Communications and Networking for Smart Grid...

Communications and Networking

for Smart Grid Systems

Dusit Niyato

Nanyang Technological University (NTU), Singapore

Rose Qingyang Hu

Utah State University

Ekram Hossain

University of Manitoba, MB, Canada

Yi Qian

University of Nebraska-Lincoln

1

IEEE GLOBECOM 2011, Houston, USA December 9, 2011

Tutorial Outline

1. Introduction, Background, and Overview of Smart Grid Systems

2. Data Communication Requirements in Smart Grid

3. Communication Architectures, Area Networks, and Components for Smart Grid

4. Data Communications and Networking in Smart Grid

5. Cyber Security and Privacy in Smart Grid Communications Infrastructure

6. Field Trials and Case Studies

7. Open Issues and Future Research Directions

8. Summary

2 IEEE GLOBECOM'11

Introduction • What is smart grid?

– Smart grids – add communication capabilities and intelligence to

traditional grids

• What enables smart grids?

– Intelligent sensors and actuators

– Extended data management system

– Expanded two way communications between power generation,

distribution, and customers

– Network security

– etc.

IEEE GLOBECOM'11

3

Smart Grid: The “Energy Internet”

2-way flow of electricity and information

Standards Provide a Critical Foundation 4

Transition

• Transition from traditional power grid to smart grid

IEEE GLOBECOM'11

5

Motivations

Smart Grid Enables:

• Higher Penetration of

Renewables

• Smart Charging of

Electric Vehicles

• Consumers to Control

Energy Bills

• Efficient Grid

Operations &

Reduced Losses

• Reduced Distribution

Outages

• Improved System

Reliability & Security

IEEE GLOBECOM'11

6

http://images.google.com/imgres?imgurl=http://fuel-efficient-vehicles.org/energy-news/wp-content/uploads/chevy-volt-concept-08.jpg&imgrefurl=http://fuel-efficient-vehicles.org/energy-news/?p=355&usg=__VtDJ6oE8l4-QFhKMA1Go_DYvznE=&h=600&w=800&sz=38&hl=en&start=55&um=1&itbs=1&tbnid=saroAx2bI6ZmlM:&tbnh=107&tbnw=143&prev=/images?q=chevy+volt&start=40&um=1&hl=en&sa=N&ndsp=20&tbs=isch:1

http://www.google.com/imgres?imgurl=http://news.cnet.com/i/bto/20080729/insight.jpg&imgrefurl=http://news.cnet.com/greentech/?keyword=Tendril&usg=__Fw6lu5bRtiDW35ToSjLTWFI9mx0=&h=177&w=216&sz=8&hl=en&start=2&um=1&itbs=1&tbnid=y4hXAtPZSKkG-M:&tbnh=88&tbnw=107&prev=/images?q=electricity+in+home+display&um=1&hl=en&tbs=isch:1

http://www.google.com/imgres?imgurl=http://stopthecap.com/wp-content/uploads/2009/10/Bucket_Truck-Pole_Repair.jpg&imgrefurl=http://stopthecap.com/tag/fairpoint/&usg=__0wqCOWlSwl_x9m_SMG65vS_RttY=&h=1500&w=1500&sz=364&hl=en&start=16&um=1&itbs=1&tbnid=yjgeQCHlAMv0VM:&tbnh=150&tbnw=150&prev=/images?q=utility+line+outage+bucket+truck&um=1&hl=en&tbs=isch:1

Primary objectives

• National integration

• Self healing and adaptive –Improve distribution and

transmission system operation

• Allow customers freedom to purchase power based on

dynamic pricing

• Improved quality of power-less wastage

• Integration of large variety of generation options

IEEE GLOBECOM'11

7

Economic and social benefits

• Provide Customer Benefits

• Reduce Peak Demand

• Increase Energy Conservation & Efficiency

• Reduce Operating Expenses

• Increase Utility Worker Safety

• Improve Grid Resiliency and Reliability

• Reduce Greenhouse Gas Emissions

• Promote Energy Independence

• Promote Economic Growth & Productivity

IEEE GLOBECOM'11

8

More Background on Smart Grid

IEEE GLOBECOM'11

9

Distributed Generation

• Hybrid Energy Resource

– Fossil-Fuel

– Wind

– Solar

– Bio-Mass

– Batteries

– Capacitors

– Flywheel

– Etc.

IEEE GLOBECOM'11

10

Smart Metering

(b) Microsoft Hohm (a) Google PowerMeter

• Automatic Metering

– Automatic Meter Reading (AMR)

– Automated Metering Management (AMM)

– Advanced Metering Infrastructure (AMI)

Example smart metering systems:

IEEE GLOBECOM'11

11

Intelligent electronic devices (IEDs)

• Protection relay

• Auxiliary relay

• Cheap contractors

• Remote terminal units

• Circuit breaker monitor

• Revenue meters

• Solar flare detectors

• Power quality monitors

• Phasor measurement units

• Communication processors

• Communication alarm

• Etc.

GE CFD Intel 4004

12

Monitoring and Controlling

– Supervisory Control And Data Acquisition (SCADA)

– Energy management system (EMS)

– Information and Communications Technology (ICT)

IEEE GLOBECOM'11

13

Telecontrol

• Different protocols for different operations

– Proprietary protocols (more than 100)

– Standards

• SCADA

• Modbus

• DNP

• IEC61850

IEEE GLOBECOM'11

14

SCADA Protocols

• Siemens quad 4 meter

• CONITEL 2000

• CONITEL 2100

• CONITEL 3000

• CONITEL 300

• HARRIS 5000

• HARRIS 5600

• HARRIS 6000

• UCA 2.0 or MMS

• PG & E 2179

• MODBUS

• DNP3

• ICCP

• IEC 61850

IEEE GLOBECOM'11

15

General Protocols

• MODBUS -Primitive without security and not very

extensible

• DNP3 –Advanced SCADA protocol

• DNP1 and 2 are proprietary protocols

• IEC 61850 the most used protocol for new implementations

• ICCP

IEEE GLOBECOM'11

16

Groups working on smart grids

• UCA International user group www.ucaiug.org

• International electrochemical commission www.iec.ch

• Electric power research institute www.epri.com

• Intelligrid consortium and architecture www.intelligrid.epri.com

• IEEE smart grid www.smartgrid.ieee.org

• NIST csrc.nist.gov

IEEE GLOBECOM'11

17

http://www.ucaiug.org/

http://www.iec.ch/

http://www.epri.com/

http://www.intelligrid.epri.com/

http://www.smartgrid.ieee.org/

http://csrc.nist.gov/

http://csrc.nist.gov/

Smart Grid Communication Architecture

IEEE GLOBECOM'11

18

NIST Conceptual Reference Model for

Interoperability

IEEE GLOBECOM'11

19

Communication Media

• Urge for new FCC allocation for smart grids

• PLC –Power line carriers

• Ethernet

• WLAN

• ZigBee

• Bluetooth

• Optical fiber

• Microwave

• Etc.

IEEE GLOBECOM'11

20

Data Communications in Smart Grid Optimal Network(s)

• Broadcast data (Demand Response, price signals, emergency events, etc.)

– Low volume, infrequent

– Can use currently available communication infrastructure (cellular, broadband, WiFi, Pager) with standard internet security measures

• Real-time Consumption Data (high volume, frequent)

– Useful primarily for real-time control & usage information to consumer

– We favor meter premises where displays & controllers can locally act upon this data along with pricing information

• Minimizes risk (privacy & network stability) and maximizes benefit from real time info.

• Raw Billing Data (reading when price changes)

– Utility operations

• Aggregate Data

– Comparison over time & among neighbors, best practices, consumption pattern recognition, suggest corrective actions, etc.

• Utility or third party cloud-based applications operating on anonymous summary data

• Little risk for privacy or network stability in case of breach of security

• Can use standard internet communication with standard security measures

• T&D: relatively few points (substations); mission-critical, but already connected

Broadband

Cellular

WiFi, etc.

Direct meter

to HAN

AMI

Internet portal

Existing

connectivity

IEEE GLOBECOM'11

21

Match Info To The Communications Medium

Information

category

Smart Grid Signals Detailed Consumption Data

Examples ToU pricing, critical peak pricing,

reliability, carbon content, etc.

Periodic meter readings (e.g., once

a minute)

Location of

information

Utility servers connected to Internet Embedded meter hardware

Evolution potential High, as new applications arise (e.g.,

PHEVs, micro-grids)

Very low

Optimal approach

AMI-centric approach

Communication

medium

General telecom infrastructure

(Internet): broadband, cellular,

municipal WiFi, etc.

Specialized embedded hardware

(short-range radio, power-line

carrier, etc.)

IEEE GLOBECOM'11

22

Priority and types of information

Communication model , source: NIST Vol 1

IEEE GLOBECOM'11

23

Data Communication Requirements

in Smart Grid

IEEE GLOBECOM'11

24

Requirements

• Latency

• Bandwidth

• Interoperability

• Scalability

• Security

• Standardization

IEEE GLOBECOM'11

25

Latency

• The real-time operational data communications in smart grid include online

sensor/meter reading and power system control signals.

• The communication is characterized by the fact that most of interactions must

take place in real time, with hard time bound.

• The communication requirements define the design of the technical solutions.

• For real-time sensing/metering purposes, reading messages should be

transmitted within a very short time frame.

– For instance, the maximum allowed time is in the range of 12-20 ms, depending on

the type of protection scheme which origins from the fact that the disconnection of

fault current should within approximately 100 ms.

• Power System Control signals mainly include supervisory control of the power

process on secondary or higher levels. These systems are of the kind

SCADA/EMS.

– Measured values must not be older than 15 seconds, when arriving at the control

center. Breaking information shall arrive no later than 2 seconds after the emergency

event has occurred

IEEE GLOBECOM'11

26

Bandwidth

• As more and more interconnected intelligent elements are added to the

electricity network with the evolution of the smart grid, the

communication infrastructure should be able to transport more and more

messages simultaneously without severe effect on latency.

• The network bandwidth must increase faster than the demand of these

interconnected intelligent elements in the network.

• An Example: (A. Aggarwal, S. Kunta, P. K. Verma, “A proposed communications

infrastructure for the smart grid,” in Innovative Smart Grid Technologies (ISGT), 2010,

pp. 1-5.) – Model the communication bandwidth requirements for a moderate size electricity distribution system. In this

model, a distribution substation is connected to 10,000 feeders and each feeder connects to 10 customers.

– Assuming that every electric meter generates a message every second to the distribution substation, the total is

100,000 messages per second. The feeders themselves will generate messages to each other and to the distribution

substation.

– The authors in this paper modeled the messages in the smart grid arriving at servers located at the control center

as M/M/1 traffic. Then, the transmission line bandwidth is evaluated over 100 Mbps through the M/M/1 queuing

model. It can be observed that this situation results in a very poor bandwidth utilization of the transmission

facilities as well.

– Unfortunately, a higher level of utilization will not permit meeting the assumed latency constraint.

IEEE GLOBECOM'11

27

Interoperability

• The ability of 2 or more networks, systems, devices, applications, or

components to communicate & operate together effectively, securely, &

without significant user intervention

– Communication requires agreement on a physical interface &

communication protocols

– Exchanging meaningful & actionable information requires common

definitions of terms & agreed upon responses

– Consistent performance requires standards for the reliability, integrity,

and security of communications

– Interoperability may include:

• “Plug and play”: connect them & they work together

• Interchangeability: Ability to readily substitute components

IEEE GLOBECOM'11

28

Interoperability (cont‟d)

IEEE GLOBECOM'11

29

Scalability

• Phenomena in Smart Grid Communication

IEEE GLOBECOM'11

30

Standards • EISA 2007 Directs National Institute of Standards & Technology (NIST) to:

– Coordinate the development of model standards for interoperability of smart grid devices and systems

• Create flexible, uniform, and technology neutral standards

• Enable traditional resources, distributed resources, renewables, storage, efficiency, and demand

response to contribute to an efficient, reliable grid

• EISA Directs FERC, when sufficient consensus, to:

– Adopt standards necessary to insure smart-grid functionality and interoperability in the interstate transmission of electric power, and regional and wholesale electricity markets

– EISA did not expand FERC‟s Federal Power Act authority to enforce standards

• State Commissions:

– May adopt standards by regulation, separately or in parallel with FERC

– May consider standards when approving utility investments

• Considerations for Regulators:

– Ensuring interoperability & security, without impeding innovation

– Consistent action will influence the vendor community

– Vendors often will follow standards that are not legally mandated

– SGIP standards reflect efforts to build broad stakeholder consensus

IEEE GLOBECOM'11

31

Standardization (cont‟d)

• IEEE

– IEEE P2030

• Power Engineering Technology

• Information Technology

• Communications Technology

• IEC – IEC 61968 - Distribution Management

– IEC 61970 - Common Information Model

– IEC 60870 - Inter-control Center Communication Protocol

– IEC 62210 - Data and Communication Security

– IEC 62357 - Reference Architecture

– IEC 61850 - Standard for Design of Substation Automation

• IEC 61850-7-420 - Integration of Distributed Energy Resources

• IEC 61850-7-410 - Integration of Hydro Resources

– IEC 61400 - Integration of Wind Farms to Utility Communication Network

– IEC 62056 - Communication

IEEE GLOBECOM'11

32

Security

• DISA Security Technical Implementation Guides (STIGs)

• FIPS 201

• North American Electrical Reliability Corporation-Critical

Infrastructure Protection (NERC CIP)

• National Infrastructure Protection Plan (NIPP)

• IEEE 1402

• International Society of Automation(ISA)

• ISO 17799

• NIST GWAC

– DEWGs

• Home-to-Grid (H2G)

• Building-to-Grid (B2G)

• Industrial-to-Grid (I2G)

• Transmission and Distribution (T&D)

• Business and Policy (B&P)

IEEE GLOBECOM'11

33

Summary

Application Security Bandwidth Reliability Coverge Latency Back-up Power

Advanced Metering Infrastructure High 14-100 kbps per node 99.0-99.99% 20-100 % 2000 ms 0-4 hours

AMI Network Management High 56-100 kbps 99.00% 20-100% 1000-2000 ms 0-4 hours

Automated Feeder Switching High 9.6-56 kbps 99.0-99.99% 20-100% 300-2000 ms 8-24 hours

Capacitor Bank Control Medium 9.6-100 kbps 96.0-99.00% 20-90% 500-2000 ms 0 hours

Charging Plug-In Electric Vehicles Medium 9.6-56 kbps 99.0-99.90% 20-100% 2000 ms - 5 min. 0 hours

Demand Response High 56 kbps 99.00% 100% 2000 ms 0 hours

Direct Load Control High 14-100 kbps per node 99.0-99.99% 20-100 % 2000 ms 0-4 hours

Distributed Generation High 9.6-56 kbps 99.0-99.99% 90-100% 300-2000 ms 0-1 hour

Distribution Asset Management High 56 kbps 99.00% 100% 2000 ms 0 hours

Emergency Response Medium 45-250 kbps 99.99% 95% 500 ms 72 hours

Fault Current Indicator Medium 9.6 kbps 99.00-99.999% 20-90% 500-2000 ms 0 hours

In-home Displays High 9.6-56 kbps 99.0-99.99% 20-100% 300 -2000 ms 0-1 hour

Meter Data Management High 56 kbps 99.00% 100% 2000 ms 0 hours

Network Protection Monitoring Medium - High 56-100 kbps 99.00-99.999% 100% 2000-5000 ms 0 hours

Outage Management High 56 kbps 99.00% 100% 2000 ms 0 hours

Price Signaling Medium 9.6-56 kbps 99.0-99.90% 20-100% 2000 ms - 5 min. 0 hours

Real-time Pricing High 14-100 kbps per node 99.0-99.99% 20-100 % 2000 ms 0-4 hours

Remote Connect/Disconnect High 56-100 kbps 99.00% 20-100 % 2000-5000 ms 0 hours

Routine Dispatch Medium 9.6-64 kbps 99.99% 95% 500 ms 72 hours

Transformer Monitoring Medium 56 kbps 99.00-99.999% 20-90% 500-2000 ms 0 hours

Voltage and Current Monitoring Medium 56-100 kbps 99.00-99.999% 100% 2000-5000 ms 0 hours

Workforce Automation Medium 256-300 kbps 99.90% 90% 500 ms 8 hours

CURRENT FUNCTIONAL REQUIREMENTS

National Broadband Plan: RFI Communications Requirements

Comments of Utilities Telecom Council, July 12, 2010

IEEE GLOBECOM'11

34

Challenges for Smart Grid Communication Infrastructure

• Complexity

• Efficiency

• Reliability

• Security

IEEE GLOBECOM'11

35

Complexity

• Need to support multi-physics approach

• Need to support multidisciplinary approach

• Need to support dynamic and reconfigurable model

level definition

• Need to provide visualization to support system

analysis

• Need to provide support for uncertainty propagation

IEEE GLOBECOM'11

36

Efficiency

• Better Telemetry

• Faster Controls

• More Robust Controls

• Embedded Intelligent Devices Communication

• Integrated And Secure Communications

• Enhanced Computing Capabilities

• Internet Technology

37

Reliability

• Renewable Resources

• Demand Response

• Load Management

• Storage Devices

IEEE GLOBECOM'11

38

Security

• Information security domains

– Public, supplier, maintainer domain

– Power plant domain

– Substation domain

– Telecommunication domain

– Real-time operation domain

– Corporate IT domain

• SCADA

– De-coupling between operational SCADA/EMS and admin IT

– Governmental coordination on SCADA security

• Threats to

– AMI (similar to WSN)

– SCADA

IEEE GLOBECOM'11

39

Communication Architectures

• Communication Architecture and Model for Distribution

Network

• Home-Area Networks (HANs)

• Neighborhood-Area Networks (NANs)

• Wide-Area Networks (WANs)

• Sensor and Actuator Networks (SANETs)

40 IEEE GLOBECOM'11


Communication Architecture and Model for Distribution

Network

41 IEEE GLOBECOM'11

DAU/

NAN GW

Generation

MDMS

Tansmission and Distribution Customer

Premises

Transmission

Substation

Solar

EnergyWind

Turbines

Smart Meter/

HAN GW

Communication Core Network

E.g., TCP/IP Network, WiMax, Cellular (GSM or CDMA), Ethernet

Control CenterControl Center Control Center

HAN

NANWAN

Ele

ctri

cal

Infr

ast

ruct

ure

Co

mm

un

ica

tio

n

Infr

ast

ruct

ure

Customer Networks

E.g., ZigBee, WiFi,

PLC

Legends: DAU=Data Aggregator Unit, MDMS=Meter Data Management System, HAN=Home Area Network,

NAN=Neighborhood Area Network, WAN=Wide Area Network , GW = Gateway

Electric FlowInformation Flow

Distribution

Substation

Distribution

Feeder

Sensor Network

Advanced Metering Infrastructure (AMI)

Last Mile Connection

Transmission

Feeder


Communication Architecture and Model for

Distribution Network

• Smart grid follows the same electrical architecture

• Electricity is delivered from the generation to consumers

through transmission and distribution substations

• Transmission substation delivers electricity from power

generation plant over a high voltage transmission line (over

230kV) to the distribution substation

• Distribution substation converts the electric power to

medium voltage level

• Distribution feeder then converts the medium voltage to

lower level for distributing to the consumer‟s end

42 IEEE GLOBECOM'11


Customer Premise and Customer Network

43 IEEE GLOBECOM'11

NAN1

HAN3HAN1HAN2

Smart Meter

(HAN Gateway)

DAU/NAN GW

Bluetooth/ZigBee/

WiFi

BACnet, KNX,

PLC protocol

Smart Devices

(e.g., AC) With

Sensors

Display

HAN3HAN1

HAN2

NAN2

DAU/NAN GW

MDMS

Control

Center

HAN3

Control

Center

MDMS


Customer Premise and Customer Network

• Data aggregator unit (DAU) also referred to as NAN GW

acts as a data sink to collect and relay the information from

the consumer side to meter data management system

(MDMS)

• MDMS will provide storage, management, and processing

of meter data for proper usage by other power system

applications and services

44 IEEE GLOBECOM'11


Home-Area Networks (HANs)

• HAN (sometimes referred to as Premise Area Network

(PAN) or a Building Area Network (BAN)) is the smallest

subsystem in the hierarchical chain of smart grid

• HAN provides a dedicated demand side management

(DSM), including energy efficiency management, and

demand response by proactive involvement of power users

and consumers

• HAN consists of smart meter, smart devices with sensors

and actuators, and in-home display for energy management

system (EMS)

– EMS will provide means of reducing energy consumption by

monitoring and controlling different electrical appliances

45 IEEE GLOBECOM'11


General Structure

46 IEEE GLOBECOM'11

Electric supply from

Transmission

HAN

Gateway

In-Home

Display

DAU/NAN

Sensors

Light

Temperature

Voltage

Wired/Wireless

Connection

(e.g., Zigbee,

BACnet

Smart

Devices

Smart

Devices

Smart

Devices

Actuators

ActuatorsActuators

(Smart Meter

or

Dedicated

in-home

Gateway)

Enabling Communications Technologies

• Short Range Wireless Technologies

– Wi-Fi, Bluetooth, ZigBee, Z-Wave

• Z-Wave:

– Proprietary wireless standard designed for home control automation,

specifically to remote control applications in residential homes

– Z-Wave was originally developed by Zensys A/S and is being

marketed by Z-Wave Alliance

– Z-Wave wireless protocol provides reliable and low-latency

communication of small data packets within HANs

– Z-Wave also uses a mesh networking approach with source routing

47 IEEE GLOBECOM'11


Enabling Communications Technologies

• Z-Wave:

– Bandwidth: 9,600 bit/s or 40 kbit/s

– Modulation: GFSK

– Range: Approximately 100 feet (or 30 meters)

– Frequency band: The Z-Wave Radio uses the 900 MHz ISM band

• 908.42 MHz (United States)

• 868.42 MHz (Europe)

• 919.82 MHz (Hong Kong)

• 921.42 MHz (Australia/New Zealand)

48 IEEE GLOBECOM'11


Enabling Communications Technologies: Wireless Technologies

49 IEEE GLOBECOM'11

Key Criteria WiFi Bluetooth ZigBee Z-Wave

Feature - Designed for providing wireless

connection for accessing Internet

and is direct replacement to

traditional Ethernet

network

-Designed for consumer electronics to

provide short-range wireless

communication

to connect a wide range of devices

easily and quickly

- Designed specifically for industrial and

home automation for connecting

sensors,

monitors and control devices

- Designed for home automation,

specifically to remote control

applications in residential home such as

light, entertainment systems, etc

Frequency Band - 2.4/5 GHz - 2.4 GHz - 2.4 GHz, 915MHz and 868MHz - 900 MHz

Standards - International Standard (IEEE

802.11 a/b/g/n)

- Open Standard

- International Standard (IEEE 802.15.1)

- Open Standard

- International Standard (IEEE 802.15.4)

- Open Standard

- Proprietary Standard (Z-Wave

Alliance and Zensys)

- Closed standard

Speed - 54 Mbps ( 802.11. b/g)

- 150 Mbps (802.11 n)

- 2.1 Mbps (V 2.0)

- 20 Mbps (V 3.0, recently released)

- 250 Kbps - 9600 bits/s

Range - 70m (indoor) to 250m (outdoor) - 10m - 70m (indoor) to 400m (outdoor) - 30m (indoor) to 100m (outdoor)

Power Consumption - High - Lower than WiFi - Lower than WiFi and Bluetooth - Almost same as ZigBee

Maximum Nodes - 2007 - 8 - > 64000 - 232


Enabling Communications Technologies: Wireless Technologies

50 IEEE GLOBECOM'11

Key Criteria WiFi Bluetooth ZigBee Z-Wave

Security - WEP (Wired Equivalent privacy)

-WPA (Wi-Fi Protected access)

- WPA2

- E0 stream cipher

- More Secure than WiFi

- 128 AES (Advanced

Encryption Standard )

keys

- 3 DES(Triple Data Encryption

Standard)

Strength - Easy to deploy,

equipment costs

dropping rapidly

- Supports mesh topology

- Most popular protocol for transferring

data and wireless alternative to RS-

232 data cables

- Supports ring topology

- Low power requirements and

implementation costs

- Particularly designed for use in

industrial and home automation or

security applications

- Scalable and flexible


- Low power, low latency, and low cost

- Less interference due to use of sub-

GHz frequency

- Higher propagation range 2.5 times

the 2.4 GHz signal


Concern - High power consumption

- Higher data latency

- Additional security layer should be

implied to use WiFi within HAN

-Lack of proper installation in consumer

portal context such as fire alarm, security

sensors, etc

- Does not support mesh networking

- Limited range and low data rates

- Interference due to overlapping with

WiFi standard

- Low data rates

- Requires to add devices into network

manually

- Slightly higher installation cost than

ZigBee

- Offers less flexibility due to close

nature



Enabling Communications Technologies: Wired Technologies

51 IEEE GLOBECOM'11

Key

Criteria

X10 HomePlug GP BACnet KNX

Feature - Simple and popular protocol designed for

providing simple automation functionality such as

on and off.

-Designed specifically for smart grid to

provide lower power consumption

- A data communication protocol that

attempts to unifies all the proprietary

communication protocol into single

communication language

- Global standard protocol designed

basically for home automation and

control

Wireless

Support

- Yes

- 310 MHz U.S. 433 MHz European

- Yes

- Recent ZigBee/HomePlug initiatives

- - Yes

- KNX RF (868.3 MHz)

Standards - De facto Standard

- Open Standard

- International Standard (IEEE 1901) - ANSI/ASHRAE 135-2008

- ISO 16484-5

- Open Standard

- CENELEC EN 50090 and CEN EN

13321-1

- ISO/IEC 14543-3

- GB/Z 20965, ANSI/ASHRAE 135

Speed - 20 bits/s - 4 to 10 Mbps - Depends on choice of LAN technology

used

- wired 9.6 kbps

- wireless 16.4 kbps

Maximum

Nodes

- 256 - 253 (theoretically)

- 10 (Practically)

- No limit - 57600 network nodes for wired

connection


Enabling Communications Technologies: Wired Technologies

52 IEEE GLOBECOM'11

Key

Criteria

X10 HomePlug GP BACnet KNX

Security - lack of encryption - AES pro 128

Security (128-bit triple AES encryption

and a time lock)

- Assume that all devices are sitting

behind a firewall

- EIBsec

Strength - Commonly used with variety of equipment

available in the market

- No installation cost as uses power line

-Ubiquitous reach throughout the home

environment

-Interoperability with consumer home

networking

-Low-cost and low-power network

interfaces

-Cross-compatibility between wired and

wireless Smart Grid applications

- Well established as an enabler for

commercial building automation

technologies

- Already has the needed functionality

for energy management and load control

- Independent of current LAN or WAN

technologies

-Scalable

- Interoperable with other KNX

products

- Hardware/software independent

- Well established promoter which

provide any application for home

control

- Compatible with any buildings

Concern - Extreme low data rate

- Lack of standard and security

- Limited functionality

- Prone to interference from neighbors using the

same X10 device addresses

- Limited connection (10) when

transferring data simultaneously

- Susceptible to power line interference

and old wiring in home.

- client-server system might create

bottleneck when fully deployed to

consumer premise

- Security concerns

- Object model is limited to low-level

types

- Low data rates

Neighborhood-Area Networks (NANs)

General Structure

53 IEEE GLOBECOM'11

Utility Network Back Bone

Wide Area Network

(WAN)

Neighborhood Area

Network (NAN)

HAN = Home Area Network

Cel

lula

r,IP

net

wor

k,

BPL

, WiM

AX

PLC,

ANSI C12

Neighborhood Area

Network (NAN)

Neighborhood Area

Network (NAN)

Neighborhood Area

Network (NAN)

MDMS

MDMS

DAU

HAN

DAU

HAN

DAU

HAN

DAU

HAN

Cellu

lar,IP n

etwork

,

BP

L, W

iMA

X



• NAN connects multiple HANs together

• Wired Technologies

– Power Line Communication (PLC):

• Ultra narrow band (UNB) operates in 0.3-3 kHz bands

• Narrow band (NB) PLC operates in 3-500 KHz bands

• Broadband (BB) PLC or BPL operates in 1.8-250 MHz bands

– Internet Protocol (IP)-Based Networks

– Internet Based Virtual Private Networks (Internet VPN)

• Internet VPN technology can provide reliable, secure, and robust alternative to

ensure security and QoS requirement

• Wireless Technologies

– 3G and LTE cellular Networks

– WiMAX Technology

54 IEEE GLOBECOM'11

Wide-Area Networks (WANs)

Core Communication Network and Last Mile Connectivity

55 IEEE GLOBECOM'11

Enabling Technolgies Scope Strength Concern

1. Power Line Communication - Communication Core

Network and Last Mile

Connection

- Complete control over the communication path

with extensive coverage that is solely controlled

by the utility industry

- Provides low cost solution to overlay the

communication network over already available

power lines

- Provides direct route between controllers and

other subsystem to ensure low latency

- Mature technology with many variants available

commercially

- The power line are connected to various equipments such as

motor, power supplies, which can act as noise sources that

eventually degrades the performance of PLC

- The load impedance fluctuation, and electromagnetic

interference causes signal attenuation and distortion, which can

result to failure of communication link

- Lack of standard status and government regulation due to

industry fragmentation result in high interference from other

PLC technology deployed at close range

- Cost of PLC modem are still high

- Coexistence issue from many commercial technologies

2. Internet Protocol (IP)-Based

Networks

- Communication Core

Network and Last Mile

Connection

- IP-based networks have rich convergence

capabilities which can help to connect the overall

systems and subsystems in smart grid

-Can provide QoS and reliable connection using

technologies such as DiffServ and MPLS

Security can be enhanced using technologies

(IPSec)

- In case of master/slave configuration, transmitting IP packets

from slave is not possible, which might increase the data

latency for those applications which requires fast response as in

case of smart grid

- Unless private IP-based network (e.g., Internet VPN) is used,

security remains crucial issue

Wide-Area Networks (WANs)

Core Communication Network and Last Mile Connectivity

56 IEEE GLOBECOM'11

Enabling Technolgies Scope Strength Concern

3. Wireless Communication - Communication Core Network and

Last Mile Connection

- Huge coverage area, potential for low cost

- Packet-Switched Cellular Data has lower cost and

much higher data rates

- WiMAX can support mesh networks for higher

reliability

- Utilities have to depend on these technologies

without any control over them

- Packet switch technologies are not available in all

deployed cellular structures

- Requires connection to network before transmitting

the data and might be problem in case of outage and

emergency

4. Communication and Networking

Middleware

- Communication Core Network - Hybrid network can provide better needs to

specific smart grid application

- Improves Interoperability

- Requires more research to combine technologies to

form network middleware


Standard Activities: Standard Developing Organization (SDO)

• ANSI - American National Standards Institute (www.ansi.org)

• IEC - International Electrotechnical Commission (www.iec.ch)

• IEEE - Institute of Electrical and Electronics Engineers (www.ieee.org)

• ISO - International Organization for Standardization (www.iso.org)

• ITU - International Telecommunication Union (www.itu.int)

57 IEEE GLOBECOM'11

http://www.ansi.org/

http://www.iec.ch/

http://www.ieee.org/

http://www.iso.org/


Standard Activities (1)

58 IEEE GLOBECOM'11

Standards Application Strength Concern

ANSI C12 Suite

ANSI C12.19/IEEE 1377

ANSI C12.22

- Defines utility industry end device data

tables for representing the data produced by

revenue meters.

- Standard protocol for network

communication

- Defines format of data for meter

- Provides transport independent application

level protocol for data exchange with low

overheads between nodes.

- Supports transport of C12.9 table data

- Provide authentication & encrypting the C12.9

data.

- Does not specify protocol to transport it

-Lacks full interoperability due to

specialized local profile

- Requires complexity in implementation in

clients

ANSI/ASHRAE 135/ISO 16484-5

BACnet

- Defines information model & messages as

objects for providing common language for

different proprietary protocols

- Open, mature standard with interoperability

testing developed and maintained by SDOs

- Serves as customer side communication

protocol with relevancy in price, DR/DER &

energy usage

- Object model might be limited to low

level protocols

- Requires structural view & specific profile

to address consumer portals

ANSI /EIA/CEA 709 & CEA 8521

Protocol Suite LONworks

ANSI/CEA 709.1-B

ANSI/CEA 709.2

ANSI/CEA 709.3

ANSI/CEA 709.4

- General purpose LAN protocol for

providing communication over with home &

building automation

- The Control Network

- Power Line Carrier Physical Layer

- Twisted Pair Physical Layer

- Fiber Optic Physical Layer

- Widely used matured protocol

- Specify as one of the data link & physical

layer option for BACnet

- de facto standard controlled by Echelon

with limited support in power industry

- Lack of complex object model to support

function



59 IEEE GLOBECOM'11


ZigBee/ HomePlug Smart Energy Profile Strategic alliance of ZigBee & HomePlug to

provide communication & information model

in HAN

- Interoperable between two distinct HAN technology

- Technology independent

IEC 62056 Device Language Message

Specification (DLMS) &

Companion Specification for

Energy Metering (COSEM)

- Standard representation of metering data

used for accessing and exchanging structured

data models

- Supports object modeling of application data as object

identification system (OBIS) and the Open Systems

Interconnection (OSI) model

- Matured and internationally recognized standard

- Supports variety of media such as PSTN, GSM network,

PLC and recently ZigBee protocols

IEEE 1901 - Broadband communications over

Powerline medium access control

(MAC) and physical layer (PHY)

Protocols for HAN and also access

application

- High speed (>100 Mbps) communication for devices

using frequency below 100 MHz.

- Uses inter-system protocol (ISP), which allow device to

coexist with devices based on ITU-T G.hn standard

- Initiate harmonization and coexistence of PLC with

other technologies

- Has backward compatibility with HomePlug standard

- Short range due to higher

attenuation of the medium as a

result of using broadcast channels

above 80 MHz



60 IEEE GLOBECOM'11


ITU-T G.hn/G.9960 Home Networking

Standard

- In-home networking over power lines,

phone lines, and coaxial

cables

- Designed especially for HAN

- Use single fast Fourier

transform (FFT) OFDM modulation and low-

density parity-check code (LDPC) forward

error correction (FEC) code

- Does not address PLC access application

- Does not support HomePlug standard

ISO/IEC 15045, A residential gateway

model for Home electronic

system

- Defines specification for residential

gateway (RG) that connects HAN to network

domain outside the home basically last mile

connection

- Defines functional requirement &

architecture for RG

- Defines security requirements for connecting

to WANs

- Still under consideration by independent

organization

ISO/IEC 15067-3, Model for an energy

management system for Home

electronic system

- Defines a model for energy management

system that accommodates a range of load

control strategies

- Specifies methods for demand response that

may be implemented by an electric utility or

by a third-party supplier of energy

management services

- Supports various smart appliances


Cognitive Radio [Yu_2011] (1)

• Cognitive radio based communications architecture is

presented for the smart grid

• Cognitive radio allows unlicensed (secondary) user to

access spectrum licensed to licensed (primary) user

– Improve spectrum utilization

– Improve spectrum efficiency

• The proposed architecture is motivated by

– Explosive data volume

– Diverse data traffic

– Need for QoS support

61 IEEE GLOBECOM'11



• Proposed Network Architecture

62 IEEE GLOBECOM'11



63 IEEE GLOBECOM'11

Cognitive area network

Home area network (HAN) Neighborhood area network (NAN)

Wide area network (WAN)

Spectrum band Unlicensed band Licensed band Licensed band

Network topology Centralized/decentralized Centralized Centralized

Network users Smart meters/sensors/acuators HGW

HGWs, NGWs spectrum broker

Featured strategy Cross-layer spectrum sharing Hybrid dynamic spectrum access

Optimal spectrum leasing

Key techniques Access control, power coordination

Guard channel, spectrum handoff

Join spectrum management



• Dynamic Spectrum Sharing in Cognitive HAN

– HGW will connect to the HAN, which in turn will connect to

external networks (e.g., Internet and NAN)

– Within a HAN, the HAN cognitive gateway (HGW) manages the

license-free spectrum bands to provide optimal data rate with low

interference

– HGW enables other devices and sensors to join the network, assigns

channel and network addresses to each device, and coordinates the

communications between the devices within the HAN

64 IEEE GLOBECOM'11



• Cognitive Communications in Neighborhood Area Network

(NAN)

– NAN Cognitive gateway (NGW) connects several HGWs from

multiple HANs together

– Hybrid dynamic spectrum access (H-DSA) is proposed

– Some licensed spectrum bands are leased/bought from a

telecommunication operator, and these bands are used as licensed

access for the HGWs to ensure the QoS of data communications

– The NGW distributes these licensed bands to the HGWs according

to the transmission demand

– However, if licensed spectrum bands are not enough to meet the

demand, unlicensed access is also needed for the HGWs to improve

the capacity and throughput of the NAN

– In unlicensed access, the HGWs and NGW could be considered

secondary users 65

IEEE GLOBECOM'11



• Cognitive Communications in Wide Area Network (WAN)

– In WAN, each NGW is a cognitive node with the capability to

communicate with the control center through frequency space

unused by a licensed primary user

– Control center is connected with cognitive radio base stations

– Spectrum broker controls sharing the spectrum resources among

different NANs to enable coexistence of multiple NANs

– Joint WAN/NAN spectrum management is proposed by minimizing

the maximum dropping probability of data connection in NAN

66 IEEE GLOBECOM'11

Sensor and Actuator Networks (SANETs)

Applications of Data Sensing in Smart Grid

• Power Generation

– WSN called WiMMS unit is deployed in the wind turbine structure [Wang_2007] to

provide information about dynamic behavior of wind turbine and response to loading

– For energy storage, lead-acid batteries will be used, and sensor network can be used

to monitor temperature, voltage, and current

• Power Transmission and Distribution

– Data sensing can be used to monitor substations, transformers, underground lines,

and overhead lines

• Power Consumption

– Smart meter acts as a sensor node and records the electricity consumption (kilo watt

hour [kWh]) and time of use (TOU)

67 IEEE GLOBECOM'11


Requirements for Data Sensing and Communication

• Sensor and Actuator Requirements

– Longer life span

– Reliability and energy-efficiency

– Cost-effectiveness and secured operation

• Data Collection Requirements

– Machine readable format

– Contain the temporal information including the time-stamp

– Identification of location

• Requirements for Communication Networks

– Distributed operation

– Interoperability

– Scalability

– Security

68 IEEE GLOBECOM'11


SANET in Transmission Line Monitoring [Hung_2010]

• The linear sensor network for transmission line is analyzed

• Accelerometer (inclination and cable position and tilt), magnetic field

sensor (current and power quality), strain sensor, and temperature sensor

are considered

69 IEEE GLOBECOM'11


Approaches for Data Sensing

• Phasor Measurement Units

– Phasor measurement units (PMUs) (also referred as synchrophasors) measure the

electrical waves, using a common time source for synchronization

– IEEE Standard C37.118-2005 deals with issues concerning the use of PMUs in

electric power systems

• Compressive Sensing

– Compressive sensing (CS) is proposed which links data acquisition, compression,

dimensionality reduction, and optimization together

– CS senses less and computes more to obtain the useful data

• Decentralized and Cooperative Sensing

– Distributed information processing and control are needed in power system

operations

– For example, distributed state estimation methods have been considered for decades

with the goal of reducing the computational burden at the central control by

distributing the tasks across the system.

70 IEEE GLOBECOM'11


Approaches for Data Communication

• Cooperative Communications

– Cooperative communications refer to the techniques in which multiple nodes help

each other (e.g., in wireless mesh, ad hoc, and sensor networks) to relay or forward

data packets to their destinations

– Cooperative wireless sensor network (IEEE 802.15.4 ZigBee) is used to provide data

transmission in urban-scale smart grid environment [Ullo_2010]

– Secure and reliable collaborative communication scheme for advanced metering

infrastructure (AMI) is introduced [Yan_2011]

– Multihop wireless network is used to connect smart meters with AMI to transfer

meter data to a local collector

• Cognitive Radio

– CR-based wireless sensor network using the 802.15.4 ZigBee standard is proposed in

[Sreesha_2011]

– In the design, a coordinator is used to provide the synchronization and control of data

transmission, while a spectrum sensor is used to support frequency agility so that the

transmission can be adapted based on the wireless channel condition

71 IEEE GLOBECOM'11

Data Communications and Networking in

Smart Grid

• Demand Response Management (DRM)

• Home Energy Management System (HEMS)

• Advanced Metering Infrastructure (AMI)

• Wide-Area Measurement Systems (WAMSs)

72 IEEE GLOBECOM'11

Demand Response Management (DRM)

• DRM is the programs implemented by utility companies to

control the energy consumption at customer side

73 IEEE GLOBECOM'11

Permanent Days Seconds Time

Optimality

Optimized

infrastructure

Optimized

schedule

Temporary

adjustment

Energy

efficiency

TOU

Market

DR

Physical

DR

Spinning

reserve


• Energy efficiency focuses on users and behavioral changes

to achieve more efficient energy usage

– Users buy appliance with energy reduction feature

74 IEEE GLOBECOM'11


• Smart pricing or time of use (TOU)

– Customers (re)arrange their energy consumption to minimize costs

• Market demand response

– Direct load control (DLC): utility or grid operator control energy

consumption of consumers

– Interruptible/curtailable rates: customers has a contract with limited

sheds feature from utility

– Emergency demand response programs: customers voluntarily adjust

energy consumption based on emergency signals (e.g., blackout)

– Demand bidding programs: customers can bid for curtailing at

attractive price

75 IEEE GLOBECOM'11


• Physical demand response

– Grid management and emergency signals (on the utility side)

– Signal if the grid (power lines, transformers, and substations) are in

a reduced performance due to maintenance or failure

• Spinning Reserves (SR)

– Generators are online, synchronized to the grid, that can increase

output immediately in response to a major outage and can reach full

capacity [Hirst_1998]

76 IEEE GLOBECOM'11


• Energy efficiency vs. demand response

77 IEEE GLOBECOM'11

Original load

Time

Energy consumption

Energy efficiency

Demand response

without rebound Demand response

with rebound


Residential load management [Mohsenian-Rad_2010]

• Residential load management programs usually are to

reducing consumption and shifting consumption

• In direct load control (DLC), utility company sets up an

agreement with its customers

• Utility company can manage and control remotely the

operations and energy consumption of certain household

appliances

– Lighting

– Thermal and cooling system

– Refrigerators

– Pumps

78 IEEE GLOBECOM'11


Smart Pricing [Mohsenian-Rad_2010]

• With smart pricing, energy consumers are encouraged to

individually and voluntarily manage their loads

– Reducing their consumption at peak hours

• Critical-peak pricing (CPP), time-of-use pricing (ToUP),

and real-time pricing (RTP) can be used

• For example, in RTP, the price of electricity varies at

different hours of the day

– Prices are usually higher during the afternoon, on hot days in the

summer, and on cold days in the winter

79 IEEE GLOBECOM'11

Home Energy Management System (HEMS)

• HEMS acts as the subset of energy management system

(EMS) and together with smart meter provides a necessary

interface to the HAN for better energy management

80 IEEE GLOBECOM'11

Heating and cooling 49%

Water heater 13%

Refrigerator 5%

Dishwasher 2%

Clothes washer & Dryer 6%

Lighting 10%

Electronics 7% Other 8%

http://www.energystar.gov/


• HEMS (or EMS) sets a certain user limit threshold based on

the information about real-time price-responsive load

management and consumption history (i.e., collected from

smart meter) to control the energy usage of appliances

• HEMS is generally integrated into HAN to offer a channel

for the consumers to interact with the electrical power grid

• HEMS may reside in the smart meter or in an independent

gateway such as residential gateway and network adapters

• HAN contains many electrical appliances (e.g., routers, TV,

AC, computers, etc) which provide different services, e.g.,

wireless access, VoIP calls, ambient temperature control

• These services can be controlled by using different power

control elements (PCEs) such as Ethernet switch, PSTN,

and DSL modem 81

IEEE GLOBECOM'11


• Example: GE Demand Reduction Approach

82 IEEE GLOBECOM'11

Price Event Signal to

Smart Appliance

Smart Appliance will indicate

to consumer

Price Event has occurred

Smart Appliance will

recommend to delay

start

Over

Ride?

Over

Ride?

Run Normal operating mode

Consumer Choice

Initiate delayed start function

Initiate peak reduction mode

Data communications Electric supply from

Transmission

HAN

Gateway

In-Home

Display

DAU/NAN

Sensors

Light

Temperature

Voltage

Wired/Wireless

Connection

(e.g., Zigbee,

BACnet

Smart

Devices

Smart

Devices

Smart

Devices

Actuators

ActuatorsActuators

(Smart Meter

or

Dedicated

in-home

Gateway)


Machine-to-Machine Communications [Niyato2011]

• Network design issue of M2M communications for a home

energy management system (HEMS) is considered

• The network architecture for HEMS to collect status and

power consumption demand from home appliances is

introduced

• Optimal HEMS traffic concentration is presented and

formulated as the optimal cluster formation

83 IEEE GLOBECOM'11


Machine-to-Machine Communications

• Network model

84 IEEE GLOBECOM'11

Concentrator

Base stationInternet

backhaul

Control

center

Service area with wide area network (WAN)

Neighborhood area network (NAN)

Home area

network (HAN)

Smart meter


Machine-to-Machine Communications

• Optimal cluster is determined

• The average cost per node under different packet generation

rates is shown

85 IEEE GLOBECOM'11

0.4 0.6 0.8 1 1.2 1.4 1.6 1.80

2

4

6

8

10

12

14

16

18

20

Packet generation rate (packets/minute)

Avera

ge c

ost

per

node

Optimal formation

Fixed formation

Cluster size = 10

Cluster size = 5

Cluster size = 1

Cluster size = 2

Cluster size = 4

Cluster size = 3

Advanced Metering Infrastructure

• AMI acts as the gateway for access enabling the

bidirectional flow of information and power in support of

distributed energy resource (DER) management or

distributed generation (DG) and consumer participation

• AMI will provide near real-time consumption data including

fault and outage to the utility control center

• AMI supports time-based and dynamic tariffs such as Time

of Use (TOU), Real-Time Pricing (RTP), and Critical Peak

Pricing (CPP)

• AMI consists of several different components

– Smart meters and data aggregator units (DAUs))

– Hierarchical area networks (e.g., home-area networks (HANs) and

neighborhood-area networks (NANs), and wide-area networks

(WANs))

86 IEEE GLOBECOM'11


• Comparison

87 IEEE GLOBECOM'11

Manual/Automatic Meter Reading (AMR)

AMI

Pricing Fixed price and measure total consumption only

Total consumption Time-of-use Critical peak pricing Real-time pricing

Other demand response None Load control Demand bidding Demand reserves Critical peak rebates

Customer feedback Monthly bill Monthly bill Monthly detailed report Web display In-home display

Customer bill savings Turn off appliances manually Turn off appliances Shift appliances off peak Manual or automatic control

Outages Customer phone calls Automatic detection Verification of restoration at individual home level

Distribution operations Use engineering models Dynamic, real-time operations


Benefit of AMI [Liu_2010]

• Fault Location, Isolation and Service Restoration (FLISR)

– AMI will be able to automatically report loss of power, and the information can be

used to assist locating the fault location

• Emergency Load Shedding

– AMI helps to shed large amounts of load very quickly (within seconds) to avoid

power system instability and loss of system integrity (e.g., during bulk power grid

emergencies)

• Distribution System Planning and Analysis

– AMI provides accurately metered data for all customers on the feeder from billing

records, and this information will enable the system to prepare much more accurate

short term load forecast

• Continuous Condition Monitoring

• Equipment and System Performance Forecasting

• Automated “Triggering” for Maintenance and Work Assignments

• Substation and Line Monitoring

88 IEEE GLOBECOM'11


Wireless Broadband Architecture [Mao_011] and Key Design

Issues

• Address Depletion

– For AMI, a very large number of new subscriber devices, i.e. smart meters, will need

address for communications

• Traffic Scheduling

– Critical alarm indication data should be reported immediately and not be queued until

the next scheduled connected period

• Congestion Control

– A very large numbers of SM give rise to potential “traffic burst” scenarios which

arise when large numbers of devices are simultaneously (or near simultaneous)

reporting or reacting to a common event

89 IEEE GLOBECOM'11


Service-Oriented AMI [Chen_2010]

• Service-oriented approach to AMI aiming at solving the

intercommunication problem and meanwhile providing a trust and

secure environment for smart grids

– System integration and cooperation are done through service composition.

– Generic service interfacing method is designed to develop standardized

– services for heterogeneous power systems

– Role-based access control mechanism is used to guarantee secure access

90 IEEE GLOBECOM'11


Reliability Analysis

• Reliability analysis of the wireless communications system

in the smart grid can be performed

• Availability performance can be obtained given the random

failure of the system devices

• Availability measure can be used to calculate the cost of

power-demand estimation error and damage of power

distribution equipment if its failure cannot be reported

• Redundancy design approaches can be developed to

minimize the cost of failure as well as the cost of

deployment of the wireless communications system in the

smart grid

91 IEEE GLOBECOM'11



92 IEEE GLOBECOM'11

Neighborhood area

network (NAN)

Home area

network (HAN)

HAN gateway and

smart meter

NAN gateway Data aggregator

unit (DAU)

Meter data-management system (MDMS)

NAN with gateway

redundancy

Power distribution

equipment


Reliability Analysis: Operation of a power system

93 IEEE GLOBECOM'11

Smart meter estimates power

demand in the next period

(e.g., using power scheduling)

HAN gateway sends power

demand collected from smart meter

to the corresponding NAN gateway to

forward to DAU and subsequently MDMS

MDMS buys additional

power supply in

economic dispatch stage Power demand is added into

amount of power to be supplied

Power demand of each house

is received by MDMS?

MDMS uses mean power

consumption of that house to

compute amount of

power to be supplied

MDMS buys power supply

in unit commitment stage

MDMS checks if power

supply is enough or not?

No

Yes

Yes

No


Reliability Analysis: Operation of a power system • If the power demand of any house is not received by the MDMS (e.g.,

due to failure of the HAN gateway, the NAN gateway, or the DAU), the

MDMS uses historical data to compute the aggregated power demand

• x% of mean power-consumption1 of those houses is used as the

estimated demand

94 IEEE GLOBECOM'11

Power consumption (kWh)

Pro

bab

ility

dis

trib

ution

Estimated power demand (i.e., reserved

power from unit commitment stage) for

x=100% of mean

Cost of under-reservation

Cost of over-reservation

0


Reliability Analysis: Availability • Availability of a component/device/system is the probability that the

component/device/system has not failed or repaired and it can operate

normally

• Uptime is also known as the mean time between failure (MTBF)

• Downtime is known as the mean time between repair (MTBR)

• Failure rate can be obtained a 1-Availability

95 IEEE GLOBECOM'11


Reliability Analysis: Availability • Dependence diagram (DD) determines the contribution of each

component to the availability of the system

• The components can be connected in parallel and/or series

96 IEEE GLOBECOM'11

Radio interface Single board computer Adaptor

Power

Software

Power

Control unit Radio interface Power Metering engine

Dependence diagram of smart meter and home area network gateway

Dependence diagram of neighborhood area network gateway

Node B Radio network

controller (RNC)

Service gateway

support node (SGSN)

GPRS gateway support

node (GGSN)

Dependence diagram of UMTS network


Reliability Analysis: Availability • HAN gateway and a smart meter can be integrated into a single device.

The availability of a HAN gateway is computed from AHAN =

availability of metering engine × availability of control unit ×

availability of power module × availability of radio interface

• Availability of a NAN gateway is computed from ANAN = availability of

radio interface × availability of single board computer × availability of

adaptor × availability of software × (1 − (1−availability of power

module)2)

• 3G cellular base station is assumed to have the DAU functionality

whose availability is computed from: ADAU = availability of node B ×

availability of radio network controller (RNC) × availability of service

gateway support node (SGSN) × availability of GPRS gateway support

node (GGSN)

97 IEEE GLOBECOM'11


Reliability Analysis: Cost of Network Unavailability • Cost of demand-estimation error of individual house i whose connection

to the MDMS is unavailable can be obtained from

• Ei = x/100 × Meani is the power supply reserved in the unit

commitment stage

• Meani is the mean power-consumption of house I

• Maxi is the maximum power-consumption

• fA(i)(a) is the PDF of actual power demand a

• puc and ped denote the power prices in the unit commitment and in the

economic dispatch stages, respectively

98 IEEE GLOBECOM'11


Reliability Analysis: Cost of Network Unavailability

• Number of houses

• Number of redundant NAN gateways

99 IEEE GLOBECOM'11

20 40 60 80 100 120 140 160 180 2000

10

20

30

40

50

60

70

Number of houses in NAN

Cost

of

dem

and e

stim

ation e

rror

per

month

($)

Failure rate of HAN gateway = 2 days in 1 years




0 1 2 3 4 5 6 7 8 9 10160

180

200

220

240

260

280

300

Number of redundant NAN gateways

Avera

ge t

ota

l cost

per

month

($)

Failure rate of NAN gateway = 2 day in 2 years



Wide-Area Measurement Systems (WAMSs)

• WAMS is used to conduct real time monitoring and control

in dynamic power system states

• WAMS uses a synchronized phasor measurement unit

(PMU) to guarantee for security and stability of power

systems

• WAMS is typically composed of PMUs, phasor data

concentrator (PDC), control center (CC), as well as the

high-speed data communication networks

100 IEEE GLOBECOM'11


Applications [Naduvathuparambil_2002]

• State estimation: PMUs can measure and relay information

on a continuous basis to the control centers, and control

center will generate a state vector of system dynamics

• Instability prediction: Synchronized phasor measurements

can enable real-time stability analysis and instability

prediction

• Improved control of power systems: Controllers (e.g.,

variable series capacitors [VSC], universal power flow

controllers [UPFCs] and power system stabilizers) can

receive feedback from control center to regulate the grid



Data Communication

• Telephone lines

– Easy to set up and economical to use, but low speed

• Fiber-optic cables

– Immunity to RF & atmospheric interference

– Large bandwidth

• Satellites: low-earth orbiting (LEO)

– Large coverage area,

– High cost, narrow bandwidth, and large delays

• Power lines

– Uses the medium and low voltage electric supply grid for transmission of

data and voice

• Microwave links

– Easy to set up and are highly reliable

– Signal fading and multipath propagation



Centralized WAMS [Shahraeini_2011]

• All data resources send data to control center (CC)

• After processing the received data, appropriate decisions are

made and related commands are sent back to controllable

devices



Decentralized WAMS [Shahraeini_2011]

• System is divided into multiple areas

• Each area has its own are control center area (ACC)

• In each area, ACC processes the acquired data and perform

control

• For the control of a system, ACCs share information among

each other through communication systems



• [Bruce_1998], [Xie_2002], [Wang_2010]

– Synchronized phasor measurement unit (PMU)

– Phasor data concentrator (PDC)

– Ring interface unit (RIU)

– Control center (CC)


Fault tree analysis of WAMS



• Availability is calculated from

• Ai is the availability of the th PMUs-PDC working group

• AijPMU is the availability of PMU j in PMUs-PDC working group i

• Mi is the number of PMUs in group I

• AiRN is availability of regional communication network

• AiPDC is availability of PDC device



Cyber Security for Smart Grid

• Introduction

• Why do we need cyber security

• Adversaries

• Threats

• Impacts

• How to achieve cyber security

• Survey some solutions

IEEE GLOBECOM'11

107

Transmission

TOP1 – Operational Information

Distribution

DIST1 - Operational Information

DISTx – Operational Information

Customers Generation

GEN1 - Operational Information

GENx - Operational Information

Current Electric Grid – Islands of Technology

TOPx – Operational Information

IEEE GLOBECOM'11

108

Convergence of Enterprise & Operations IT

Enterprise Systems

Web Applications

Control Systems

Protection Systems

Information Technology Operations Technology

AMI

DSM

OMS

GIS

Smart Grid Technology

Integration counters key security principals of isolation and segregation

Cyber Secure

Integration counters key security principals of isolation and segregation

Convergence of Enterprise & Operations IT

IEEE GLOBECOM'11

109

Smart Grid – Connectivity with Security

Transmission Distribution Customers Generation

System

Operators

Conservation

Authorities

End-to-End Communications, Intelligence, and Defense-in-Depth Security

AMI DSM

IEEE GLOBECOM'11

110

Why do we need cyber security ?

• Network security is a priority and not a add on for smart

grids

• Protecting control center alone - not enough

• Remote access to devices

• QoS requirement from security system

• Safety (line worker public and equipment)

• Reliability and availability

111

Drivers

Increasing Number

Of Systems and

Size of Code Base

Control Systems

Not Designed with

Security in Mind

Increasing Use of

COTS Hardware

and Software

New Customer

Touch Points into

Utilities

New 2-Way

Systems

(e.g. AMI, DSM)

Increasing

Interconnection

and Integration

Increased Attack Surface

Increased Risk to Operations

112

Threats-I

Example from 2006 SANS SCADA Security Summit, INL

1. Hacker sends an e-mail with malware

2. E-mail recipient opens the e-mail and the

malware gets installed quietly

3. Using the information that malware gets,

hacker is able to take control of the e-mail

recipient’s PC!

4. Hacker performs an ARP (Address Resolution

Protocol) Scan

5. Once the Slave Database is found, hacker sends

an SQL EXEC command

6. Performs another ARP Scan

7. Takes control of RTU

Internet

Admin

Acct

Opens

Email with

Malware

Admin

Send e-mail

with malware

Slave

Database

Operator

Operator

Master

DB

RTU

Perform

ARP Scan

SQL

EXEC

Perform

ARP Scan

113

Example from AMRA

Webinar, Nov ’06

“The Active Attacker”

Threats-II

U N I V E R S I T YU N I V E R S I T Y

AMI WAN AMI WAN AMI WAN

Communications

Network

(WAN)

Communications

Network

(WAN)

Data Management

Systems

(MDM/R)

Retailers

3rd Parties

AMCC

(Advanced Metering

Control Computer)

Attacker

Cyber

Penetration

Attacker

Controls the

Head End

Attacker

Performs

Remote

Disconnect

114

Impacts-I

Meter

EMS

AMI

Network

HAN

Energy

Service

Provider

Wide Effect, High Impact on

the Grid, Attacker may be

Remote

Local Effect, Narrow Impact,

Attacker Needs to be Local

Energy Consumption Data

Demand Response Trigger

Utility

Back Office

Direct Energy Information Access from Meter and Local Control in

Customer Premises has Lowest Risk

The Impact of a Security Breach*

* Does not represent the difficulty or ease of executing the breach.

115

Impact-II

Threat Attacker

Location

Impact

Spread

Impact

Effect

AMI Network

Compromised Remote Wide Network

Stability

DR Manipulated in

―Cloud‖*

Remote Wide Network

Stability

Customer Privacy

Breached in ―Cloud‖* Remote Wide Loss of Privacy

HAN Compromised Local Narrow

Local

Nuisance

* “Cloud” refers to both a Utility Back Office and Energy Service Provider

IEEE GLOBECOM'11

116

Cost of Power

Disturbances:

$25 - $188 billion

per year

~$6 billion lost

due to 8/14/03

blackout

Northeast Blackout – August 14, 2003

• Affected 55 million

people

• $6 billion lost

• Per year $135

billions lost for

power interruption

http://en.wikipedia.org/wiki/Northeast_Blackout_of_2003

IEEE GLOBECOM'11

117

http://en.wikipedia.org/wiki/Northeast_Blackout_of_2003

Adversaries

• Hostile States

• Hackers

• Terrorist /Cyber terrorists

• Organized crime

• Other criminal elements

• Industrial competitors

• Disgruntled employees

• Careless and poorly trained employees

IEEE GLOBECOM'11

118

Classification of attacks

• Component based attacks

• Protocol based attack

IEEE GLOBECOM'11

119

COMPONENT BASED ATTACK -STUXNET

• Specifically programmed to attack SCADA and could reprogram

PLC‟s

• Zero day attack

• Highly complex

• 0.5 Mb file transferred able to multiply

• Targets- Iran nuclear plants ,Process plants in Germany and ISRO

India

Source: wikipedia

IEEE GLOBECOM'11

120

COMPONENT BASED ATTACK - SCADA attacks

• Internal attacks Employee

Contractor

• External attacks Non specific- malware , hackers

Targeted Special knowledge – former insider

No special knowledge –hacker terrorist

Natural disaster

Manmade disasters

IEEE GLOBECOM'11

121

SCADA – vulnerability points

• Unused telephone line – war dialing

• Use of removable media – stuxnet

• Infected Bluetooth enabled devices

• Wi-Fi enabled computer that has Ethernet connection to scada

system

• Insufficiently secure Wi-Fi

• Corporate LAN /WAN

• Corporate web server email servers internet gateways

IEEE GLOBECOM'11

122

SCADA-CYBER ATTACKS

• Web servers or SQL attacks

• Email attacks

• Zombie recruitment

• DDOS attacks

IEEE GLOBECOM'11

123

Protocol based attacks

• All protocols runs on top of IP protocol and IP protocol has its own set of

weakness

• DNP3 implements TLS and SSL encryption which is weak

• The protocol is vulnerable to out-of-order, unexpected or incorrectly formatted

packets

• A significant weakness for IEC 61850 is that it maps to MMS (Manufacturing

message specification)as the communications platform, which itself has a wide

range of potential vulnerabilities

IEEE GLOBECOM'11

124

Challenges

• The challenge is complex and continuously changing

• Legacy systems need to be protected

• Number and geographic location of end points

• Relationship to physical security

• Systems are 7x24 and critical

• The human element / social engineering

IEEE GLOBECOM'11

125

Challenges („cont.)

• Scale

• Legacy devices

• Field location

• Culture of security through obscurity

• Evolving standards and regulations

IEEE GLOBECOM'11

126

How to achieve cyber security?

• Security by obscurity

• Trust no one

• Layered security framework

• Efficient firewall

• Intrusion detection

• Self healing security system

IEEE GLOBECOM'11

127

Types of Cyber Security Solutions

• Reactive vs. Proactive – Reactive

o Incident response plan

o Applied for general purpose computers more

– Proactive Security for embedded computers

• High assurance boot

• Secure software validation

• Secure association termination if found

infected

• Device assentation

IEEE GLOBECOM'11

128

Solution - Incidence response plan

IEEE GLOBECOM'11

Attack

Prevention Services

Containment Services

Detection &

Notification Services

Recovery &

Restoration Services

129

Solution - Defense in Depth

• Perimeter Protection

– Firewall, IPS, VPN, AV

– Host IDS, Host AV

– DMZ

– Physical Security

• Interior Security

– Firewall, IDS, VPN, AV

– Host IDS, Host AV

– IEEE P1711 (Serial Connections)

– NAC

– Scanning

• Monitoring

• Management

• Processes

IDS Intrusion Detection System

IPS Intrusion Prevention System

DMZ DeMilitarized Zone

VPN Virtual Private Network (encrypted)

AV Anti-Virus (anti-malware)

NAC Network Admission Control

IEEE GLOBECOM'11

130

Solution –Control Network

Internet

Enterprise Network

Control Network

Field Site Field Site Field Site

Partner

Site

VPN

VPN

FW

FW

IPS

IDS

Scan

AV

FW IPS

P1711

FW

AV Host IPS Host AV Proxy

Host IDS Host AV

IDS Scan

NAC

NAC

• Defense in Depth

• Access Control

• Secure connections

• Link to Physical

• Security Management

• Apply same approach

to other Smart Grid

elements

Key Points:

131

Solution – Key management

• Issue of key management – Scale

• PKI with trusted computing elements- considerable

amount of security

• Embedded vs. general-purpose computing

IEEE GLOBECOM'11

132

PKI Infrastructure

133

Issues with PKI

• Updating the keys

• Parameter generation

• Key distribution

• Staffing for key management

IEEE GLOBECOM'11

134

Solution – Attack trees

IEEE GLOBECOM'11

135

Calculation of cyber security conditions (omega)

IEEE GLOBECOM'11

136

Rules for Conditions 1, 2, and 3

Conditions Rules

Condition 1 The system is free of intrusion attempt that

is concluded from the electronic evidences

in the system

Condition 2 At least one or more countermeasures are

implemented to protect an attack leaf.

Condition 3 At least one or more password policies are

enforced corresponding to each attack leaf.

Weighing factor for password policy

IEEE GLOBECOM'11

137

Calculations of vulnerability index

• Leaf VI : max( total countermeasures implemented

/total countermeasures available x ω , ω x weighing

factor of password policy)

• Scenario vulnerability index : Product of its leaf

vulnerability indices

• System vulnerability index is the max of all

scenario vulnerabilities indices

IEEE GLOBECOM'11

138

State estimation attack - introduction

• State estimation is to determine the optimal estimate for the complex voltages at each bus based on real-time analog measurements. – The state typically refers to bus voltage magnitudes

and phase angles

• Bad data processing is to detect measurement errors, and identify and eliminate them if possible. – It is effective against random noises, but

– It lacks the ability to detect intentionally coordinated bad data

• That conforms to the network topology and physical laws

IEEE GLOBECOM'11

139

State estimation attack - 1

• Attack on state estimation [Giani_2011]

– By compromising some line meters, sending wrong information about voltage / current status

• Force the energy management system to make wrong balancing operations that causes outage

– Main characters of the attack

• Sparse attacks are common (unobservable attacks)

– Large number of coordinated attacks can be detected by a bad data detection algorithm

• [Giani_2011] A. Giani, E. Bitary, M. Garciay, M. McQueenz, P. Khargonekarx, and K. Poolla, “Smart

Grid Data Integrity Attacks: Characterizations and Countermeasures”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

140

Main contributions

• An efficient detection algorithm for – Case I : the attackers compromise

• Two power injection meters coordinately

• Arbitrary number of line meters

• The algorithm require O(n2×m) flops – n is the number of buses , m is the number of line meters

– Case II • Limited number of coordinated meters for attack (i.e., 3, 4, or 5)

• All lines are metered

• The algorithm requires O(n2) flops

• Countermeasures – Using known-secure PMUs for counteracting the attacks

– Demonstrate that p+1 PMUs are enough to neutralize a collection of p cyberattacks

• The positions of PMUs need to be carefully chosen

IEEE GLOBECOM'11

141


• Study the vulnerability of the state estimator to attacks performed against the communication infrastructure [Vukovic_2011]

• Use the security metrics defined by them to show – how various network and application layer mitigation

strategies can be used • to decrease the vulnerability of the state estimator

• Background – An attacker that wants to change the measurement on one

substation might have to change several other measurements • To avoid a bad data detection (BDD) alarm

• [Vukovic_2011] O. Vukovic, K-C Sou, G. Dan, and H. Sandberg, “Network-layer Protection Schemes

against Stealth Attacks on State Estimators in Power Systems”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

142

Main ideas

• Substation is the weak point – Measurement data are usually collected through

substations

– An attacker can access and modify all data that traverses a substation

– The authors proposed to assess the importance of each substation with respect to state estimation

• Security metrics – Substation attack impact

• The number of measurements on which an attack can perform a stealth attack

– Measurement of attack cost • Minimum number of substations that have to be attacked in

order to perform attack against the measurement

IEEE GLOBECOM'11

143

Main contributions

• Protective methods

– Network layer solutions

• Single-route routing vs. Multi-path routing

• Modify single-route path to decrease the vulnerability

of the system

• Multi-path routing could reduce the maximum attack

impact by 50%

– Application layer solutions

• Data authentication increases the attack cost

– The solutions are very realistic

IEEE GLOBECOM'11

144


• This paper introduced a procedure that aims to achieve network-wide optimal attack detection and state estimation [Tajer_2011]

• The procedure is distributed – Different controlling agents distributed across the

network carry out the attack detection and system recovery tasks through

• local processing and message passing, and

• An iterative process

– Distributed state estimation method can reduce the computational burden on the centralized control system

• Using a decompose-merge approach

• [Tajer_2011] A. Tajer, S. Kar, V. Poor, and S. Cui, “Distributed Joint Cyber Attack Detection and State

Recovery in Smart Grids”, Proceedings of IEEE Globecom 2011.

IEEE GLOBECOM'11

145

Main contributions

• Reliable detection + reliable estimate of the

false injected data

– Means that the system can still obtain relatively

accurate estimation of the data in spite of attacks

– Different from works that avoid data to be

compromised

– Used an information theoretic method

IEEE GLOBECOM'11

146


• This paper [Esmalifalak_2011] demonstrate an attack method that – Inject false data with low detectability

– Without knowledge of the network topology

– Makes the inference from the correlations of line measurements

• But assume that the attackers can break into the SCADA system

• Main contributions – Demonstrate that an attacker can estimate both the system

topology and power states just by observing the power flow measurements

– Independent component analysis (ICA) is used • to infer the linear structure of the power flow measurements

• [Esmalifalak_2011] M. Esmalifalak, H. Nguyen, R. Zheng, and Z. Han, “Stealth False Data Injection using Independent Component

Analysis in Smart Grid”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

147

Performance evaluation

• The authors demonstrated that

– The ICA based attack is almost unobserserable

– The random attack is easy to be detected • Real – no attack

• Estimated – ICA based attack

IEEE GLOBECOM'11

148

General security technique used in smart grid

IEEE GLOBECOM'11

149

Message authentication code aggregation

• Message Authentication Code (MAC) is used to authenticate each message [Kolesnikov_2011] – To prevent en route accidental and malicious data

corruption

– Aggregate MAC is often used • Since the communication channel capacity is often small,

and

• The data size is short compared to the MAC code

– The aggregate MAC is not resilient to denial-of-service (DOS) attacks

• [Kolesnikov_2011] V. Kolesnikov, W. Lee, and J. Hong, “MAC Aggregation Resilient to DoS

Attacks”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

150

Main contributions

• The authors proposed a new authentication mechanism for the wireless sensor data – Securely combine authentication tags computed by

sensors • So that the aggregate tag is much shorter than the

concatenation of the constituent tags, but

• Provides same strong security guarantees

– Resilient to denial-of-service (DOS) attacks • A DoS attacker will only be able to disrupt a portion of the

data – Only the data he relays

• His point of insertion can be estimated based on which part of aggregate MAC is corrupted.

IEEE GLOBECOM'11

151

Secure energy routing

• The authors of [Zhu_2011] developed a novel secure energy routing mechanism – for securely and optimally sharing renewable

energy in smart microgrids

– It can detects most internal attacks by using message redundancy

• Spoofed route signaling

• Fabricated routing messages

• [Zhu_2011] T. Zhu, S. Xiao, Y. Ping, D. Towsley, and W. Gong, “A Secure

Energy Routing Mechanism for Sharing Renewable Energy in Smart Microgrid”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

152

Intrusion detection systems for home area

networks • This paper [Jokar _2011] presents a layered specification-based

intrusion detection system (IDS) – Designed to target ZigBee technology

– Addressed the physical and MAC layer • Normal behavior of the network is defined through selected specifications

extracted from the IEEE 802.15.4 standard

• Deviations from the defined normal behavior is viewed as a sign of malicious activities

• The performance analysis demonstrated that the designed IDS provides a good detection capability against known attacks – The same is expected for unknown attacks

• Since the design of the IDS is based on anomalous event detection

• [Jokar _2011] P. Jokar, H. Nicanfar, V. Leung, “Specification-based Intrusion Detection for Home

Area Networks in Smart Grids”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

153

Privacy of electricity usage information

IEEE GLOBECOM'11

154

Privacy-preserving authentication

• Privacy requirement: to preserve the privacy of the consumers, the electric usage information is hidden from the substations [Chim_2011] – But it should be known by the control center

• Pseudo identity is used

• Authentication requirement on each smart meter – To ensure requests are sent from valid users

• The authentication process is made very efficient by means of Hash-based Message Authentication Code (HMAC) – The overhead is only 20 bytes per request message

• Under attack, the substation allows 6 times more valid messages to reach the control center – when compared to the case without any verification

• [Chim_2011] T. Chim, S. Yiu, L. Hui, and V. Li, “PASS: Privacy-preserving Authentication Scheme

for Smart Grid Network”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

155

Privacy-utility tradeoff

• Existing privacy preservation solutions for user‟s electricity usage data have also not quantified the loss of benefit (utility) of data dissemination [Rajagopalan_2011]

• Using tools from information theory, a new framework is presented that abstracts both the privacy and the utility requirements of smart meter data.

• For a stationary Gaussian Markov model of the electricity load, it is shown that the optimal utility-and-privacy preserving solution requires filtering out frequency components that are low in power – this approach encompass most of the proposed privacy approaches

• [Rajagopalan_2011] S. Rajagopalan, L. Sankar, S. Mohajer, and V. Poor, “Smart Meter

Privacy: A Utility-Privacy Framework”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

156

Cooperative state estimation for preserving privacy

• This paper [Kim_2011] presents a cooperative state estimation technique that protects the privacy of users‟ daily activities. – By exploiting the kernel of an electric grid configuration

matrix

– Obfuscate the privacy-prone data without compromising the performance of state estimation

• The power consumption measurement is well obfuscated such that the consumers do not fully disclose their private behavioral information in the first place, and

• the obfuscated data retain the necessary information such that the state vector can be accurately estimated from the perturbed measurement

• [Kim_2011] Y. Kim, E. Ngai, and M. Srivastava, “Cooperative State Estimation for Preserving Privacy of User Behaviors in Smart Grid”, Proceedings of IEEE SmartGridComm 2011.

IEEE GLOBECOM'11

157

Summary on Cyber Security for Smart Grid

• Different security constraints that makes securing smart grids a difficult

problem

• Several highly efficient adversaries

• Use existing protocols like IP with known vulnerabilities and work

around to using new protocols with unknown vulnerabilities

• Use of layered security architecture and attack tree‟s for efficient

security and risk assessment

IEEE GLOBECOM'11

158

Field Trials and Case Studies for Smart Grid

Communication Infrastructures Smart Power Grid

• SDC

Smart Renewable

• W2B

Smart Electricity Service

• S&C‟s CES

Smart Transportation

• PHEV/EV

Smart Consumer

• MDM

• MYPOWER

IEEE GLOBECOM'11

159

SmartGridCityTM – Boulder, Colorado

“The fundamental component for making the smart grid work will be a robust and dynamic communications network; providing the utility the ability for real-time, two-way communications throughout the grid and enabling interaction with each component from fuel source to end use” (Xcel Smart Grid White Paper)

Collaborating to Build the Next Generation Utility

160

Status of SGC

City - City of Boulder - 100,000 people, 50,000

homes

Smart Meters - 14,398 as of 1/28/09

Premises - 16,616 BPL enabled homes as of

1/28/09

Telecom Fiber - 120 miles planned by June 2009

Delivery Dates - build out complete by 6/30/2009

Systems - plug and play demand and generation

response (in process)

161

Demand Management

• Reduce spinning reserves

• Generation following (not demand response)

• Availability-based pricing

• Automated generation dispatch

Renewables Management

• Align demand to availability

• Manage intermittency

• Opt for type of energy use

• Supply-based pricing

Asset Management

• Improve field efficiency

• Real-time asset status & control

• Expanded reliability

• Extended asset life

Premise Management

• Automated device response control

• Real-time pricing (device-level)

• New services and products

• Enable customer choice

SMARTGRIDCITY – Key Values

162

SmartGridCity-Objectives

Xcel Objective Measurement CURRENT Smart Grid Impact

Improving Customer Satisfaction by

reducing customer minutes out of service

Reduce SAIDI by 10% Distribution Automation

Analysis & Reporting of:

Incipient transformer failure

Secondary neutral failure

Voltage exceptions

Transformer Overload

Underground remote fault detection

Outage notification & restoration

Empowering Customers to Reduce

Electricity Usage

Decrease usage by 2.5% 2-way thermostat control

Demand response portals

Meter consumption reporting

Reduce Service and Billing Expense,

Increase Revenue Assurance

Up to 50% annually Call center meter pings

Automated meter reading

Proactive maintenance (reduced O&M)

Decrease System Losses Reduce CO2 emissions up to 500,000 tons

annually

System Optimization

Conservation voltage reduction

Volt/Var Control

Phase Load Balancing

Asset Optimization Reduce capital investment and

distribution/substation maintenance up to $32

mil annually

Substation monitoring

Targeted asset replacement (system

reports)

Develop a Smart Grid City Consortium

Framework

Seamless integration of applications and

business process

Open GridTM Platform

Develop a Regulatory Framework to

Recover Smart Grid Investment

TBD Smart Grid Value Model

163

Smart Grid Operational Impact

Trees

8%

Xformer

25%

Xformer

Lead/Connection

10%

Secondary Brkr

Tripped

30%

Secondary

2%Secondary/Xformer

9%

Arrestor

1%

Xformer Tap

1%

Capacitor

2%

Sub LV Bus

Voltage

6%

Secondary Neutral

Connection

6%

Examples of items detected by a Smart Grid:

Smart Grid Solutions:

● 24x7 real-time distribution

network monitoring in use

● Dispatching work crews to repair

problems detected by CURRENT

Smart GridTM

● Underground fault detection

installed

● Successful distribution

automation switching trial

94% of the incidents detected avoided customer complaints

54% of the incidents detected avoided outages

SmartGridCityTM

Consortium

164

Smart Renewable Grid Balancing

Renewable Integration

Outage Support

Capital Cost Avoidance

Emissions Savings

Transmission Support

Firm Renewable Power Pricing

Graph from John P.

Benner, Manager, PV

Industry Partnerships,

National Renewable

Energy Laboratory,

303-384-6496

165

Wind 2 Battery (W2B) Project Description

• 1 MW NaS Battery

System • Can deliver 1 MW for 7 hrs

• Power Conditioning Equipment

• Wind farm/grid interconnection

• Local and remote data and

communication equipment

• Two Phases of Study • Understand how system could

optimize wind farm economies

• Understand how system could

optimize utility integration of

wind resources

166

Smart Electricity Service

167

COMMUNITY ENERGY STORAGE (CES)

Growth of Customer-Owned

DG (solar)

• Availability?

• Reliability?

• Safety?

• Dispatch?

“Net Zero” or “Near Zero”

Customers and Areas

• Own their generation (solar or

wind)

• Grid-Independent (with storage)

• Third-party storage service

could take them off the utility

grid

168

S&C‟S CES PROJECT-HARDWARE OVERVIEW

CES is a small distributed energy storage unit connected to the secondary of

transformers serving a few houses or small commercial loads

Key Parameters Value

Power (active and reactive) 25 kVA

Energy 25-75 kWh

Voltage - Secondary 240 / 120V

Battery - PHEV Li-Ion

Round Trip AC Energy

Efficiency

> 85%

25 KVA

169

S&C‟s CES Project-A “Virtual” Substation Battery

Communication and Control Layout for CES

CES Control Hub

Power Lines Communication and Control Links

CES CES CES CES

CES is Operated as a Fleet providing Multi-MW, Multi-hour Storage

Grid Benefits:

• Load Leveling at substation

• Power Factor Correction

• Ancillary services

Local Benefits:

• Backup power

• Voltage correction

• Renewable Integration

Integration

Platform

Utility Dispatch

Center/ SCADA

Substation

170

SpeedNET Network Management

171

SpeedNet™ Radios:

A Leading Solution for Self-Healing Applications

Features Benefits

Self healing—peer-to-peer

mesh network

Reliable performance even if a

communication point is lost

Multi-level security Improved performance, less susceptible

to interference

Low latency High speed communications—shorter

restoration times

Assignable messaging

priority

Effectively serves both AMI backhaul

and DA applications

172

Smart Transportation

Electric Drive &

Electronic Components

Vehicle Stability

Control

Functional Safety

& durability of the FEV Communication

Architecture for Energy,

Communication &

thermal management,

Energy / Power Storage

Systems

Vehicle 2 Grid

Interface

Integration of the FEV in

cooperative transport

Infrastructure

173

Electric Drive Vehicles

• Until now, base growth of 1% per year for USA system – At 25% of US vehicle fleet is “only” 2% of total MW*hr (but

billions of $ in generation and distribution costs)

– On distribution a car‟s 6 KW connection for an average home‟s

peak usage of 3 KW is +200% & is very significant

http://www.ornl.gov/info/ornlreview/v41_1_08/regional_phev_analysis.pdf

174

2007 Xcel Energy / NREL PHEV Study

Scenarios Production Cost Capacity Cost Avoided Gasoline Emissions Distribution Impacts

Do Nothing Good Worse* Good Better Worse*

Delay to 10pm Better Best Good Good Best

Optimized to Off-peak Best Best Good Worse Best

Opportunity Charging Worse Worse* Best Best Worse*

• For any utility:

Time of charging matters…

Coincident peak loading matters…

Tailpipe versus upstream emissions matter…

* Could be mitigated with control technology / incentives

For Xcel Energy with night time coal base

load:

Smart Charge after 10 PM avoids Capital

Costs and Green House Gasses 175

2008 Xcel Energy / NREL PHEV Study

• 6 Converted Ford Escapes (3 fleet, 3 personal use) and driven 40 miles per day (as do 85% of US commuters) at $7500 / car

• Results (yet not statistically significant)

Used only top 1/3 of 25 mile battery pack (parallel hybrid) Averaged over 6 months, 56.84 MPG in a SUV at $0.03 vs $0.11* Extremely consistent availability (except Sunday post 5:00 PM) Plugged In MORE often over time (from 50% to 80% over 6 months) Availability to utility at 60% - 85% with all factors considered Infrastructure is EVERYWHERE - “power to the curb” is there but what is the “tipping point”?

* at $2.00 / gal gas for 18 MPG for 12,000 per year at with $0.08 / kW*hr

* payoff at $7,500 cost to implement is 93,750 miles or 7.8 years while GM’s Volt is expected to have 140 MPG or 3.2 year payoff

Photo by ASC Designs 303-522-0066

176

Impacts from PHEVs & EVs

Without SmartCharging:

130 new power plants needed with 25% PHEV/EV penetration (source: ORNL), but still 40% less emissions when “filled” with coal based generation

With SmartCharging:

Theoretically ZERO new power plants needed (source: ORNL) until 73% of total fleet with generation “valley fill”

With SmartCharging:

Reduce to 85% fewer car emissions by reducing total number of power plants (source: NREL, and being studied by Xcel Energy)

177

Smart Consumer

178

Meter Data

179

Meter Data Management (MDM)

Multiple data

sources

Accurate and timely

data

Secure data

storage

Create and disseminate information

• AMI

• Manual

Readings

• SCADA

• OMS

• MWF

• Other

• Validating,

Editing and

Estimating (for

hourly data)

• Standards and

rules for service

order creation

• Proactive

assurance of

data availability

• Audit trail

• Securely manages

1,000 times more

data/meter than CIS

or AMI systems can.

• Tags for weather,

demographic and

other operational

characteristics

• Manage and access

non-traditional meter

data, e.g., PQ, volts,

etc.

• Interface to billing

systems

• Interface for Customer

Service Reps

• Create TOU billing

summaries

• Provide summary data

• Support operation &

planning needs

• Platform for customer

web presentment

180

Combined data flow

Data input,

validation and

warehouse

AMI/AMR

Systems:

• RF

• PLC

• Drive-by

Other inputs:

• Handhelds

• SCADA

• Manual data

• Weather data

• ???

CIS Systems:

• NISC

• SEDC

• Daffron

• Others

Analytics:

• Revenue

Protection

• System loss

analysis

• Planning

• Cost of Service

• Others

Operational

Support for AMI:

• Business rules

• Service order

interpretation

Web

Presentment:

• Meter data

• Customer and

billing data

• Demographic or

other data MDMS

G&T:

• Data for M&V of load control

• Class level data from each EMC

• Demographic data for planning

• Other

181

MDM Vendors

SIEMENS

182

http://www.itron.com/pages/index.asp

http://www.aclaratech.com/

http://jobs.oracle.com/

http://www.nisc.coop/index.jsp

myPower Pricing Pilot Overview

Control Group myPower Sense myPower Connection

Customers 450 Residential 379 Residential 319 Residential

Rate* RS TOU-CPP (RSP) TOU-CPP (RSP)

Equipment

Electric interval meter Electric interval meter Electric interval meter

Programmable thermostat

Two-way communications

infrastructure - PLC, RF, Hybrid

Customer Education

and Communication

N/A Mail

E-mail

Telephone

Mail

E-mail

Telephone

Signal to thermostat

Usage and Billing

Information

N/A Internet Internet

* RS = Residential Service, TOU-CPP = Time-of-Use, Critical Peak Pricing

183

myPower Time-of-Use – Critical Peak Pricing

(TOU-CPP)Summer 2007 Pricing Plan Weekdays

June - September

0

4

8

12

16

20

24

28

9 AM 1 PM

Time of Day

Pri

ce

in

ce

nts

pe

r kW

h

9 AM

8.7¢Medium

Price

(Base

Price)

23.7 ¢High

Price

(On-

Peak)

8.7¢Medium

Price

(Base

Price) 3.7¢

Low Price

(Night Discount)

6P M 10 PM

$1.46Critical

Price

Standard Residential Rate

WeekendsJune - September

0

4

8

12

16

20

24

28

9 AM

Time of Day

Pri

ce

in

ce

nts

pe

r k

Wh

9 AM

8.7¢Medium

Price

(Base Price)

3.7¢ Low Price

(Night Discount)

10 PM

Standard Residential Rate

184

myPower Sense Customers

Time-of-Use and Critical Peak Impacts

Source: myPower Pricing Pilot results based on 2006 and 2007 data through September 30, 2007

Customers who received no in-home technology were able to

reduce On-Peak period demand on critical peak days by up to

20%, even if they do not have Central AC.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

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 Ending

Average kW

per Customer

CPP

TOU

Baseline

Night Base On-Peak Base

With Central AC on Summer Peak Days

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

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 Ending

Average kW

per Customer

CPP

TOU

Baseline

Night Base On-Peak Base

Without Central AC on Summer Peak Days

185

myPower Connection and myPower Sense Customers

Summer Period Energy Savings Estimates

• Both the myPower participant and the Control Group customers showed increases in summer usage compared to prior years

• The increase in usage in the myPower participants‟ segments was significantly smaller than the Control Group.

• An overall energy savings estimate is developed by examining the difference between the Control Group‟s and participant groups‟ increase in energy use.

Source: myPower Pricing Pilot results based on 2006 and 2007 data through September 30, 2007

Customers who participated in myPower achieved summer period

energy savings in the range of 3-4%.

Variable

Control

Group

Change in

Use

Participant

Group

Change in

Use

Summer

Energy Savings

from TOU

(Percent)

Total Summer

Energy Savings

from TOU

(kWh per Cust)

myPower Connection 5.2% - 1.9% = 3.3% 139

myPower Sense with

Central AC 5.2% - 1.5% = 3.7% 144

myPower Sense without

Central AC 6.4% - 2.1% = 4.3% 127

186

Prototype on WSN for line monitoring

• Use a hierarchical communication topology [Casey_2011] – Avoid single point failure of sensors that is common in a multi-

hop sensor network

• Main characters of the implemented system – Sensors

• Self –Configurable

• Remote-controllable

• Able to adjust the data sampling frequency automatically – E.g. Increase the sampling frequency from 10 minutes to 5 seconds when a

fault is detected

– Gateway • Does not forward sensor packets until

– A full WLAN packet (about 18 sensor packets) has been accumulated, or

– Timeout happens

• [Casey_2011] P. Casey, N. Jaber, and K. Tepe, “Design and Implementation of a Cross-Platform Sensor Network

for Smart Grid Transmission Line Monitoring”, Proceedings IEEE SmartGridComm 2011.

187

Hardware and software implementation

• Hardware – Gateway

• Encompasses a ZigBee mote on Crossbow MIB510 programming board connected to a laptop

• ZigBee mote: Crossbow Micaz mote that utilizes the Chipcon CC2420 radio

• Linksys WUSB54GC as the WLAN interface

– Sensor node • A standalone ZigBee mote with a sensor board (Crossbow

MTS300CA)

• Software – TinyOS-2.x for the sensor

– Ubuntu 8.10 for the laptop (gateway)

188

Hierarchical communication topology

• Using ZigBee for communications between sensors and gateways

• Using 802.11 to build a mesh network among gateways

• Control center is the sink node

• The communication system for line monitoring is reliable since

– Both ZigBee and WLAN are reliable for this smart grid application

189

Field trial on PLC for smart meter applications

• PRIME (PoweRline Intelligent Metering Evolution) – A narrowband power line communications (PLC) technology

targeted for use in smart metering applications

– Use OFDM techniques and well-known forward error correction mechanisms, novel discovery and network-building MAC procedures

– Allow for cost-effective, seamless integration with recognized standard metering protocols such as DLMS/COSEM

– could become a globally recognized industry standard

• This paper [Berganza_2011] presents results obtained from real-field multi-vendor deployments with PRIME-compliant interoperable implementations at Iberdrola network in Spain.

• [Berganza_2011] I. Berganza, A. Sendin, A. Arzuaga, M. Sharma, and Badri Varadarajan, “PRIME on-field

deployment - First summary of results and discussion”, Proceedings IEEE SmartGridComm 2011.

190

Main lessons learned from the field trials

• Signal interference due to misconfigurations

– Two concentrators were deployed on the two

transformers in a same substation

• Beacons collide in the time domain

– Some service nodes are jumping between the two subnetworks

• Should only set one concentrator, and set others as

switches

• Unreliable communications when not all meters

governed by a substation are PRIME meters

– The signal-to-noise ratio might not be high enough

191

PLC communication for remote areas

• This paper [Kikkert_2011]describes an accurate SWER line model – Single Wire Earth Return (SWER) lines are used in Australia,

USA, South Africa and many other countries to provide power to remote communities

– The model demonstrate the severe signal channel degradation that can occur due to line branches and coupling networks

– The model is verified with measurements from two sites in Australia.

– Data rates are at 22.8 kbps on a 14 km SWER line • when the attenuator is set to less than or equal to 15 dB attenuation

– Predict that PLC communication systems using G3-PLC modems on SWER lines in excess of 2000 km are feasible

• [Kikkert_2011] C. Kikkert, “Effect of Couplers and Line Branches on PLC

Communication Channel Response”, Proceedings IEEE SmartGridComm 2011.

192

Device communications using SCADA systems

• Communications in traditional power grid are mainly enabled by a centralized supervisory control and data acquisition (SCADA) system

• In [Lu_2011], they establish a monitoring system for a Solid State Transformer (SST) in a micro smart grid - Green Hub – To verify that SCADA system can be used to support such an

application

– The one megawatt Green Hub system is a power electronics based power system in the FREEDM systems center at the North Carolina State University.

• It is established to demonstrate salient features and capabilities of the FREEDM system on renewable energy generation, distribution, storage and management

• [Lu_2011] X. Lu, W. Wang, A. Juneja, and A. Dean , “Talk to Transformers: An Empirical Study

of Device Communications for the FREEDM System”, Proceedings IEEE SmartGridComm 2011.

193

Implementation of SST monitoring system

• In the network domain, a control center is connected to the SST

controller

– via a Local Area Network (LAN)

• DNP3 is overlayed over TCP/IP in he implementation

– DNP3 (distributed network protocol 3.0) is a widely-adopted SCADA protocol

194

Conclusions and lessons

• Conclusion

– The DNP3 based SCADA system can be used in the smart grid

• for the device monitoring and control

• Lessons

– A careful optimization is crucial to reduce the total delay

• By optimizing every time-consuming part of every system component

• Delay is the primary concern for most smart grid applications

– The DNP3-based monitoring system is not suitable for more time stringent applications like relay protection

• The architecture is too complex and induce extra delay

195

Open Research Issues

• Cost-Aware Data Communication and Networking

Infrastructure

• Quality-of-Service (QoS) Framework

• Optimal Network Design

• Need of Secured Communication Network Infrastructure

• Plug-in Hybrid Electric Vehicle (PHEV)



Cost-Aware Data Communication and Networking

Infrastructure

• There is a cost in retrieving the real-time information (e.g.,

power pricing, metering data, and surveillance data), which

increases with the increase in frequency of inquiry

• However, the performances such as latency, bandwidth,

reliability must be met

• The cost optimization for data monitoring and transferred

must be performed



Quality-of-Service (QoS) Framework

• The QoS in smart grid can be defined by accuracy and

effectiveness with which different information such as

equipment‟s state, load information, and power pricing are

delivered timely to the respective parties

• QoS framework can be developed by identifying the

specific QoS requirements and priorities for specific

communication network in smart grid


Maximum Latency Communication Type ≤ 4 ms Protective relaying Sub-seconds Wide area situational awareness monitoring Seconds Substation and feeder supervisory control and data acquisition (SCADA) Minutes Monitoring noncritical equipment and marketing pricing information Hours Meter reading and longer-term pricing information Days/Weeks/Months Collecting long-term usage data


Optimal Network Design

• Dedicated network can be built to support the QoS- and

security-sensitive smart grid applications (e.g., status

monitoring and time-of-use report)

• Optimal network devices, their connections, and protocols

have to be chosen to avoid congestion and failure

• Shared network (e.g., cellular service) can be used to

support noncritical smart grid applications (e.g., billing)

• Integration of dedicated and shared network can be explored



Need of Secured Communication Network Infrastructure

• If smart grid is attacked, the hackers can penetrate the

network and alter critical system parameters which could

destabilize the grid in an unpredictable way causing

nationwide crisis

• Intrusion detection and prevention for smart grid (e.g., AMI

and WAMS)

• Public key infrastructure (PKI) for smart grid



Plug-in Hybrid Electric Vehicle (PHEV)

• With the use of electric power, PHEV has lower operational

cost and smaller emission of CO2

• PHEV requires electric charging from charging station

• To ensure stabilized load, electric power has to be supplied

according to the demand from PHEV

• Communications intrastructure for PHEV charging can be

proposed (e.g., [Erol-Kantarci 2011])

– Utility company communicates with substation control center (SCC)

using WiMAX and charging station using wireless mesh network

– SCC decides to accept or refuse the charging request from PHEV


Conclusion

• Smart grid will be a crucial technology to improve the

efficiency of the power grid

• There are many issues related to data communications and

networking


References (1) [Niyato_2011] D. Niyato, L. Xiao, and P. Wang, "Machine-to-machine communications for home energy management system in smart

grid," IEEE Communications Magazine, vol. 49, no. 4, pp. 53-59, April 2011.

[Bruce_1998] A. G. Bruce, “Reliability analysis of electric utility SCADA systems,” IEEE Transactions on Power Systems, vol. 13, no. 3,

pp. 844-849, August 1998.

[Xie_2002] Z. Xie, G. Manimaran, V. Vittal, A. G. Phadke, and V. Centeno, “An information architecture for future power systems and its

reliability analysis,” IEEE Transactions on Power Systems, vol. 17, no. 3, pp. 857- 863, August 2002.

[Wang_2010] Y. Wang, W. Li, and J. Lu, “Reliability analysis of wide-area measurement system,” IEEE Transactions on Power Delivery,

vol. 25, no. 3, pp. 1483-1491, July 2010.

[Erol-Kantarci 2011] M. Erol-Kantarci, J. H. Sarker, and H. T. Mouftah, "Communication-based Plug-In Hybrid Electrical Vehicle load

management in the smart grid," IEEE Symposium on Computers and Communications (ISCC), pp. 404-409, June 2011-July 2011.

[Liu_2010] E. Liu, M. L. Chan, C. W. Huang, N. C. Wang, and C. N. Lu, "Electricity grid operation and planning related benefits of

advanced metering infrastructure," International Conference on Critical Infrastructure (CRIS), pp.1-5, September 2010.

[Mao_011] R. Mao and V. Julka, "Wireless Broadband Architecture Supporting Advanced Metering Infrastructure," IEEE Vehicular

Technology Conference (VTC Spring), pp.1-13, May 2011.

[Chen_2010] S. Chen, J. Lukkien, and L. Zhang, "Service-oriented Advanced Metering Infrastructure for Smart Grids," Asia-Pacific

Power and Energy Engineering Conference (APPEEC), pp.1-4, March 2010.

[Yu_2011] R. Yu, Y. Zhang, S. Gjessing, C. Yuen, S. Xie, and M. Guizani, "Cognitive radio based hierarchical communications

infrastructure for smart grid," IEEE Network, vol. 25, no. 5, pp. 6-14, September-October 2011.

[Wang_2007] Y. Wang, J. P. Lynch, and K. H. Law, “A wireless structural health monitoring system with multithreaded sensing devices:

Design and validation,” in Structure and Infrastructure Engineering, vol. 3(2), pp. 103-120, 2007.

[Ullo_2010] S. Ullo, A. Vaccaro, and G. Velotto, “The role of pervasive and cooperative sensor networks in smart grids communication,”

in Proceedings of IEEE Mediterranean Electrotechnical Conference (MELECON), pp. 443-447, April 2010.

[Yan_2011] Y. Yan, Y. Qian, and H. Sharif, “A secure and reliable in-network collaborative communication scheme for advanced

metering infrastructure in smart grid,” in Proceedings of IEEE Wireless Communications and Networking Conference (WCNC), pp.

909-914, March 2011.

[Sreesha_2011] A. A. Sreesha, S. Somal, and I-Tai Lu, “Cognitive radio based wireless sensor network architecture for smart grid utility,”

in Proceedings of IEEE Long Island Systems, Applications and Technology Conference (LISAT), May 2011.


References (2) [Hung_2010] K. S. Hung, W. K. Lee, V. O. K. Li, K. S. Lui, P. W. T. Pong, K. K. Y. Wong, G. H. Yang, and J. Zhong, "On Wireless

Sensors Communication for Overhead Transmission Line Monitoring in Power Delivery Systems," IEEE International Conference

on Smart Grid Communications (SmartGridComm), pp. 309-314, October 2010.

[Naduvathuparambil_2002] B. Naduvathuparambil, M. C. Valenti, and A. Feliachi, "Communication delays in wide area measurement

systems," Proceedings of the Thirty-Fourth Southeastern Symposium on System Theory, pp. 118- 122, 2002.

[Shahraeini_2011] M. Shahraeini, M. H. Javidi, and M. S. Ghazizadeh, "Comparison Between Communication Infrastructures of

Centralized and Decentralized Wide Area Measurement Systems," IEEE Transactions on Smart Grid, vol. 2, no. 1, pp. 206-211,

March 2011.

[Mohsenian-Rad_2010] A. Mohsenian-Rad, V. W. S. Wong, J. Jatskevich, R. Schober, and A. Leon-Garcia, "Autonomous Demand-Side

Management Based on Game-Theoretic Energy Consumption Scheduling for the Future Smart Grid,” IEEE Transactions on Smart

Grid, vol. 1, no. 3, pp. 320-331, December 2010.

[Hirst_1998] E. Hirst and B. Kirby, “Unbundling Generation and Transmission Services for Competitive Electricity Markets: Ancillary

Services”, NRRI-98-05, National Regulatory Research Institute, Columbus, OH, Jan. 1998

[Giani_2011] A. Giani, E. Bitary, M. Garciay, M. McQueenz, P. Khargonekarx, and K. Poolla, “Smart Grid Data Integrity Attacks:

Characterizations and Countermeasures”, Proceedings of IEEE SmartGridComm 2011.

[Vukovic_2011] O. Vukovic, K-C Sou, G. Dan, and H. Sandberg, “Network-layer Protection Schemes against Stealth Attacks on State

Estimators in Power Systems”, Proceedings of IEEE SmartGridComm 2011.

[Tajer_2011] A. Tajer, S. Kar, V. Poor, and S. Cui, “Distributed Joint Cyber Attack Detection and State Recovery in Smart Grids”,

Proceedings of IEEE Globecom 2011.

[Esmalifalak_2011] M. Esmalifalak, H. Nguyen, R. Zheng, and Z. Han, “Stealth False Data Injection using Independent Component

Analysis in Smart Grid”, Proceedings of IEEE SmartGridComm 2011.

[Kolesnikov_2011] V. Kolesnikov, W. Lee, and J. Hong, “MAC Aggregation Resilient to DoS Attacks”, Proceedings of IEEE

SmartGridComm 2011.

[Zhu_2011] T. Zhu, S. Xiao, Y. Ping, D. Towsley, and W. Gong, “A Secure Energy Routing Mechanism for Sharing Renewable Energy

in Smart Microgrid”, Proceedings of IEEE SmartGridComm 2011.

[Jokar _2011] P. Jokar, H. Nicanfar, V. Leung, “Specification-based Intrusion Detection for Home Area Networks in Smart Grids”,

Proceedings of IEEE SmartGridComm 2011.


References (3) [Chim_2011] T. Chim, S. Yiu, L. Hui, and V. Li, “PASS: Privacy-preserving Authentication Scheme for Smart Grid Network”,


[Rajagopalan_2011] S. Rajagopalan, L. Sankar, S. Mohajer, and V. Poor, “Smart Meter Privacy: A Utility-Privacy Framework”,


[Kim_2011] Y. Kim, E. Ngai, and M. Srivastava, “Cooperative State Estimation for Preserving Privacy of User Behaviors in Smart Grid”,


[Casey_2011] P. Casey, N. Jaber, and K. Tepe, “Design and Implementation of a Cross-Platform Sensor Network for Smart Grid

Transmission Line Monitoring”, Proceedings IEEE SmartGridComm 2011.

[Berganza_2011] I. Berganza, A. Sendin, A. Arzuaga, M. Sharma, and Badri Varadarajan, “PRIME on-field deployment - First summary

of results and discussion”, Proceedings IEEE SmartGridComm 2011.

[Kikkert_2011] C. Kikkert, “Effect of Couplers and Line Branches on PLC Communication Channel Response”, Proceedings IEEE

SmartGridComm 2011.

[Lu_2011] X. Lu, W. Wang, A. Juneja, and A. Dean , “Talk to Transformers: An Empirical Study of Device Communications for the

FREEDM System”, Proceedings IEEE SmartGridComm 2011.


Communications and Networking for Smart Grid...

Documents

Transcript of Communications and Networking for Smart Grid...