flexibility assessment metrics - nerc

Aidan TuohyGrid Operations and Planning, EPRI

NERC ERSTF Meeting12/10/2014

Flexibility Assessment Metrics

2© 2014 Electric Power Research Institute, Inc. All rights reserved.

EPRI Project: Strategic and Flexible Planning

• Framework in how to think about flexibility from planning perspective– Multi-Level Analysis– Reliability and economics– Metrics to assess flexibility

requirements and flexibility adequacy

– Tool to perform analysis– Case studies to prove concepts

• Considering multiple aspects– Resource adequacy– Transmission– Energy limited (DR, Storage, Hydro)– System operations

-1000 -800 -600 -400 -200 0 200 400 600 800 100010-2

10-1

100

101

MW ramp

Fre

qu

en

cy (

% o

f tim

e)

no wind1 GW wind2 GW wind

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

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

0

100

200

300

400

500

600

700

800

900


Operational Flexibility in Planning and Operations• Planning for flexibility

– Need to ensure in planning time frame that there is sufficient operational flexibility provided

– Similar to capacity/resource adequacy (though more complex)– Consider different resources and the network, as well as how

the system operates– Trade offs between reliability and economics

• Operations with flexibility– Once you have the resources, how do you operate them?– How do you incentivize resources?– New reserve products, new scheduling techniques, etc

Need to consider operational issue in planning to ensure system reliability


Flexibility Considerations & Metrics

• Many Regions (Regulators + ISO+ Utilities) Considering Future Flexibility Needs Now – Planning and Operations time frame

• Other systems experiencing similar needs Germany, Spain, New York, Hawaii etc.

• New flexible resources becoming deployable in the bulk system

California•Flexible Resource Adequacy•Flexi-Ramp Market Product•Long Term Procurement Plan

Ireland• Long Term Flexibility Incentives

Oregon• Integrated Resource Planning Process

MISO• Market Rule Changes to

Incentivize Flexibility

All areas are different, but shared learnings help all improve methods


Flexibility Metrics for system planning

• Multi-Level Approach– Levels 1 and 2 screening

metrics– Levels 3 and 4 detailed metrics

• Four detailed metrics for planning time frame:– Periods of Flexibility Deficit – Expected Unserved Ramping – Well-being analysis – Insufficient Ramping Resource

Expectation

• Post-processed metrics based on simulation or historical data

Level 1• Variability Analysis &

Flexibility Requirement

Level2• Resource Flexibility

Calculation

Level 3 • System Flexibility

Metrics

Level 4• Transmission and Fuel

Constrained Flexibility

Results from different case studies presented here


Level 1: Variability Analysis – what flexibility is required?

Determine ramping with different time horizons, for demand, wind and net demand

Percentiles used to determine likelihood (99.9 here)


Time of Day and Year Example Ramp Rates

1 2 3 4 5 6 7 8 9 101112131415161718192021222324

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

0

200

400

600

800

1000

1200

1400

1600

1800• Max 1 hour up ramps for each hourly interval in a given month

• Based on example data from Northwest US

• Calculate for different time intervals, up and down ramping

Flexibility resources will need to be available at different times of day and year depending on system requirements


Level 2: What flexibility is available?

Determine Ramping Available in Each Hour of the YearDifferent time scales, need to make assumptions

about intertie and energy limited resources


Need to consider operational aspects…

How you operate the system – including reserves – impacts on the availability of flexible capacity


Level 3: What is the net flexibility after ramps?

Examine either net available flexibility or against extreme ramps


Level 3: Metric 1: Periods of Flexibility Deficit

30%

40%

50%

60%

70%

80%

90%

100%

0 2 4 6 8 10 12 14 16

Perio

ds o

f Fle

xibi

lity

Def

icit

(%)

Time Horizon (Hours)

Example for extreme ramping requirements (97th percentile)


Level 3: Metric 2. Expected Ramping Deficit

Expected value of ramping deficit values observed in each direction and time horizon

∑=

=−+>−+ =

8760

1/,0||/, 8760

1 i

iDCit i

DeficitERD


Example Conclusions from Studies

• “Flexibility shortages seen over 60 minute interval”– Higher resolution data would be beneficial

• “Peak at 540 minutes indicates that the system may need to add additional capacity”– Unlikely that system leaves long start units offline when available– In that case the system had interties assumed inflexible

• Assumptions on intertie flexibility may alleviate issue

• Frequency of shortages for <1 hour shows that it may be operational issue driving results– Operational changes may be sufficient to meet ramping

requirements reserve etc.– Longer term mitigation measure: intertie operation or new capacity


Ongoing work

• Impact of transmission constraints on system flexibility adequacy– Working with Ecco International to demonstrate how metrics can

consider flexibility and how transmission can be a flexibility resource

• Flexibility from Energy Limited Resources– How do demand response, storage, hydro etc provide flexibility?

• Continued Tool Development– Adding new functions (some based on existing algorithms, some new)– Improving interface/ease of use

• Case Studies to demonstrate framework– Detailed case study w/BPA – Screening case studies for TVA & Southern


Conclusions

• Many Applications of such an approach– Reliability assessments– Wind and solar integration studies– Flexibility assessment for resource planning – May be applied in the operational planning time frame

• E.g. maintenance scheduling, hydro scheduling

• Flexibility metrics have been or will be made public through various journal and conference publications– White paper available on epri.com – “Metrics for Quantifying Flexibility in Power Systems”

• InFLEXion flexibility assessment tool available to project funders

• EPRI is engaging with a wide variety of parties in this area, e.g. PG&E collaborative effort, IEEE, LBNL, NWPCC


Together…Shaping the Future of Electricity

1

WWSIS - 3: Western Frequency Response and Transient Stability Study

GE EnergyNicholas W. Miller (PM)Miaolei ShaoSlobodan PajicRob D’Aquila

NRELKara Clark (PM)

NERC ERSTF BriefingAtlantaDecember 10-11, 2014

The draft report is under review by the TRC and by DOE.

Therefore, all of the results and statements in this presentation MUST be regarded as preliminary and subject to further review and modification.

Western Wind and Solar Integration Study, Phase III : Transient Stability and Frequency Response. Subject to Final Review and Approval by DOE

Key Points

What: Stability – for the 1st minute after a big disturbance –is critically important limitation in the West

Why: Widespread worry that lots of wind and solar, especially combined with lots of coal diretirements will irreparably disrupt grid stability.

In the context of ERSTF: will essential reliability services be affected (i.e. depleted, altered, enhanced...)

What we learned: The Western Interconnection can be made to work well with both high wind and solar and substantial coal displacement, using good, established planning & engineering practice and commercially available technologies.

2


Team….

Who:

– Project Co-funded by DOE Wind and Solar Programs

– Project Management by NREL: Kara Clark

– Subcontract to GE Energy Consulting

Technical Review Committee:

• North American Electric

Reliability Corporation

• PacifiCorp

• Public Service of New

Mexico

• Western Area Power

Administration

• Tucson Electric Power,

• Western Electricity

Coordinating Council,

• California ISO

• Xcel Energy

• Sacramento Municipal

Utility District

• Arizona Public Service,

• Bonneville Power

Administration

• Western Governors

Association

• Electric Reliability

Council of Texas

• Utility Variable-

generation Interest

Group

• DOE

• Electric Power Research

Institute

• Sandia NL

• Lawrence Berkeley NL

• Iowa State University

• University College

Dublin

• Arizona State University


Critical Disturbances in the West

Pacific DC

Intertie

Selected by Technical Review Committee:

• Palo Verde Nuclear Plant(2 of 3 units for ~2,750 MW)

• Pacific DC Intertie (Maximum north-to-southpower flow ~3,100 MW)


WECC-Wide Summary(1)

Light Spring Base

(2)

Light Spring High Mix

Light Spring Extreme Sensitivity

Wind (GW) 20.9 27.2 32.6

Utility-Scale PV (GW) 3.9 10.2 13.5

CSP (GW) 0.9 8.4 8.3

Distributed PV (GW) 0 7.0 10.4

Total (GW) = 25.7 52.8 64.8

Penetration(3)

(%) = 21% 44% 53%

Light Spring Load Study ScenariosBase Case High Mix Case

(1) Western Electricity Coordinating Council includes parts of Canada and Mexico, (2) Provided by WECC, (3) Penetration is % of total generation for this snapshot.


Heavy Summer Load Study ScenariosBase Case High Mix Case

WECC-Wide Summary(1)

Heavy Summer Base

(2)

Heavy Summer High Mix

Wind (GW) 5.6 14.3

Utility-Scale PV (GW) 1.2 11.2

CSP (GW) 0.4 6.6

Distributed PV (GW) 0.0 9.4

Total = 7.2 41.5

Penetration(3)

(%) = 4% 20%

(1) Western Electricity Coordinating Council includes parts of Canada and Mexico, (2) Provided by WECC, (3) Penetration is % of total generation for this snapshot.


Frequency Response Analysis

7Preliminary Results: Not for Further

Distribution or Citation


Frequency Response with High Renewables

Interconnection frequency response > 840 MW/0.1Hz threshold in all cases.No under-frequency load shedding (UFLS).

Disturbance: Trip 2 Palo Verde units (~2,750MW)

3

2

Light Spring Base


Light Spring Extreme

23

1

1


Wind Plant Frequency Responsive Controls

• Inertial control responds– to frequency drops only

– in 5-10 second time frame

– uses inertial energy from rotating wind turbine to supply power to system

– requires energy recovery from system to return wind turbines to nominal speed

– more responsive at higher wind speeds

– ERSTF: this is Fast Frequency Response, NOT System Inertial Response

• Governor control responds– to both frequency drops and increases

– in 5-60 second time frame

– requires curtailment to be able to increase power

– ERSTF: this is either Fast Frequency Response, or Primary Frequency Response (depending on aggresiveness of the control)


Frequency Control on Wind Plants

Light Spring High MixLight Spring High Mix with governor control*Light Spring High Mix with inertial control*Light Spring High Mix with both controls


40% of wind plants (i.e., new ones) had these controls, for a total of 300 MW initial curtailment out of 27GW production.

1

2

3

4

1

2

34


Frequency Control on Utility-scale PV Plants


Light Spring High Mix with governor controls on

utility-scale PV plants

~80% of utility-scale PV plants (i.e., new ones) had these controls, for a total of 820 MW initial curtailment out of 10.2 GW production.


1

2

2

1

ERSTF: 820 MW of Fast Frequency Response

FRO Base Hi-Mix Wind

Governor

Control

Wind

Inertial

Control

Wind

Governor

and

Inertial

Controls

Utility-

scale PV

Governor

Control

Energy

Storage

with

Governo

r Control

Extreme

Hi-Mix

WECC 840 1352 1311 1610 1323 1571 2065 1513 1055


Fault Ride Through Needed with High Levels of DG

Pessimistic

Pessimistic approximation to worst case 1547 under-voltage tripping(88%, no delay)

Pacific DC Intertie trips

Widespread, commonmode tripping of DG (i.e. distributed solar PV results in system collapse

DG with LVRTDG without LVRT

Disturbance: Trip Pacific DC Intertie

1

2

1

2


Frequency Response Conclusions

For the conditions studied, system-wide frequency response can be maintained with high levels of wind and solar generation with both traditional and non-traditional approaches.

Traditional transmission system reinforcements to address local stability, voltage, and thermal problems include:

• Transformers • Shunt capacitors, (dynamic reactive support)• Local lines

Traditional approaches to meeting frequency response obligations are to commit synchronous generators with governors and to provide all response within an individual balancing authority area

Non-traditional approaches are also effective at improving frequency response including:

• Sharing frequency response resources• Frequency-responsive controls on inverter-based resources

• Wind• Utility-scale PV • CSP• Energy storage, (demand response) There are caveats in report


Transient Stability Analysis

14Preliminary Results: Not for Further

Distribution or Citation


Heavy Power Transfer Affects Response More than High Wind and Solar

Disturbance: Trip Pacific DC Intertie… NO RAS enabled

Heavy summer Base

Heavy summer Base with high COI flows

Heavy summer High Mix with high COI flows

High power transfer drives performance in both Base case and High Renewables case.

1

2

3

2

3

1

California Oregon Interface Power Flow (MW)

4,800 MW


Transient Stability in Northeastern WECC

L

Aeolus

500kV

Large Coal Plants


0

5000

10000

15000

20000

25000

30000

35000

LSP Base LSP HiMix LSP

HiMixXtrm

LSP Base LSP HiMix LSP

HiMixXtrm

DSW NorthEast

WIND

Steam

PV

PSH

Other

NUC

HYDRO

GEO

GasCT

CSP

Coal

CCPP

Bio

Light

Spring

Base

Coal Displacement in Light Spring Scenarios

Ge

ne

rati

on

pro

du

ctio

n (G

W)

PV=photo voltaic, PSH=pumped storage hydro, NUC =nuclear, GEO=geothermal, GasCT=gas fired combustion turbine, CSP=concentrating solar power, CCPP=combined cycle power plant, Bio=biomass

Light

Spring

High

Mix

Light

Spring

Extreme

Sensitivity

Desert Southwest

Light

Spring

Base

Light

Spring

Extreme

Sensitivity

Light

Spring

High Mix

Northeast (of the West)

Co

al

Co

al


System Non-Synchronous Penetration (SNSP)

• Percent of non-synchronous generation (i.e.,

inverter-based generation like wind and solar)

compared to synchronous generation in a system

• EirGrid (Irish grid operator) presently has 50% cap

on the amount of non-synchronous generation

allowed at any time

• ERSTF: a SNSP cap is similar to a SIM, but reflects

restrictions on short-circuit strength as well as

inertia


Synchronous vs. Non-synchronous

OK

Fail

Inverter-based MVA

Synchronous MVA

80% drop in Coal Dispatch. This

case passes Aeolus fault test.

90% drop in Coal Dispatch. This case

needs further reinforcement

Ge

ne

rati

on

co

mm

itm

en

t (b

ase

d o

n M

VA

ra

tin

g)

Heavy Summer Base

Heavy Summer Base

Heavy Summer Base

Heavy Summer Base





California Desert Southwest Northeast Northwest

Light Spring Base

Light Spring Base

Light Spring Base

Light Spring Base



Light Extreme Sensitivity







Dave Johnson Voltage

Light Spring Base


Light Spring Extreme

Light Spring Extreme with

synchronous condenser

conversion

Synchronous Condenser Conversion Results in

Acceptable Performance in Extreme Sensitivity

Reinforcements for Extreme sensitivity: 3 condensers total ~1700MVA plus ~500 MVArshunt banks.

Disturbance: Aeolus bus fault and line trip

1

3

2

4

1

2

3

4ERSTF: this is

transient voltage

collapse.

Pessimistic

dynamic load

model plays a

key role

ERSTF: 80%

reduction in coal

dispatch still

stable


Transient Stability Conclusions

For the conditions studied, system-wide transient stability can be maintained with high levels of wind and solar generation with both traditional and non-traditional approaches.

Traditional transmission system reinforcements to address stability, voltage, and thermal problems include:

• Transformers • Shunt capacitors, (dynamic reactive support)• Local lines

Non-traditional approaches are also effective at improving transient stability including:

• Synchronous condenser conversions• New wind and solar controls

There are caveats in report.


Study Conclusions

The Western Interconnection can be made to work

well in the first minute after a big disturbance with

both high wind and solar and substantial coal

displacement, using good, established planning and

engineering practice and commercially available

technologies.

The following detailed conclusions were word-

smithed by Technical Review Committee and include

the appropriate caveats.


Frequency Response ConclusionsFor the conditions studied:

• System-wide FR can be maintained with high levels of wind and solar generation

if local stability, voltage, and thermal problems are addressed with traditional transmission

system reinforcements (e.g., transformers, shunt capacitors, local lines).

• Limited application of non-traditional frequency-responsive controls on wind,

solar PV, CSP plants, and energy storage are effective at improving both frequency

nadir and settling frequency, and thus FR. Refinements to these controls would further

improve performance.

• Individual BA FR may not meet its obligation without additional FR from resources

both inside and outside the particular area. As noted above, non-traditional approaches

are effective at improving FR. Current operating practice uses more traditional approaches

(e.g., committing conventional plants with governors) to meet all FR needs.

• Using new, fast-responding resource technologies (e.g., inverter-based controls) to

ensure adequate FR adds complexity, but also flexibility, with high levels of wind and solar

generation. Control philosophy will need to evolve to take full advantage of easily

adjustable speed of response, with additional consideration of the location and size of the

generation trip.

• For California, adequate FR was maintained during acute depletion of headroom from

afternoon drop in solar production, assuming the ability of California hydro to provide FR.


Transient Stability Conclusions

For the conditions studied:

• System-wide transient stability can be maintained with high levels of wind and solar generation if local stability, voltage, and thermal problems are addressed with traditional transmission system reinforcements (e.g., transformers, shunt capacitors, local lines). With these reinforcements, an 80% reduction in coal plant commitment, which drove SNSP to 56%, resulted in acceptable transient stability performance.

• With further reinforcements, including non-standard items such as synchronous condenser conversions, a 90% reduction in coal plant commitment, which drove SNSP to 61%, resulted in acceptable transient stability performance.

• Additional transmission and CSP generation with frequency-responsive controls are effective at improving transient stability.


Other Conclusions• Accurate modeling of solar PV, CSP, wind, and load behavior is extremely important when

analyzing high-stress conditions, as all of these models had an impact on system

performance.

• Attention to detail is important. Local and locational issues may drive constraints on both FR

and transient stability.

• The location of generation tripping, e.g., DG vs. central station, is not as important as the

amount of generation that is tripped. However, widespread deliberate or common-mode DG

tripping after a large disturbance has an adverse impact on system performance. It is

recommended that practice adapt to take advantage of new provisions in IEEE 1547 that

allow for voltage and frequency ride-through of DG to improve system stability.

• Further analysis is needed to determine operational limits with low levels of synchronous

generation in order to identify changes to path ratings and associated remedial action

schemes, as well as quantify the impact of DG on transmission system performance.

• Because a broad range of both conventional and non-standard operation and control

options improved system performance, further investigation of the most economic and

effective alternatives is warranted. This should include consideration of the costs and

benefits of constraining commitment and dispatch to reserve FR, as well as the capital and

operating costs of new controls and equipment.


Thank [email protected]

flexibility assessment metrics - nerc

Documents