HLSS603Wk5%

SNLResilienceApril29.pdf

Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for

the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000

Energy Infrastructure Resilience

Framework and Sector-Specific Metrics

The Purpose of This Exercise

• The President mandated a Quadrennial Energy

Review to be jointly conducted by several US

Departments.

• The concepts on resilience being discussed

today will establish a foundation for a national

roadmap in resilience, including:

– Strategic national thrusts

– R&D thrusts

Defining Resilience

Presidential Policy Directive (PPD) 21 “the ability to prepare for and adapt to changing conditions and withstand

and recover rapidly from disruptions. Resilience includes the ability to

withstand and recover from deliberate attacks, accidents, or naturally

occurring threats or incidents.”

-PPD-21: Critical Infrastructure Security and Resilience

“without some numerical basis for assessing resilience, it would be

impossible to monitor changes or show that community resilience has

improved. At present, no consistent basis for such measurement exists. We

recommend therefore that a National Resilience Scorecard be established.”

-Disaster Resilience: A National Imperative, National Academy of Sciences

Goals For Today

� Begin a discussion about how to measure resilience

� Explore a general framework for developing energy resilience metrics

� Discuss ‘prototype’ resilience metrics for Oil, Gas, and Electricity

� Review plausible use-cases for electricity resilience metrics

� Collaboratively outline next steps

Takeaway Points

- R&D is needed to address this critical national

problem.

- Metrics are needed to enable resilience goals

and decisions for our US national strategy.

- The proposed framework applies common

principles across energy sectors

- We’re looking forward to your help!

SCENARIO CONCEPTS FRAMEWORK CASES SUMMARY

Lo a

d S

e rv

e d

( M

W )

Time

Total Load Served,

Nominal Conditions

Nominal Load Served

Illustrative Scenario: Nominal Conditions

Illustrative Scenario: Hurricane

Lo a

d S

e rv

e d

( M

W )

Time

Load Served, Hurricane

Nominal Load Served System #1 load served

Illustrative Scenario: Impact on Load Served

Hurricane affects ability to provide grid services

Lo a

d N

o t

S e

rv e

d (

M W

)

Time

Load Not Served, Hurricane Hurricane Load Not Served

R e

co ve

ry e

ff o

rt (

M a

n p

o w

e r)

Time

Labor, Hurricane

Hurricane

Recovery Effort

Lo a

d S

e rv

e d

( M

W )

Time

Load Served, Hurricane

Nominal Load Served System #1 load served

Illustrative Scenario: Hurricane Impacts

Hurricane damage yields significant impacts 10

Resilience-Enhancing Activities

� Utility prepares for hurricane

• Pre-positions recovery supplies

• Key assets outside of flooding areas

• Charges battery reserves

� While trying to cope with effects of damage, the utility

• Brings backup generation online

• Reconfigures lines to circumvent damaged assets

• Uses battery and reservoir discharge

� More rapid, less resource-intensive recovery

Lo a

d S

e rv

e d

( M

W )

Time

Load Served, Hurricane

Nominal Load Served System #1 load served

System #2 load served

Illustrative Scenario: Performance of a more resilient system

A more resilient system exhibits improved performance

Lo a

d S

e rv

e d

( M

W )

Time

Load Not Served, Hurricane

System #1

System #2

R e

co ve

ry E

ff o

rt (

M a

n p

o w

e r)

Time

Labor, Hurricane System #1

System #2

Comparison of Performance Indicators

N u

m b

e r

o f

d a

m a

ge d

a ss

e ts

Time

Damaged Assets, Hurricane

System #1

System #2

Translation to consequence

Performance Indicators

Alternative units:

Safety

Economics

Population affected

Etc…

Total Consequence 1

Total Consequence 2

Uncertainty

P ro

b a

b il

it y

o f

C o

n se

q u

e n

Consequence

Distribution of Consequence, Hurricane

System #1

System #2

Uncertain:

Disruption impacts

System response

Interdependencies with other systems

Resource availability

Etc…

Enabling Decisions

0.01

0.02

0.03

0.04

0.05

0.06

0 50 100 150 200 250

P ro

b a

b il

it y

o f

C o

n se

q u

e n

Consequence

Distribution of Consequence, Hurricane

System #1

System #2

Mean Mean

Extreme Values:

System 1

System 2

SCENARIO CONCEPTS FRAMEWORK CASES SUMMARY

Definition of a Metric

• A metric is a measure of something

– The unit ‘inch’ measures distances

– ‘Miles per hour’ measures speeds

• Metrics should not be confused with the values

that populate them

– 60 mph is an actual speed, where 60 populates the

metric

• We will be making a ‘speedometer’ for resilience

Resilience Complements Reliability

• Reliability is commonly applied to electric power, but

is informally applied to oil and gas sectors.

• This work does not seek to re-define, displace, or

extend existing reliability metrics

• We define resilience to be risk-based, with focus on

includes high consequences low probability threats

What Resilience Metrics Have to Do

– inform decision making

– provide validity (they properly discriminate)

– are repeatable (robustness to uncertainty)

– are feasible (implementable)

– be useable in a planning or operating context

– allow for uncertainty quantification

– be useable in an analytic context (such as an

optimization algorithm)

– the resiliency framework must be scalable

Metrics Inform Better Decision Making

Broad Categories of Decision Making For Energy

Infrastructure Systems

1. Policy decisions- how to direct national strategy

2. Planning decisions- whether to inform capital

investments

3. Operational decisions- informing real-time

decision making

SCENARIO CONCEPTS FRAMEWORK CASES SUMMARY

MEASUREMENTS e.g. voltage, frequency

power flow

PERFORMANCE

INDICATOR CONSEQUENCES METRICS

INFORMED

DECISION MAKING

POPULATING RESILIENCE METRICS

From Measurements to Performance

Indicators

Forecasts &

Measurements

Multiple Models / Algorithms

State

Estimation

Transient

Simulation

Dynamic

Simulation

• Voltage

• Frequency

• Power

• Load Forecasts

• Renewable Gen. Forecasts

• Topology

New

Algorithms

and Tools

• Transfer capability

• Generation shortages

• Stability margins

• ...

Impending threat

Performance Indicators

From Performance Indicators to to

Consequences

Performance

Indicators with

estimates of

uncertainty

Multiple Models / Algorithms

Financial

models

GIS tools

Sector-

specific

network

models

Conversion to

one or many

consequence

distributions

Human

behavior

estimators

Consequence

distributions

Lost revenue

Fatalities

Ability to

provide

public

services

Impact to

overall

economy

The Form of Resiliency Metrics

• Our proposed

resilience metrics

take the form of

probability density

functions

Many PDFs exist for the

same system. They

reveal resilience for

different threats and

different consequences

5E-09

1E-08

1.5E-08

2E-08

2.5E-08

3E-08

3.5E-08

0 10 20 30 40 50 60 70 80 90 100

We may set a

goal to

eliminate or

reduce

consequences

P ro

b a

b ili

ty C

o n

se q

u e

n ce

s fo

r a

C a

t 5

H u

rr ic

a n

This line

represents

system

resiliency

Consequences in Damage in Billions of Dollars

-OR-

Consequences in Lives Lost

-OR-

Consequences in Environmental Damage

CLOSING THE LOOP

Make

Decisions

Decision

Support

Analysis

Decisions Implemented

into the System

Populated Resiliency

Metrics

SCENARIO CONCEPTS FRAMEWORK CASES SUMMARY

ELECTRIC POWER USE CASE

Goal: Deciding between two different system improvements

Model: IEEE 14 Bus System

Hurricane

Hurricane winds and flooding disrupt operations

Damage Area

Performance Indicators:

Load and Labor

Consequences Total Cost:

$24M

Include Uncertainty: Baseline

Metric = Mean of the Distribution

Cost (Millions $)

P ro

b a

b il

it y

Resilience Comparison: Design Decision

Metric = Mean of the Distribution

Backup Generator

Buried Lines

Baseline

Cost (Millions $)

P ro

b a

b il

it y

OIL USE CASE

Goal: Reassess system resilience after changes

Oil System Earthquake Example

National Transportation Fuel Model Transmission Pipelines,

Refineries, and Terminals

The DHS/SNL National Transportation Fuels Model was used for this simulation example

New Madrid Earthquake Performance

Indicators

Convert Output to Consequence

Convert using

• Consequence model

• Distribution of outcomes from multiple

simulations

Economic consequences (million $)

D a

il y

C o

n su

m p

ti o

n S

h o

rt fa

ll (

kb b

Compare Resilience: Assessment over time

Economic consequences (million $)

P ro

b a

b ili

Prior to 2009 Midwest refineries increased use of crude

from resulting in increased resilience to a New Madrid

earthquake

0.1

0.2

0.3

0.4

0.5

0.6

2009

2003

NATURAL GAS USE CASE

Goal: Select policy for use rules of asset in emergency

Natural Gas Earthquake Example

North American Natural Gas Network

San Andreas Earthquake Performance

Indicators

This calculation was performed using the Gas Pipeline Competition Model (GPCM), which was

developed by, and licensed from, Robert Brooks Associates Consulting (RBAC).

Convert Output to Consequence

Convert using

• Consequence model

• Distribution of outcomes from multiple

simulations

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

50 60 70 80 90

P ro

b a

b il

it y

Economic Consequences (million USD)

100

0 1 2 3 4 5 6 7

Time (days)

LA B

a si

n S

u p

p ly

( %

)

Restricted Storage Withdrawals

Compare Resilience: Policy Options

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

- 50 100 150 200 250

P ro

b a

b il

it y

Economic Consequences (million USD)

Unconstrained Storage Withdrawals Constrained Storage Withdrawals

SCENARIO CONCEPTS FRAMEWORK CASES SUMMARY

Challenges

• Strategic

– Stakeholder engagement

• Interdependencies

– Common models, knowledge sharing

• R&D

– Decision support tools, consequence estimation

Energy Resilience is a National Priority

• Energy resilience metrics are needed to make

measure baselines and create goals

• Metrics should allow depth of application, but

should simplify when desired

• R&D will be needed for advanced decision

support

• Success will depend on a multi-disciplinary

team