dominated by solar PV; Gx – Generation; com. – Commercial; ind. – Industrial; min. – mining; agri. – Agricultural; res. – Residential; DSR – Demand Side Response; 1

Requires adjusted SSEG regulations (proposed lifting licensing requirement for SSEG & only requiring registration with NERSA - from 1 MW to 10 MW (or more)); 2 Does not require

regulatory changes (just communications rollout but could be further incentivised by Eskom/municipalities); 3 Will require Ministerial Determination - generators expected to be >10 MW

(technologies should be aligned with IRP 2019); 4 Will require fast-tracked procurement process; 5 In addition to capacity under construction.

Short-term (2020-2022)

Medium-term/ Long-term

(≥2023)

REIPPPP ‘power-up’200-500 MW (2020)

SSEG1 (com./ind./agri.)400-750 MW/yr

IRP 2019 (first new-build):Wind: 1 600 MW (2022)Solar PV: 1 000 MW (2022)Coal: 750 MW (2023)EG/DG: 500 MW (2023)Gas: 1 000 MW (2024)

Structural loadsheddingStage 1: 1 000 MWStage 2: 2 000 MW

IRP 2019 (first new-build):Storage: 513 MW (2022)

SSEG2 (res.)125-200 MW/yr

DG/EG3 (ind./min.)250-750 MW/yr

Standby Gx aggregator200-1 000 MW/yr

Emergency Gx4

500-2 000 MW/yr

IRP 2019 (by 2030)Wind: 14 400 MWSolar PV: 6 000 MWCoal: 1 500 MWEG/DG: 4 000 MWGas: 3 000 MW

Short-term (2020-2022)

Medium-term/ Long-term

(≥2023)

Storage - SSEG2 (res.)30-100 MW/yr / 90-300 MWh

Storage - SSEG1 (com./ind./agri.)100-375 MW/yr / 300-1125 MWh

Storage (munics)30-120 MW/yr / 90-360 MWh

DSR50-300 MW/yr

IRP 2019 (by 2030)Storage: 2 088 MW

Technical potential5

Prescribed in IRP 2019

Technical potential5

Prescribed in IRP 2019

48

Considering the options available – a customer response at scale would form the only options with fastest lead-times (likely mostly self-dispatched)

• Some solutions can be quickly implemented by customers & at scale (short lead-times)

• Immediate requirement is adjusted SSEG regulations

• Adjusted PPAs with already existing REIPPPP generation capacity

• Communications/incentives/financing for SSEG for further deployment (with requisite Eskom/municipal tariffs)

• DG/EG (ind./min.) expected to be >10 MW each, requires licensing & ministerial approval*

• Options include (from 2020 onwards):

o REIPPPP ‘power-up’: 200-500 MW

o SSEG (com./ind./agri.)1: 400-750 MW/yr

o Storage - SSEG (com./ind./agri.)1: 100-375 MW/yr / 300-1125 MWh

o SSEG (res.)2: 125-200 MW/yr

o Storage - SSEG (res.)2: 30-100 MW/yr / 90-300 MWh

o DG/EG (ind./min.)3: 250-750 MW/yr

o Storage (munics)2,3: 30-120 MW / 90-360 MWh

o Structural loadshedding: Stage 1: 1000 MW; Stage 2: 2000 MW; (or more)NOTES: Gx – Generation; com. – Commercial; ind. – Industrial; min. – mining; agri. – Agricultural; res. – Residential; DSR – Demand Side Response; * Technologies should be aligned

with IRP 2019; DSR (EWHs) could also contribute 50-150 MW/yr in the short-term but has not been explicitly considered as yet due to uncertainty; 1 Requires adjusted SSEG

regulations (proposed lifting licensing requirement for SSEG & only requiring registration with NERSA - from 1 MW to 10 MW (or more)); 2 Does not require regulatory changes (just

communications rollout but could be further incentivised by Eskom/municipalities); 3 Will require Ministerial Determination - generators expected to be >10 MW (technologies should be

aligned with IRP 2019)

Step 1 (customer response at scale)

49

Manage short-term needs with the range of options from Step 1 whilst RfI process runs in Step 2 (many dispatchable options likely to be available)

• DMRE Risk Mitigation Power Purchase Procurement Programme (RMPPPP) RfI

• RfI (and resulting RfP) should assist to fill gap following Step 1 (customer response at scale)

• Reality: Only likely to come online from 2021 onwards

• Capacity providers expected to be >10 MW - licensing & ministerial approval (technologies aligned with IRP 2019)

• Can also include some of the options in Step 1

• Options could include (from 2021 onwards):

o Standby Gx aggregator1: 200-1000 MW/yr

o Emergency Gx2,3: 500-2000 MW/yr

• Structural loadshedding: Stage 3: 3000 MW; Stage 4: 4000 MW

NOTES: Gx – Generation; com. – Commercial; ind. – Industrial; min. – mining; agri. – Agricultural; res. – Residential; DSR – Demand Side Response; 1 Requires adjusted SSEG

regulations (proposed lifting licensing requirement for SSEG & only requiring registration with NERSA - from 1 MW to 10 MW (or more)); 2 Will require Ministerial Determination -

generators expected to be >10 MW (technologies should be aligned with IRP 2019 but could include temporary/permanent engines, OCGTs, cogeneration, imports etc.);3 Will require fast-tracked procurement process.

Step 2 (addressing short-term adequacy and setting up for long-term expected energy mix

50

DMRE RfI is needed & important but is only part of the puzzle and a component of the supply/demand options that will be needed

DMRE Request for Information (RfI)

• Request for Risk Mitigation Power Purchase Procurement (Generation)

• Released: 13 December 2019

• Briefing Session: 8 January 2020

• Response: 31 January 2020

• Specification:

o Quantity: 2000 – 3000 MW

o Can include supply-side and demand-side options

o Lead-time: 3-6 months (2000 MW), 6-12 months (3000 MW)

o PPA tenure: 3 years, 5 years, 10 years, 15 year, 20 years

o Type: “Baseload Energy”, “Peaking Energy”, Mid-Merit Energy”

o Regime: “Dispatchable”, “Self-Dispatchable”

• RfP to follow once RfIs have been assessed

51

Step 3 to setup for the 2020s - immediately run procurement rounds aligned with IRP 2019 & consider reductions based on Step 1 & 2 deployments

• Immediate Ministerial Determinations aligned with IRP 2019

• Begin next procurement rounds for new capacity (lead-times necessitate this)

• Reality: Only likely to come online from 2022 onwards (more likely 2023-2024)

• Options as defined by IRP 2019 (should likely be subtracted from capacity already deployed in Step 1 and Step 2):

o Wind: 1 600 MW (2022) Wind: 14 400 MW (2030)

o Solar PV: 1 000 MW (2022) Solar PV: 6 000 MW (2030)

o Coal: 750 MW (2023) Coal: 1 500 MW (2030)

o EG/DG: 500 MW (2023) EG/DG: 4 000 MW (2030)

o Gas: 1 000 MW (2024) Gas: 3 000 MW (2030)

o Storage: 513 MW (2022) Storage: 2 088 MW (2030)

Step 3 (continued implementation of IRP 2019)

52

50

0

10

20

40

30

60

[GW]

Week 1 Week 2


System Demand

Available Capacity

Unserved Energy

Hydro

Storage

Biomass/gasSolar PV

DG

CSP

Wind Coal

Nuclear

PS

Diesel

Gas

50

0

10

40

20

30

60


20

21

59.6

IRP 2019 scenario and roll out of step 1, daytime energy shortages & diesel burn substantially reduced, but load shedding still requiredIRP 2019 with customer response at scale options (Step 1) options: Simulated hourly generation for 2 weeks in 2021

Significantly reduced daytime shortages

Increased ability to utilise pumped storageLower peaking capacity

utilisation (reduced diesel burn)

NOTE: Lower end of total estimated capacity in Step 1 is considered in scenarios); Representative week(s) shown but entire time horizon is simulated at hourly resolution in production

cost model applying unit commitment and economic dispatch principles; Sources: CSIR Energy Centre analysis

53

30

20

0

10

40

50

GW

30

0

40

10

20

50

GW

IRP 2019 comparison with customer response at scale options (Step 1) revealing notably less constrained power systemSimulated hourly generation of the total power supply in RSA for 1 week in April 2021

Mon Tue Wed Thu Fri Sat Sun

Unserved Energy

Storage

CSPDG

Solar PV Wind

Biomass/gas

PS

Hydro

Diesel

Gas

Coal

Nuclear


IRP 2019 IRP 2019 with customer response at scale (Step 1)

System Demand

Available Capacity




Significantly reduced daytime shortages

54

IRP 2019 scenario (where improved EAF occurs) that is supplemented by customer response (Step 1) will then require mostly capacity (less energy)

What is still needed to ensure system adequacy (utilising DMRE RMPPPP RfI/RfP process)? (Step 2)

• Capacity1: 1.3 GW (2021), 0.1-1.2 GW (2022-2025)

• Capacity factor <3%

NOTE:

• This scenario assumes IRP 2019 EAF & IRP 2019 demand forecast:

o EAF recovery from ≈67% in 2019 to 75.5% by 2024

o Demand forecast immediately growing to 284 TWh by 2025

• If lower customer response at scale (Step 1), additional energy will be needed (higher capacity factor)

• Structural loadshedding (Stage 3 & Stage 4) may also be necessary during 2020-2021

1 It is assumed that capacity from Step 2 will only come online in 2021 (owing to procurement process & technology lead times);


55

10

40

0

30

20

50

GW

10

0

40

30

20

50

GW

IRP 2019 comparison with customer response (Step 1) and DMRE RMPPPP (Step 2) revealing capacity needs met (notably less constrained system)Simulated hourly generation of the total power supply in RSA for 1 week in April 2021


Unserved Energy

Solar PVStorage

DG CSP

NuclearWind

Biomass/gas

PS

Hydro

DMRE

Diesel

Gas

Coal


IRP 2019 with customer response at scale (Step 1)

IRP 2019 with customer response (Step 1) and DMRE RMPPPP (Step 2)

System Demand

Available Capacity




56

30

0

40

60

20

10

50

[GW]

Week 1 Week 2


System Demand

Available Capacity

Unserved Energy

Storage PS

DG

CSP

Solar PV

Wind Hydro

Biomass/gas

Gas

Diesel Nuclear

Coal

0

10

40

20

30

50

60


59.6

20

21

Reduced daytime shortages but still extensive gap

Only evening capacity shortage



Updated scenario - with updated EAF & demand, with customer response (Step 1) reduces daytime shortages & diesel burn, but load shedding remainsUpdated scenario with customer response at scale options (Step 1): Simulated hourly generation for 2 weeks in 2021

57

30

10

0

40

20

50

GW

0

40

10

30

20

50

GW

Updated scenario shows significantly less constrained system with the roll out of customer response at scale (Step 1) Simulated hourly generation of the total power supply in RSA for 1 week in 2021


DGUnserved Energy

Solar PVStorage

CSP

Wind

Biomass/gas

PS

Hydro

Diesel

Gas

Coal

Nuclear


Updated scenario Updated scenario with customer response at scale (Step 1)

System Demand

Available Capacity




Extensive daily energy shortages


58

Updated scenario will require significant additional capacity & energy through to 2025 in addition to customer response at scale (Step 1)

Now… What is still needed to ensure system adequacy (utilising DMRE RMPPPP RfI/RfP process)? (Step 2)

• Capacity1: 2.8 GW (2021), 1.9 GW (2022) and 1.6-3.9 GW (2023-2025)

• Capacity factor 8% (2021), 10% (2022), 1-3% (2023-2025)

NOTE:

• This scenario assumes an Updated EAF & demand forecast:

o EAF from ≈67% in 2019 to ≈65% by 2025

o Demand forecast initially flat & growth to 267 TWh by 2025

• If lower customer response at scale (Step 1), additional energy will be needed (higher capacity factor)

• Structural low-level loadshedding (Stage 3 & Stage 4) may also be necessary during 2020-2021

1 It is assumed that capacity from Step 2 will only come online in 2021 (owing to procurement process & technology lead times);


59

30

20

0

10

40

50

GW

0

10

40

30

20

50

GW


Storage

Unserved Energy DieselHydroDG CSP

Solar PV Wind

CoalBiomass/gas

PS GasDMRE Nuclear


Updated scenario with customer response (Step 1) and DMRE RMPPPP (Step 2)

Eskom fleet EAF (week) = 58%

Updated scenario with customer response at scale (Step 1)

Eskom fleet EAF (week) = 58%

System Demand

Available Capacity

DMRE RMPPPP providing peaking/mid-merit capacity

Updated scenario comparison of customer response (Step 1) and additional DMRE RMPPPP (Step 2) revealing notably less constrained power systemSimulated hourly generation of the total power supply in RSA for 1 week in 2021




60

Thank you

61

References

Loadshedding data

• Eskom twitter account: https://twitter.com/Eskom_SA

• ESP (Twitter app): https://sepush.co.za/

• National Energy Regulator of South Africa (NERSA). (2008). Inquiry Into the National Electricity Supply Shortage and Load Shedding. Retrieved from http://pmg-assets.s3-website-eu-west-1.amazonaws.com/docs/080528nersa.pdf

COUE

• Department of Energy (DoE). (2019). Integrated Resource Plan (IRP 2019). Retrieved from http://www.greengazette.co.za/pages/national-gazette-37230-of-17-january-2014-vol-583_20140117-GGN-37230-003

Eskom fleet Energy Availability Factor (EAF)

• Eskom Annual Reports: http://www.eskom.co.za/OurCompany/Investors/IntegratedReports

• MTSAO: Fabricius, C., Sigwebela, N., Damba, S., Mdhluli, S., & Rambau, P. (2019). MEDIUM-TERM SYSTEM ADEQUACY OUTLOOK. Retrieved from http://www.eskom.co.za/Whatweredoing/SupplyStatus/Documents/MediumTermSystemAdequacyOutlook2019.pdf

• Eskom System Adequacy Reports: http://www.eskom.co.za/Whatweredoing/SupplyStatus


• Department of Energy (DoE). (2018). INTEGRATED RESOURCE PLAN 2018. Retrieved from http://www.energy.gov.za/IRP/irp-update-draft-report2018/IRP-Update-2018-Draft-for-Comments.pdf

https://twitter.com/Eskom_SA

https://sepush.co.za/

http://www.greengazette.co.za/pages/national-gazette-37230-of-17-january-2014-vol-583_20140117-GGN-37230-003

http://www.eskom.co.za/OurCompany/Investors/IntegratedReports

http://www.eskom.co.za/Whatweredoing/SupplyStatus/Documents/MediumTermSystemAdequacyOutlook2019.pdf

http://www.eskom.co.za/Whatweredoing/SupplyStatus


http://www.energy.gov.za/IRP/irp-update-draft-report2018/IRP-Update-2018-Draft-for-Comments.pdf

62

References

Capacity and energy mix to 2030 (IRP 2019)


DMRE Risk Mitigation Power Purchase Procurement Programme (RMPPPP) RfI

• Department of Mineral Resources and Energy (DMRE). (2019). REQUEST FOR INFORMATION IN RESPECT OF THE DESIGN OF A RISK MITIGATION POWER PROCUREMENT PROGRAMME. Retrieved from http://www.energy.gov.za/files/docs/MTPPP%20Generation%20-%20DMRE%20Request%20For%20Infromation%20(13%20December%202019).pdf


http://www.energy.gov.za/files/docs/MTPPP%20Generation%20-%20DMRE%20Request%20For%20Infromation%20(13%20December%202019).pdf

63

Disclaimer

© CSIR 2020. All rights to the intellectual property and/or contents of this document remain vested in the CSIR.

This document is issued for the sole purpose for which it is supplied. CSIR will take no responsibility or incur any

liability for the use of the information contained herein if used for any other purpose. No part of this publication

may be reproduced, stored in a retrieval system or transmitted, in any form or by means electronic, mechanical,

photocopying, recording or otherwise without the express written permission of the CSIR. It may also not be

resold, hired out or otherwise disposed of by way of trade in any form of binding or cover than that in which it is

published.