Smart Grid Vision: Vision for a Holistic Power Supply and Delivery Chain

Stephen Lee

Senior Technical Executive Power Delivery & Utilization

November 2008

2© 2008 Electric Power Research Institute, Inc. All rights reserved.

Smart GridTwo Way Communications….Sensors…….Intelligence

Smart Smart GridGrid

Need an Objective Assessment of the Potential for Smart Transmission and the Path to Achieve it

Hype

3© 2008 Electric Power Research Institute, Inc. All rights reserved.

End-uses & DR

Distribution SystemTransmission System

Energy Storage

Fuel Supply System

Fuel Source/Storage

Power Plants

Renewable Plants

Data CommunicationData Communication

Wide Area ControlWide Area Control

Sensors

Controllers

ZIP

M

Dynamic Load ModelsDynamic Power Plant Models

End-to-End Power Delivery Chain Operation & Planning

Monitoring, Modeling, Analysis, Coordination & ControlMonitoring, Modeling, Analysis, Coordination & Control

4© 2008 Electric Power Research Institute, Inc. All rights reserved.

Alarm Management

Most Needed Capabilities Requiring Research

• Hierarchical Integration of Entire Supply & Delivery Chain

• Optimal End-to-End Dispatch under Uncertainties

• Dynamic Models of Generators and Loads

• Online Alarm Root-Cause Diagnostics

• Prevention of Cascading Outages, Safety Nets

• Fast System Restoration After Blackouts

ProtectionSCADA

Data Management

Network Management

Security

ChainIntegration

OutageManagement

Dynamic Models

End-to-End Dispatch

Restoration Protection

5© 2008 Electric Power Research Institute, Inc. All rights reserved.

Source: California ISO

Tehachapi, California Wind Generation in April – 2005

0

100

200

300

400

500

600

700

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

Hour

MW

Average

Each Day is a different color.

Day 29

Day 5 Day 26

Day 9

Could you predict the energy production for this wind park either day-ahead or 5 hours in advance?

6© 2008 Electric Power Research Institute, Inc. All rights reserved.

Methods of Coping with Wind Uncertainty

• Short-Term– Better wind forecasting– Carry more operating and spinning

reserve to handle up and down ramps of wind output

– Rapid coordination with demand response and energy storage

• Long-Term– Build more energy storage, e.g.,

Compressed Air Energy Storage (CAES)

– Controllable demand response– Holistic planning of transmission,

generation and demand• Virtual Service Aggregator

0

50

100

150

200

250

300

350

Sep06 1

2:0

0

Sep06 1

4:2

4

Sep06 1

6:4

8

Sep06 1

9:1

2

Sep06 2

1:3

6

Sep07 0

0:0

0

Sep07 0

2:2

4

1-min MW

fcst 1

fcst 2

fcst 3

Potential wind curtailment

CAES

7© 2008 Electric Power Research Institute, Inc. All rights reserved.

Potential Role of the Virtual Service Aggregator

TraditionalPower Plants

TraditionalPower Plants

RenewableResources

RenewableResources

EnergyStorage

EnergyStorage

TransmissionGrid

TransmissionGrid

Real RegionalControl Center

Real RegionalControl Center

VirtualService

Aggregator

Σ

VirtualService

Aggregator

Σ

End Usesand

DistributedResources

End Usesand

DistributedResources

FinancialSettlement ofNet Difference

Power Flow

FinancialTransaction

8© 2008 Electric Power Research Institute, Inc. All rights reserved.

Alarm Management

Alarm Management

• Need to diagnose root-cause of alarm messages

• Need to link diagnosis to operator procedure

• Current EMS alarm management uses technologies of the 1970s

• Need to integrate all sources of data and messages, through a hierarchical approach

9© 2008 Electric Power Research Institute, Inc. All rights reserved.

Why Accurate Load and Generator Models Are Needed?

• Inadequacy of current model data– Inaccurate voltage recovery

simulation after disturbances– Uncertainty about generator

reactive power capabilities• Implications

– Uncertainty about the stability margin of the power grid

– Unaware of real risk of cascading blackouts or voltage collapse, or

– Under utilization of available stability margin for greater economic benefits

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20

0.2

0.4

0.6

0.8

1

Time (seconds)

Vol

tage

(pu

)

Hassayampa 500 kV

MeasuredSimulated H=0.3Simulated H=0.03

Simulated H=0.05Simulated H=0.1Simulated H=0.2Simulated H =0.5

Southern Co.’s GenVARRTM

10© 2008 Electric Power Research Institute, Inc. All rights reserved.

Major Power System Disturbances

0

10

20

30

40

50

60

70

0 10 20 30 40 50 60

Duration in Hours

Lo

ad

Lo

st in

GW

1 - 2003 NE

2 - 1965 NE

3 - 1977 NYC

4 - 1982 WSCC 5 - 1996 WSCC

6 - 1996 WSCC

7 - 1998 MW

Restoration Objectives: ▪ Minimize Duration of Outages ▪ Minimized Unserved Loads ▪ Avoid Equipment Danage

Source: Mike Adibi, NSF/EPRI Workshop on Understanding and Preventing Cascading Failures in Power Systems, Oct 28, 2005.

Effective System Restoration Can Reduce The Societal Impact Of Widespread Blackouts

•Operators need online decision support for restoration strategies

•How can automation be used to improve system restoration?

11© 2008 Electric Power Research Institute, Inc. All rights reserved.

Prevention of Cascading Outages – Safety Nets

• Application of SynchroPhasor Measurements for Controlled Separation, Load Shedding and Generation Rejection

– Controlled separation is an effective last resort to mitigate severe cascading failures

– Voltage Instability Load Shedding

– Online risk monitoring of potential cascading outages

12© 2008 Electric Power Research Institute, Inc. All rights reserved.

Thank you