1

California ELCC Analysis in SERVM

Astrape Consulting

6/2/2017

2

Shifting Load Shape as More Renewable Are Added

0

10,000

20,000

30,000

40,000

50,000

60,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

MW

Hour

Gross Load 33% Net Load 43% Net Load

Increasing penetration initially flattens load shape as it shifts

peak timing

Further increases steepen load shape, resulting in more value for energy

limited resources

3

SERVM Results

Preliminary Draft Results

4

Weighted Average Location of Solar Locations Used for Each Region

5

Output as a Function of Longitude for Average August Hour 16

y = -0.0392x - 4.0148R² = 0.9009

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

-120.0 -118.0 -116.0 -114.0 -112.0 -110.0 -108.0 -106.0 -104.0 -102.0

Sola

r O

utpu

t

Longitude

Malibu, CA

Portales, NM

West East

Congress, AZ

6

Output as a Function of Latitude for Average August Hour 16

y = -0.0084x + 0.9296R² = 0.507

0.5

0.52

0.54

0.56

0.58

0.6

0.62

0.64

0.66

0.68

32.0 34.0 36.0 38.0 40.0 42.0 44.0

Sola

r O

utpu

t

Latitude

Long Beach, CA

Platora, NV

South North

Olancha, CA

7

33% Northern and Southern CA Average August Day Comparison

-20%

-15%

-10%

-5%

0%

5%

10%

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1 6 11 16 21

Del

ta (%

)

Sola

r O

utpu

t

Hour of Day

Northern CA Southern CA Delta (N-S)

8

33% Southern CA and Southwest Average August Day Comparison

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%

30%

35%

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1 6 11 16 21

Del

ta (%

)

Sola

r O

utpu

t

Hour of Day

Southern CA Southwestern U.S. Delta (S-SW)

9

ELCC Approximation Method Illustration

0

10000

20000

30000

40000

50000

60000

Jan-83 Feb-83 Apr-83 May-83 Jul-83 Sep-83 Oct-83 Dec-83

MW

Date

Gross Load Net Load

Difference in Peaks = 4,960 MWPeak Gross Load = 52,421 MW

Peak Net Load = 47,461 MW

10

Approximation Method Comparison to SERVM Results – 33% RPS Case

ELCC Case and TechnologyNLP Method

(Average Annual) (%)

SERVM Results (%)

Average/-RPS 29.08 28.94Average/-Wind 18.27 21.03Average/-Solar 28.60 32.75

Marginal CA-N/-Wind 26.25 21.49Marginal CA-N/-Fixed PV 4.08 13.36

Marginal CA-N/-Tracking PV 8.74 21.12Marginal CA-N/-BTMPV 3.78 11.56

Marginal CA-S/-Wind 22.25 14.43Marginal CA-S/-Fixed PV 1.80 9.58

Marginal CA-S/-Tracking PV 4.00 15.24Marginal CA-S/-BTMPV 1.29 7.73

Marginal NW/Wind 42.89 40.26Marginal SW/Wind 17.57 23.75

Marginal SW/Fixed PV 0.64 8.12Marginal SW/Tracking PV 1.93 12.35

11

Approximation Method Comparison to SERVM Results – 43% RPS Case

ELCC Case and TechnologyNLP Method

(Average Annual) (%)

SERVM Results (%)

Average/-RPS_43 17.97 20.17Average/-Wind_43 17.96 22.50Average/-Solar_43 14.79 19.56

Marginal CA-N/-Wind_43 18.44 27.09Marginal CA-N/-FixedPV_43 0.15 4.16

Marginal CA-N/-TrackingPV_43 0.85 8.28Marginal CA-N/-BTMPV_43 0.33 4.74

Marginal CA-S/-Wind_43 20.69 22.06Marginal CA-S/-FixedPV_43 0.02 3.61

Marginal CA-S/-TrackingPV_43 0.16 3.91Marginal CA-S/-BTMPV_43 0.02 2.00

Marginal NW/-Wind_43 43.27 43.06Marginal SW/-Wind_43 21.41 29.93

Marginal SW/-FixedPV_43 0.00 0.69Marginal SW/-TrackingPV_43 0.00 2.99

12

RPS Example Dispatch – 33 and 43%

40,000

41,000

42,000

43,000

44,000

45,000

46,000

47,000

48,000

49,000

50,000

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Net

Loa

d (M

W)

Hour

Conventional Battery/PSH Demand Response

38,000

39,000

40,000

41,000

42,000

43,000

44,000

45,000

46,000

47,000

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Net

Loa

d (M

W)

Hour

Conventional Battery/PSH Demand Response

13

Instantaneous Load Compared to Hourly Integrated Load

-400

-300

-200

-100

0

100

200

300

5 10 15 20 25 30 35 40 45 50 55 60

MW

Time (Minutes)

Without Solar With Solar

1

California ELCC Analysis in SERVM

Astrape Consulting

6/2/2017

2

Shifting Load Shape as More Renewable Are Added

0

10,000

20,000

30,000

40,000

50,000

60,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

MW

Hour

Gross Load 33% Net Load 43% Net Load

Increasing penetration initially flattens load shape as it shifts

peak timing

Further increases steepen load shape, resulting in more value for energy

limited resources

3

SERVM Results

Preliminary Draft Results

4

Weighted Average Location of Solar Locations Used for Each Region

5

Output as a Function of Longitude for Average August Hour 16

y = -0.0392x - 4.0148R² = 0.9009

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

-120.0 -118.0 -116.0 -114.0 -112.0 -110.0 -108.0 -106.0 -104.0 -102.0

Sola

r O

utpu

t

Longitude

Malibu, CA

Portales, NM

West East

Congress, AZ

6

Output as a Function of Latitude for Average August Hour 16

y = -0.0084x + 0.9296R² = 0.507

0.5

0.52

0.54

0.56

0.58

0.6

0.62

0.64

0.66

0.68

32.0 34.0 36.0 38.0 40.0 42.0 44.0

Sola

r O

utpu

t

Latitude

Long Beach, CA

Platora, NV

South North

Olancha, CA

7

33% Northern and Southern CA Average August Day Comparison

-20%

-15%

-10%

-5%

0%

5%

10%

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1 6 11 16 21

Del

ta (%

)

Sola

r O

utpu

t

Hour of Day

Northern CA Southern CA Delta (N-S)

8

33% Southern CA and Southwest Average August Day Comparison

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%

30%

35%

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1 6 11 16 21

Del

ta (%

)

Sola

r O

utpu

t

Hour of Day

Southern CA Southwestern U.S. Delta (S-SW)

9

ELCC Approximation Method Illustration

0

10000

20000

30000

40000

50000

60000

Jan-83 Feb-83 Apr-83 May-83 Jul-83 Sep-83 Oct-83 Dec-83

MW

Date

Gross Load Net Load

Difference in Peaks = 4,960 MWPeak Gross Load = 52,421 MW

Peak Net Load = 47,461 MW

10

Approximation Method Comparison to SERVM Results – 33% RPS Case

ELCC Case and TechnologyNLP Method

(Minimum Annual) (%)

NLP Method (Maximum Annual)

(%)

NLP Method (Average Annual)

(%)

SERVM Results

(%)

Average/-RPS 17.12 49.52 29.08 28.94Average/-Wind 1.80 57.21 18.27 21.03Average/-Solar 18.18 36.28 28.60 32.75

Marginal CA-N/-Wind 0.38 72.38 26.25 21.49Marginal CA-N/-Fixed PV 0.00 48.75 4.08 13.36

Marginal CA-N/-Tracking PV 0.00 71.03 8.74 21.12Marginal CA-N/-BTMPV 0.00 39.48 3.78 11.56

Marginal CA-S/-Wind 1.11 80.49 22.25 14.43Marginal CA-S/-Fixed PV 0.00 19.38 1.80 9.58

Marginal CA-S/-Tracking PV 0.00 22.31 4.00 15.24Marginal CA-S/-BTMPV 0.00 11.97 1.29 7.73

Marginal NW/Wind 0.00 91.00 42.89 40.26Marginal SW/Wind 0.00 58.00 17.57 23.75

Marginal SW/Fixed PV 0.00 16.25 0.64 8.12Marginal SW/Tracking PV 0.00 19.50 1.93 12.35

11

Approximation Method Comparison to SERVM Results – 43% RPS Case

ELCC Case and TechnologyNLP Method

(Minimum Annual) (%)

NLP Method (Maximum Annual)

(%)

NLP Method (Average Annual)

(%)

SERVM Results

(%)

Average/-RPS_43 10.66 30.50 17.97 20.17Average/-Wind_43 1.74 56.85 17.96 22.50Average/-Solar_43 9.18 18.30 14.79 19.56

Marginal CA-N/-Wind_43 0.21 45.22 18.44 27.09Marginal CA-N/-FixedPV_43 0.00 1.50 0.15 4.16

Marginal CA-N/-TrackingPV_43 0.00 5.15 0.85 8.28Marginal CA-N/-BTMPV_43 0.00 2.73 0.33 4.74

Marginal CA-S/-Wind_43 0.90 65.98 20.69 22.06Marginal CA-S/-FixedPV_43 0.00 0.76 0.02 3.61

Marginal CA-S/-TrackingPV_43 0.00 2.92 0.16 3.91Marginal CA-S/-BTMPV_43 0.00 0.70 0.02 2.00

Marginal NW/-Wind_43 0.00 92.00 43.27 43.06Marginal SW/-Wind_43 0.00 58.00 21.41 29.93

Marginal SW/-FixedPV_43 0.00 0.00 0.00 0.69Marginal SW/-TrackingPV_43 0.00 0.00 0.00 2.99

12

RPS Example Dispatch – 33 and 43%

40,000

41,000

42,000

43,000

44,000

45,000

46,000

47,000

48,000

49,000

50,000

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Net

Loa

d (M

W)

Hour

Conventional Battery/PSH Demand Response

38,000

39,000

40,000

41,000

42,000

43,000

44,000

45,000

46,000

47,000

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Net

Loa

d (M

W)

Hour

Conventional Battery/PSH Demand Response

13

Instantaneous Load Compared to Hourly Integrated Load

-400

-300

-200

-100

0

100

200

300

5 10 15 20 25 30 35 40 45 50 55 60

MW

Time (Minutes)

Without Solar With Solar

Incorporating Adequacyinto

Resource Planning for the Pacific Northwest

North American Electric Reliability CorporationProbabilistic Assessment Working Group Meeting

Tampa, FLJune 1, 2017

Biomass2%

Coal11%

Hydro55%

Natural gas15%

Nuclear2%

Wind15%

63,100 MW Installed Generating Capacity (Pacific NW 2017)

Biomass3%

Coal12%

Market3%

Hydro67%

Natural gas6%

Nuclear4%

Wind5%

Expected Dispatch of Pacific NW Generation(Average Hydro Conditions)

When available, surplus hydro generation is used to displace more expensive resources

PNW also relies on market purchases

PNW Power Planning Process

4

Regional Portfolio Model

New and ExistingGeneratingResources

AURORAElectricity Prices

GENESYSHydro +

Adequacy

Demand ForecastShort -TermLong-Term

Energy EfficiencyPotential

Assessment

Financial andOther Assumptions

What is GENESYS• Monte-Carlo hourly chronological simulation

model for PNW generating resources

• Analyze a single operating year (Oct-Sep)• Stick and bubble representation for transmission

• Random Variables• River flows• Temperatures (load variation)• Wind generation (correlated to temperature)• Forced outages

Transmission in GENESYS

• NW region includes: East (E) West (W)

• Solid lines indicate transmission into and out of the region

• Not a power-flow analysis

• East/west transmission capability varies based on BPA data

• Southern interties have fixed transfer capability

• SW modeled as import market only

C

EW

CJ

SWNC

SC

6

GENESYS Flow DiagramLoad

GENESYS

Wind/Solar

Thermal Data

Curtailments

Contracts

•Hydro Data•Sustained

Peaking

•ThermalDispatch

•Purchases•HydroGeneration•Cost

•Elevations•Outflows•Spill

Standby •LOLP•ASCC•ARM

Yellow highlighted data fed to RPM

January 1930

-5000

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

1 27 53 79 105

131

157

183

209

235

261

287

313

339

365

391

417

443

469

495

521

547

573

599

625

651

677

703

729

Hour in Month

Meg

awat

ts

Net Demand NW Thermal NW Hydro Unserved Net Imports

Simulated Dispatch January

8

Assessing Adequacy• Run every combination of temperature and

streamflow (80 times 77 = 6,160)

• Count only existing resources, expected energy efficiency savings and only the planned resources that are sited and licensed

• Energy Efficiency savings are in the load forecast

• Count the number of simulations (games) that have at least one curtailment

Loss of Load Probability

6160 Simulations

Out of 6160 simulations, 308 had curtailment events (red bins)

Loss of Load Probability (LOLP) = 308/6160 = 5 percent

10

2022 Annual LOLP (%)

Market 3400 3000 2500 2000

High Load 18.1 19.0 21.0 23.6

Med Load 5.8 6.2 7.2 8.6

Low Load 1.8 2.1 2.7 3.4

RED Squares = Inadequate (LOLP > 5%)GREEN Squares = Adequate (LOLP < 5%)Circled Square = Reference Case, LOLP = 7.2% = Inadequate for 2022

2022 Annual EUE (MW-hours)

Market 3400 3000 2500 2000

High Load 6113 6801 8159 10434

Med Load 1085 1448 2038 2942

Low Load 358 533 844 1305

2022 Annual LOLH (Hours)

Market 3400 3000 2500 2000

High Load 7.0 7.3 8.1 9.6

Med Load 1.2 1.3 1.6 2.1

Low Load 0.3 0.4 0.6 0.8

2022 Annual CVaR Peak (MW)

Market 3400 3000 2500 2000

High Load 2362 2551 2921 3436

Med Load 1053 1247 1625 2120

Low Load 397 519 730 1054

2022 Annual CVaR Energy (GW-hours)

Market 3400 3000 2500 2000

High Load 104 113 131 158

Med Load 22 29 40 58

Low Load 7 11 17 26

GENESYS Inputs to the RPM

16

Regional Portfolio Model

GENESYS

Monthly Hydro Generation

Sustained Peaking Capability

Adequacy Reserve Margin

Associated System Capacity Contribution

Incorporating Adequacy1. Pass adequacy parameters to RPM Use Associated System Capacity

Contribution for relevant resources Build to the Adequacy Reserve Margin

2. Verification of adequacy Choose one year from one game Get loads and resources Run GENESYS to verify LOLP close to 5%

17

Next Steps

• Redeveloping GENESYS for more precision (completion date Oct 2018)

• Reevaluating current adequacy standard(Annual LOLP maximum of 5%)• Move to seasonal adequacy metric(s)• Also account for magnitude and duration• Completion by mid-2018

Variable Energy Resource Capacity Contributions Consistent With Reserve Margin and Reliability

byEugene Preston

andNoha Abdel-Karim

May 31, 2017

Modeling Variable Energy Resources

Choice – VERs as generators or negative loads?• As generators we lose time connectivity information.• As hourly load reducers we retain time connectivity.

Choice – Model a typical year or many years?• Combining several years loses year to year swings.• Combining several years loses some VER timings.• Combining VERs from several years may result in

loss of gaps in the original VER hourly data.

Modeling Variable Energy Resources Choice – Direct LOLP versus Monte Carlo?

• As long as all generators are independent, MC and Direct LOLP provide the same reliability indices provided VERs are treated as hourly load reducers.

• Sequential MC is required to model VERs, if VERs are treated as generators; however this approach results in loss of VER connectivity even when MC is used.

• Direct LOLP is at least a million times faster than MC.• Direct LOLP only requires the conventional generators

be put in a COPT, capacity outage probability table.• All conventional generators are put into the COPT.

COPT – Capacity Outage Probability Table

Net Demand = hourly demand minus sum of VER MW that hour

A Few Important Definitions

• LOLP is loss of load probability (every hour).• LOLE is loss of load expectation =

o the sum of daily max LOLPs from the COPT, oro the MC counting of days per year occurrences.

• LOLH is loss of load hours = o the sum of hourly LOLPs from the COPT, oro the MC counting of hours per year occurrences.

• LOLEV differs from LOLE in that more than one loss of load event per day can be counted.

Modeling ERCOT

• Hourly VER data for 2010 – 2015.• 2016 conventional generation = ~75000 MW• 2016 wind + solar (nameplate) = ~17000 MW• 2016 pk load = ~68000 MW for LOLE=0.1 d/y• 2026 conventional generation = ~80000 MW• 2026 wind + solar (nameplate) = ~29000 MW• 2026 pk load = ~76000 MW for LOLE=0.1 d/y

ERCOT VER RM at Different Capacity Contributions

Solar Shifts the Net Demand Peak to 8 PM

Historical Years Ratio LOLH/LOLE

Historical LOLE Deviation Increases with VER

Recommendations

• Provide reliability evaluations of VER impacts.• It’s very important to maintain chronology

between variable energy generation and load.• It will be necessary to develop and maintain

public databases of wind, solar, and hydro historical production.

• VERs should be given capacity credits from the running of loss of load probability studies.

Probabilistic Results

Matthew ElkinsSenior Economist

WECC Footprint

• 329 Members

• 38 Balancing Authorities

• 1.8 million square miles

• 82.2 million people

Reliability Threshold Margins

Existing Resources

Annual weighted reserve margin needed to maintain 99.98% reliability.

Future Resources

Existing ResourcesSub-Region

Year All #1 #2 #3 #4 #5

2017 15.7% 20.4% 12.6% 15.3% 17.3% 11.4%

2018 15.5% 20.1% 12.7% 15.1% 17.0% 11.2%

2019 15.4% 19.9% 12.7% 14.9% 16.7% 10.9%

2020 15.2% 19.8% 12.7% 14.8% 16.4% 10.7%

2021 15.1% 19.5% 12.7% 14.6% 16.0% 10.5%

2022 14.9% 19.4% 12.7% 14.4% 15.9% 10.4%

2023 14.8% 19.3% 12.7% 14.2% 15.7% 10.3%

2024 14.8% 19.1% 12.7% 14.0% 15.5% 10.2%

2025 14.6% 19.0% 12.7% 13.9% 15.3% 10.1%

2026 14.6% 18.9% 12.7% 13.7% 15.1% 10.0%Future Resources

Sub-RegionYear All #1 #2 #3 #4 #5

2017 15.8% 20.4% 12.7% 15.4% 17.3% 12.0%

2018 16.0% 20.1% 13.1% 15.9% 16.8% 13.1%

2019 16.3% 20.6% 13.4% 15.9% 16.5% 13.5%

2020 16.3% 20.5% 13.7% 15.9% 16.2% 13.3%

2021 16.2% 20.2% 13.8% 15.9% 16.1% 13.2%

2022 16.1% 20.0% 13.8% 15.9% 15.9% 13.0%

2023 16.0% 19.9% 13.8% 15.9% 15.7% 12.9%

2024 16.1% 20.1% 13.8% 15.8% 15.5% 13.0%

2025 16.1% 20.1% 13.8% 15.7% 15.3% 13.0%

2026 16.0% 20.0% 13.8% 15.6% 15.1% 12.9%

Demand-At-Risk (Hours)

Existing Resources

Number of hours where 99.98% reliability is not maintained.

Future Resources

Existing ResourcesSub-Region

Year All #1 #2 #3 #4 #52017 1 - 1 - - -2018 1 - 1 - - -2019 1 - 1 - - -2020 - - - - - -2021 - - - - - -2022 3 - 3 - - -2023 20 1 18 - 2 -2024 53 6 49 - 4 -2025 103 8 93 - 10 22026 200 9 198 6 7 3 Future Resources

Sub-RegionYear All #1 #2 #3 #4 #52017 - - - - - -

2018 - - - - - -

2019 - - - - - -

2020 - - - - - -

2021 - - - - - -

2022 - - - - - -

2023 - - - - - -

2024 - - - - - -

2025 - - - - - -

2026 19 - 19 - - -

NW-US Footprint

• 24 Balancing Authorities

• 7 Transmission Areas

Reliability Threshold Margins

Existing Resources

• Annual weighted reserve margin needed to maintain 99.98% reliability.

Future Resources

Normalized Annual Threshold Margin – Existing ResourcesArea

Year All #1 #2 #3 #4 #5 #6 #72017 20.4% 23.3% 14.6% 11.2% 17.9% 30.2% 12.9% 19.2%2018 20.1% 23.0% 14.5% 10.8% 17.9% 29.8% 12.7% 19.0%2019 19.9% 22.8% 14.4% 10.7% 17.9% 29.4% 12.3% 18.9%2020 19.8% 22.7% 14.3% 10.7% 17.8% 29.3% 12.2% 18.8%2021 19.5% 22.3% 14.2% 10.5% 17.7% 28.8% 12.1% 18.6%2022 19.4% 22.2% 14.2% 10.6% 17.7% 28.0% 11.4% 18.5%2023 19.3% 22.0% 14.1% 10.5% 17.6% 27.7% 11.4% 18.3%2024 19.1% 21.9% 14.1% 10.4% 17.5% 27.4% 11.3% 18.3%2025 19.0% 21.7% 14.0% 10.4% 17.4% 26.9% 10.6% 18.1%2026 18.9% 21.6% 14.2% 10.2% 17.3% 26.3% 10.1% 18.0%

Normalized Annual Threshold Margin – Future ResourcesArea

Year All #1 #2 #3 #4 #5 #6 #72017 20.4% 23.4% 14.7% 11.2% 17.9% 30.3% 13.1% 19.2%2018 20.1% 23.0% 14.5% 10.8% 17.9% 29.9% 12.9% 19.0%2019 20.6% 24.0% 14.5% 10.8% 17.9% 29.6% 12.1% 18.9%2020 20.5% 23.8% 14.5% 10.7% 17.8% 29.7% 12.0% 18.8%2021 20.2% 23.4% 14.4% 10.5% 17.8% 29.2% 11.9% 18.6%2022 20.0% 23.2% 14.4% 10.6% 17.8% 28.4% 11.2% 18.5%2023 19.9% 23.1% 14.6% 10.5% 17.7% 28.1% 11.2% 18.3%2024 20.1% 23.0% 16.3% 10.5% 17.6% 27.8% 11.1% 18.3%2025 20.1% 22.8% 17.0% 10.4% 17.5% 27.2% 10.4% 18.1%2026 20.0% 22.7% 17.4% 10.3% 17.4% 26.6% 9.9% 18.0%

Average Imports

Existing Resources

• Imports needed to meet 99.98% reliability threshold margin.

Future Resources - 200 400 600 800

1,000 1,200 1,400 1,600 1,800 2,000

2017 2018 2019 2020 2021 2022 2023 2024 2025 2026

Annual Average Hourly Imports (MW) - Existing Resources

-

200

400

600

800

1,000

1,200

1,400

1,600

2017 2018 2019 2020 2021 2022 2023 2024 2025 2026

Annual Average Hourly Imports (MW) - Future Resources

Demand-At-Risk (Hours)

Existing Resources

• Number of hours where 99.98% reliability is not maintained.

Future Resources

Demand-At-Risk (Hours) – Existing ResourcesArea

Year All #1 #2 #3 #4 #5 #6 #72017 - - - - - - - -2018 - - - - - - - -2019 - - - - - - - -2020 - - - - - - - -2021 - - - - - - - -2022 - - - - - - - -2023 1 1 - - - - - -2024 6 2 4 1 - - 2 -2025 8 3 7 3 - 1 6 -2026 9 4 6 4 - 4 7 -

Demand-At-Risk (Hours) – Future ResourcesArea

Year All #1 #2 #3 #4 #5 #6 #72017 - - - - - - - -2018 - - - - - - - -2019 - - - - - - - -2020 - - - - - - - -2021 - - - - - - - -2022 - - - - - - - -2023 - - - - - - - -2024 - - - - - - - -2025 - - - - - - - -2026 - - - - - - - -

Questions

Matthew ElkinsSenior Economistmelkins@wecc.biz

801-819-7657

mailto:melkins@wecc.biz

California_ELCC_Analysis_in_SERVMCalifornia ELCC Analysis in SERVMShifting Load Shape as More Renewable Are AddedSERVM ResultsWeighted Average Location of Solar Locations Used for Each RegionOutput as a Function of Longitude for Average August Hour 16Output as a Function of Latitude for Average August Hour 1633% Northern and Southern CA Average August Day Comparison33% Southern CA and Southwest Average August Day ComparisonELCC Approximation Method IllustrationApproximation Method Comparison to SERVM Results – 33% RPS CaseApproximation Method Comparison to SERVM Results – 43% RPS CaseRPS Example Dispatch – 33 and 43%Instantaneous Load Compared to Hourly Integrated Load

ELCC_Astrape_6_2_17California ELCC Analysis in SERVMShifting Load Shape as More Renewable Are AddedSERVM ResultsWeighted Average Location of Solar Locations Used for Each RegionOutput as a Function of Longitude for Average August Hour 16Output as a Function of Latitude for Average August Hour 1633% Northern and Southern CA Average August Day Comparison33% Southern CA and Southwest Average August Day ComparisonELCC Approximation Method IllustrationApproximation Method Comparison to SERVM Results – 33% RPS CaseApproximation Method Comparison to SERVM Results – 43% RPS CaseRPS Example Dispatch – 33 and 43%Instantaneous Load Compared to Hourly Integrated Load

Incorporating_Adequacy_into_Resource_Planning_for_the_Pacific_NorthwestIncorporating Adequacy�into�Resource Planning �for the Pacific Northwest�Slide Number 2Slide Number 3PNW Power Planning ProcessWhat is GENESYSSlide Number 6GENESYS Flow DiagramSlide Number 8Assessing AdequacyLoss of Load Probability2022 Annual LOLP (%)2022 Annual EUE (MW-hours)2022 Annual LOLH (Hours)2022 Annual CVaR Peak (MW)2022 Annual CVaR Energy (GW-hours)GENESYS Inputs to the RPMIncorporating AdequacyNext Steps

Variable_Energy_Resource_Capacity_Contributions_Consistent_With_Reserve_Margin_and_ReliabilityVariable Energy Resource Capacity Contributions Consistent With Reserve Margin and Reliability�Modeling Variable Energy ResourcesModeling Variable Energy ResourcesCOPT – Capacity Outage Probability TableA Few Important DefinitionsModeling ERCOTERCOT VER RM at Different Capacity ContributionsSolar Shifts the Net Demand Peak to 8 PMHistorical Years Ratio LOLH/LOLE Historical LOLE Deviation Increases with VERRecommendations

WECC_Probabilistic_Results_Sample_-_PAWG_-_20170601Probabilistic ResultsSlide Number 2Reliability Threshold MarginsDemand-At-Risk (Hours)Slide Number 5Reliability Threshold MarginsAverage ImportsDemand-At-Risk (Hours)Questions