[PPT] - Variable Renewable Energy Integration and Planning Results of Task PowerPoint Presentation

SLIDE 1

A company of Confidential Restricted Public Internal

Variable Renewable Energy Integration and Planning

Results of Task 1: Dispatch Diagnosis & Task 2: Demand & Generation Forecast Analysis

Stakeholder Dissemination Event Islamabad, April 4, 2019

x

Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized

SLIDE 2

OVERVIEW

Task 1 Dispatch Diagnosis Presentation of results and open discussion Task 2 Demand and Generation Forecast Analysis Presentation of results and open discussion

2

Chapter 2 Chapter 1

SLIDE 3

Comparison Approach Task 1 versus Task 2

3

Background: Task 1 and 3 follow different approaches due to different base and purpose Task VRE input Modelling,

ptimization

Outcome For both generation profiles PLEXOS with existing and committed system

Dispatch observing constraints / costs

Task 1: Dispatch Analysis

(2017 & 2022)

Fixed wind and PV shares

(5, 10, 14 GW…)

Least-(variable) cost to serve demand for one year (2017, 2022) No: capex, expansion transmission

Technical and commercial constraints
Implications for variable costs of overall

system → variable cost reduction due to VRE Task 3: Expansion Exercise

(up to 2040)

Cost assumptions Expansion constraints (e.g. annual limits)* Least-(all) cost for demand of 22 years All capex, transmission

(but only links between zones)

Expansion all sources (candidates) within limits*

Expansion paths generation sources
Focus: VRE shares

→ optimal (least-cost) expansion with given assumptions for given scenarios**

* Reducing complex regulatory/market frame for VRE to a set of parameters **Note: A fully fledged master plan would consider many more scenarios / sets of assumptions

SLIDE 4

Task 1: Dispatch Diagnosis

Task 1: Dispatch Diagnosis Presentation of results and open discussion

▪

Results of Dispatch Simulation

4

▪

Results of Scenario Analysis

▪

Dispatch Constraints for the Integration

f VRE

Chapter 1

SLIDE 5

Comparison of Energy Mixes

5

10 GW Case (5 GW Wind, 5 GW PV)

STs, 6.6% CCGTs, 32.7% CC DGUs, 0.1% DGUs, 0.0% Nuclear, 8.7% Coal PPs, 17.5% HPPs, 24.9% Bagasse, 4.0% Wind, 4.3% PV, 1.3%

Energy Mix

Total generation: 172.8 TWh STs, 6.4% CCGTs, 29.8% CC DGUs, 0.0% DGUs, 0.0% Nuclear, 8.7% Coal PPs, 13.5% HPPs, 24.8% Bagasse, 3.6% Wind, 8.3% PV, 4.8%

Energy Mix

Total generation: 172.8 TWh

Base Case (2.5 GW wind, 1.3 GW PV)

SLIDE 6

Dispatch Comparison: Winter

6

Bagasse Coal PPs Nuclear CCGTs OCGTs STs Gas engines HPP VRE CC DGUs

10 GW Case (5 GW Wind, 5 GW PV) Base Case (2.5 GW wind, 1.3 GW PV)

SLIDE 7

Dispatch Comparison: Summer

7

Bagasse Coal PPs Nuclear CCGTs OCGTs STs Gas engines HPP VRE CC DGUs

10 GW Case (5 GW Wind, 5 GW PV) Base Case (2.5 GW wind, 1.3 GW PV)

SLIDE 8

Reserves Overview

8

Requirement will pass to 367 MW (1,100 MW), when Karachi Coastal Nuclear will be online by the end of 2021

Primary Reserve

15 sec
5 min
1/3 largest unit

200 MW

Secondary Reserve

30 sec
30 min
1/3 largest unit

200 MW

Contingency Reserve

30 min
scheduled
Largest unit

600 MW

Response: Duration: Capacity:

e

e
e
e
e
10 sec
30 sec

SLIDE 9

Reserve Constraints: Secondary Reserve

9

50 100 150 200 250 300 350 400 450 500 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126 131 136

Reserve Requirement [MW] hour

Secondary Reserve Requirement

SecRes SecRes_SizInc SecRes_Wind SecRes_PV

Secondary ResReq = ( 1 3 𝑇𝑗𝑨𝑗𝑜𝑕_𝐽𝑜𝑑𝑗𝑒𝑓𝑜𝑢)2 + 𝑋𝑗𝑜𝑒_𝑆𝑓𝑡𝑆𝑓𝑟𝑡𝑓𝑑−𝑛𝑗𝑜 ² + 𝑄𝑊_𝑆𝑓𝑡𝑆𝑓𝑟𝑡𝑓𝑑−𝑛𝑗𝑜 ²

SLIDE 10

Reserve Constraints: Contingency Reserve

10

200 400 600 800 1000 1200 1400 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126 131 136

Reserve Requirement [MW] hour

Contingency Reserve Requirement

ContRes ContRes_SizInc ContRes_Wind ContRes_PV

Contingency ResReq = 𝑇𝑗𝑨𝑗𝑜𝑕_𝐽𝑜𝑑𝑗𝑒𝑓𝑜𝑢2 + 𝑋𝑗𝑜𝑒_𝑆𝑓𝑡𝑆𝑓𝑟𝑆𝑓𝑕𝑣𝑚𝑏𝑢𝑗𝑝𝑜 ² + 𝑄𝑊_𝑆𝑓𝑡𝑆𝑓𝑟𝑆𝑓𝑕𝑣𝑚𝑏𝑢𝑗𝑝𝑜 ²

SLIDE 11

Reserve Constraints: Primary Reserve Shortfall

11

No large change between Base Case and 10 GW Case

Provision Shortfall

SLIDE 12

Reserve Constraints: Primary Reserve Provisioning

12

Bagasse Coal PPs Nuclear CCGTs OCGTs STs Gas engines HPP VRE CC DGUs

No large change between Base Case and 10 GW Case

SLIDE 13

Reserve Constraints: Secondary Reserve

13

Provision Shortfall Provision Shortfall Base Case 2.5 GW Wind 1.3 GW PV 10 GW Case 5.0 GW Wind 5.0 GW PV

SLIDE 14

Reserve Constraints: Fuel Contracts

14

Simulations for 2022 have shown that without fuel contracts the around 116 MUSD could be saved.

Category Variable Generation Costs [MUSD] with Fuel Contracts without Fuel Contracts difference

STs 2,119.1 2,137.4 18.3 0.9% CC DGUs 20.5 21.2 0.7 3.5% DGUs 4.9 5.3 0.4 8.7% Gas engines 0.1 0.1 0.0 3.0% CCGTs 4,822.6 4,615.5

207.1
4.3%

Nuclear 134.2 134.2 0.0 0.0% Coal PPs 1,516.5 1,588.4 71.9 4.7% TOTAL 8,617.9 8,502.2

115.7
1.3%

SLIDE 15

Task 1: Dispatch Diagnosis

Task 1: Dispatch Diagnosis Presentation of results and open discussion

▪

Results of Dispatch Simulation

15

▪

Results of Scenario Analysis

▪

Dispatch Constraints for the Integration

f VRE

Chapter 1

SLIDE 16

Scenario Analysis: Analysed Cases

16

PV Wind Total VRE Base 1,286 2,450 3,736 3 GW each 3,000 3,000 6,000 5 GW each 5,000 5,000 10,000 7 GW each 7,000 7,000 14,000 9 GW each 9,000 9,000 18,000 11 GW each 11,000 11,000 22,000 Installed Capacity [MW] Cases

Bagasse capacity: 1,300 MW within all cases

SLIDE 17

Scenario Analysis: Energy Mix

17

20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 200,000 Base 3 GW each 5 GW each 7 GW each 9 GW each 11 GW each

Generation [GWh]

Energy Mix per scenario

HPPs STs Nuclear Coal PPs CCGTs CC DGUs DGUs PV Wind Bagasse

SLIDE 18

Scenario Analysis: Generation per Scenario

18

10,000 20,000 30,000 40,000 50,000 60,000

STs CCGTs CC DGUs DGUs Nuclear Coal PPs HPPs VRE

f which

Wind

f which

PV Bagasse

Generation [GWh]

Generation per scenario

Base 3 GW each 5 GW each 7 GW each 9 GW each 11 GW each

SLIDE 19

Scenario Analysis: VRE share and Curtailment

19

Same curtailment for wind and PV

0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 5,000 10,000 15,000 20,000 25,000

Share of VRE [%] Installed VRE capacity [MW]

Potential & Actual VRE share in energy mix

Potential VRE share Actual VRE share Potential VRE share Actual VRE share 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 10,000 20,000 30,000 40,000 50,000 60,000 Base 3 GW each 5 GW each 7 GW each 9 GW each 11 GW each

Curtailment Rate [%] Generation/Curtailment [GWh]

VRE Generation & Curtailment per Scenario

VRE Generation VRE Curtailment Curtaiment Rate

SLIDE 20

20

Category Variable Generation Costs [MUSD] Base 5 GW each difference

STs 2,119.1 2,095.5

23.5
1.1%

CC DGUs 20.5 2.9

17.6
86.0%

DGUs 4.9 0.6

4.3
86.8%

Gas engines 0.1 0.0

0.1 -100.0%

CCGTs 4,822.6 4,544.0

278.6
5.8%

Nuclear 134.2 134.0

0.2
0.2%

Coal PPs 1,516.5 1,215.7

300.9
19.8%

TOTAL 8,617.9 7,992.6

625.3
7.3%

Scenario Analysis: Variable Generation Cost

SLIDE 21

Scenario Analysis: System Short Run Marginal Cost

21

▪ Steep cost reduction for low VRE shares due to large savings of expensive thermal generation ▪ Declining cost reduction due to increased reserve requirements and curtailment

40.00 42.00 44.00 46.00 48.00 50.00 52.00 5,000 10,000 15,000 20,000 25,000

SRMC [USD/MWh] Installed VRE [MW]

System Short-Run Marginal Cost (SRMC)

SLIDE 22

Task 1: Dispatch Diagnosis

Task 1: Dispatch Diagnosis Presentation of results and open discussion

▪

Results of Dispatch Simulation

22

▪

Results of Scenario Analysis

▪

Dispatch Constraints for the Integration

f VRE

Chapter 1

SLIDE 23

VRE Dispatch Constraints: Overview

23

Transmission

SCADA 3
Capacity

limitations

Reserves
Responsibilities
Geographical

distribution

NTDC <> DISCOs

Reserves

Primary
Secondary
Contingency

Generation

CCGTs
Coal Power
Steam turbines
Nuclear Power
Hydro Power
Combustion

engines

Bagasse

Commercial, Legal, Regulatory, Policy

Take-Or-Pay

Agreements

Ramping
Nuclear
Bagasse
Policy

framework: Federal <> Provincial legislation

SLIDE 24

▪ Already today the reserve requirements cannot be entirely met. ▪ When Karachi Nuclear will be online the demand for reserve will go up to 1,100 MW, which represents the sizing incident, with more VRE in the system the requirement will further increase. ▪ In modern power systems, short-term reserve does cover the entire sizing incident (1,100 MW), so that the power system can ride through the loss of the largest generator without having to shed load for keeping frequency stable (today: Grid Code foresees only 1/3). ▪ The SCADA 3 system should be commissioned as early as possible to allow the Automatic Generation Control (AGC) to automatically dispatch all power plants. ▪ More power plants should participate in the provision of primary reserve (e.g. DGUs PPA flexibility restrictions to be renegotiated, nuclear technology to be assessed for flexibility, latest with new SCADA 3) and all power plants should be able to participate in secondary reserve. With this, a much larger share of reserve requirement can be supplied. ▪ Policy and investments for central VRE forecasting at NPCC are recommended as it would reduce the VRE forecasting error and hence may allow for lower reserve requirements.

Reserve Constraints

24

Primary Reserve

15 sec
5 min
1/3 largest unit

200 MW

Secondary Reserve

30 sec
30 min
1/3 largest unit

200 MW

Contingency Reserve

30 min
scheduled
Largest unit

600 MW Response: Duration: Capacity:

e

e
e
e
e
10 sec
30 sec

SLIDE 25

Generation Constraints

25

20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 200,000 Base 3 GW each 5 GW each 7 GW each 9 GW each 11 GW each Generation [GWh]

Energy Mix per scenario

HPPs STs Nuclear Coal PPs CCGTs CC DGUs DGUs PV Wind Bagasse

▪ More economic thermal power plants should participate in the provision of reserves, so that steam turbines (STs) can be switched off while assuring the provision of the reserve to keep the system stable. ▪ The cost of providing reserves could be brought down by 90% if steam turbines would not have to be dispatched for the provision of reserves. ▪ Potentially, a battery could be used for the provision of primary reserve requirement so that STs can be switched off. ▪ The contractual arrangements of gensets should be revised, so that gensets can provide flexibility to the power system. Simulations have revealed that flexible gensets may lead to system variable cost reductions of 7.6 MUSD per year. ▪ NPPs should be made more flexible in their operation not only running them at maximum load. ▪ NPPs should contribute to the provisioning of primary, secondary and contingency reserve. ▪ Simulations have revealed that flexible nuclear power plants (operation also in partial load to provide reserves) can cater for annual system variable cost reductions of 190.7 MUSD.

SLIDE 26

▪ Take-or-pay agreements require flexibility in the remaining power generation sources. ▪ Equally, the integration of increasing shares of VRE generation requires flexibility. ▪ Due the priority dispatch of VRE, however, VRE generation may possibly displace the generation of plants with take-or-pay obligation. This, however, triggers penalty costs for the violation of fuel-offtake quantities. => Possibly revise the existing take-or-pay agreements to lower the guaranteed fuel offtake quantities. For new power plants: Consider their installation without binding take-or-pay agreements ▪ Ramping capability of power plants is restricted in PPAs. These should be amended to benefit of the full technical ramping capability => reserves requirement can be better met. ▪ Though technically possible, partial load of NPPs commercially not favourable for NPPs (lower energy and higher costs) => Clarification on potential NPP partial load operation and potential additional benefit / income by provision of reserves

Commercial Constraints

26

60% 65% 70% 75% 80% 85% 90% 95% 100% 105% 110% BALLOKI BHIKI HBS Haveli Bahadar Shah Trimun Jang PORT QASIM COAL 1 PORT QASIM COAL 2 SAHIWAL COAL 1 SAHIWAL COAL 2

Fuel Offtake compared to guaranteed quantity

Base 6 GW 10 GW 14 GW 18 GW 22 GW

SLIDE 27

OVERVIEW

Task 1 Dispatch Diagnosis Presentation of results and open discussion Task 2 Demand and Generation Forecast Analysis Presentation of results and open discussion

27

Chapter 2 Chapter 1

SLIDE 28

Task 2 - Demand & Generation Forecast Analysis

Main questions

▪ Are the forecasts suitable (for the project, i.e. long term base demand)? ▪ Is there anything to further enhance forecasts immediately? What to improve beyond project? Demand ▪ What are special demand patterns / shapes with regard to VRE and general dispatch? What specific topics to analyse, review?

▪ Load shedding / suppressed demand ▪ Forecasted load curves per zone (North, Midlands, South) ▪ Subjects raised: net-metering, DSM….

Generation ▪ How is generation capacity expected to develop (resources, committed plants, delays,…)? ▪ Is there sufficient capacity to meet the peak demand throughout the future years, is there a considerable surplus expected?

28

SLIDE 29

Task 2 - Demand Forecast Analysis – Review of NTDC Forecasts

▪ Are the forecasts suitable (for the project, i.e. long term base demand)? Forecast review sample: past forecasts versus actual demand – Regression (long-term)

29

NTDC Regression forecasts versus actual demand 10/11 – 15/16

„The perfect forecast“

SLIDE 30

Task 2 - Demand Forecast Analysis – Review of NTDC Forecasts

▪ Are the forecasts suitable (for the project, i.e. long term base demand)? Concluding from analysing approach, underlying assumptions and forecast results

▪ The NTDC forecasts provide very accurate results to forecast future electricity demand for the NTDC system on a national level. ▪ They are considered suitable in terms of approach as well as data and assumptions as a basis for future expansion plans and this study.

30

SLIDE 31

100,000 200,000 300,000 400,000 500,000 600,000 Energy (sent-out) [GWh] Historic electricity demand Energy (sent-out) recorded (without LM) Historic electricity demand Energy (sent-out) computed (with LM) NTDC PMS Jun2018 "Normal Base Scenario" Energy (sent-out power plants) recorded (without LM) NTDC Regression Aug2018 "Normal Scenario" Energy (sent-out) computed (with LM) Adaped for project (Task 3) Energy (sent-out) , higher 2017/18 values and more conservative loss reduction Adaped for project (Task 3) Energy (sent-out) "actual" incl. shift from LM to served load (0% in 2027)

Forecast Assumptions for Project – Energy

▪ 4-7% growth per year (average 6%) ▪ 5-8% with current Load Management Regional slowly transferred to served demand (average 6.5%) ▪ This means annually the additional need for sent-out energy of an equivalent of a 2,000 MW conventional power plant (growing) (for comparison: in past years the additional energy need was ~ 1.300 MW power plant/a)

31

Energy demand historic, NTDC forecasts and adapted forecast 2007 – 2040

SLIDE 32

Forecast Assumptions for Project – Peak Demand

32

Peak demand 20012 – 2035

Annual growth during next 10 years by

n average

> 2,000 MW per year (for comparison: in past years between 600 and 3,000 MW)

SLIDE 33

Special Demand Patterns with regard to VRE and General Dispatch

Consumption (sales) ▪ Domestic & commercial consumption increase fastest ▪ Industry & agriculture shares shrink → Implications for load curve / load factor

33

Average growth rates by consumer groups Shares (of total consumption) by consumer groups

1969 2016

SLIDE 34

Special Demand Patterns with regard to VRE and General Dispatch

Load curves - daily & seasonal (relevant for dispatch & VRE integration) ▪ Distinguished day peak in summer → relevant for PV and net-metering → Relevant for potential regulatory intervention (time of use tariff)

34

Average load curves per month 2016/17

SLIDE 35

Special Demand Patterns with regard to VRE and General Dispatch

Load curves - daily & seasonal (relevant for dispatch & VRE integration) ▪ Distinguished day peak in summer → relevant for PV and net-metering ▪ This pattern has considerably developed in recent years

35

Plotting of hour of peak load 2016/17

July June

SLIDE 36

Special Demand Patterns with regard to VRE and General Dispatch

Absolute frequency of loads (in MW)

▪ Minimum → relevant for wind at night ▪ Export variation

36

Plotting of 2017 loads

50% 25% Times per year

SLIDE 37

Generation Expansion Assumptions – Current Installed Capacity

38

Source / Fuel Type Capacity [MW] Installed Available/Firm Thermal RLNG 7,806 7,218 Natural Gas 4,334 3,907 HFO 5,417 4,868 Imported Coal 2,640 2,428 Domestic Coal 150 30 Sub-Total 20,347 18,451 Nuclear Sub-Total 1,330 1,232 Hydro Large Scale Plants 9,638 6,080 Small Hydro 128 65 Sub-Total 9,766 6,145 Renewables Wind 1,185 1,185 Solar 386 386 Bagasse 306 295 Sub-Total 1,877 1,866 Total 33,320 27,694

SLIDE 38

Generation Expansion Assumptions

Plant retirements (existing, firm capacity + RES)

▪ Hydro: always rehabilitation ▪ Committed plants: all beyond forecast period

39

~9 GW MW

SLIDE 39

Sufficient existing and committed capacity until 2026 but NO reserve for

▪

Outages of (large) units, low / delayed hydrology, fuel supply (gas) security, delay of CODs (hydels)

▪

Higher demand growth, Load Management Regional / suppressed demand VRE not considered

Demand Supply Balancing: Monthly Peak versus Capacity (Task 2)

40

SLIDE 40

Insufficient capacity 2024 onwards if delay of

▪

Hydels 2 – 3 years

▪

Coal and nuclear 1-2 years

▪

Bagasse 1-2 years

Demand Supply Balancing: Monthly Peak versus Capacity (Task 2)

41

SLIDE 41

Demand Supply Balancing - Summary

Sufficient capacity IF

▪ Capacity side ➢No longer lasting outages of (large) units, no low hydrology and gas supply secured ➢Additional capacity added before 2026 ➢No considerable delay of hydels & nuclear (for fossil thermal probability for delay rather small) ➢VRE support peak (they do!!) → Task 3 ▪ Demand wise ➢No higher demand growth ➢Load Management Regional and other suppressed demand continues

Sufficient energy BUT

➢Probably not least-cost ➢Impact of share of take-or-pay not known in detail → Task 3 will (in part) answer this

42

SLIDE 42

Thank you!

Mr. Karsten Schmitt

Project Manager Energy Economics and Planning Department Tel.: +49 6101 / 55-1579 Fax: +49 6101 / 55-1808 Karsten.Schmitt@tractebel.engie.com

43