Making Electric Power Grids Resilient and Secure

Making Electric Power Grids Resilient and Secure

Loading
Loading Social Plug-ins...
Language: English
Save to myLibrary Download PDF
Go to Page # Page of 78

Description: Electric Power Grid, Infrastructure Security, Power Grid Vulnerabilities, Smart Grid Vulnerabilities, Stuxnet & Digital Systems, Threat Evolution, Malicious Code, Complex Interactive Networks, Enterprise Information Security, Utility Telecommunications.

 
Author: S. Massoud Amin (Fellow) | Visits: 1993 | Page Views: 6339
Domain:  Green Tech Category: Environmental Subcategory: Smart Grid 
Upload Date:
Link Back:
Short URL: http://www.wesrch.com/energy/pdfTR1AU1FU5JWSL
Loading
Loading...



px *        px *

* Default width and height in pixels. Change it to your required dimensions.

 
Contents:
Enabling a Resilient and Secure North American Electric Power Grid
Director, Technological Leadership Institute Honeywell/H.W. Sweatt Chair in Technological Leadership Professor, Electrical & Computer Engineering University Distinguished Teaching Professor

S. Massoud Amin, D.Sc.

Keynote address at the 4th International Symposium on Resilient Control Systems August 10, 2011
Material from the Electric Power Research Institute (EPRI), and support from EPRI, NSF, SNL and ORNL for my graduate students' doctoral research is gratefully acknowledged
Copyright � 2011 No part of this presentation may be reproduced in any form without prior authorization.

R&D Challenges
� Sensing and Communication � Early Fault Detection and System V&V � Systems Integration and Interoperability � Security (from embedded... to endtoend)

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

A "Sanitized" Example: Lack of awareness and inadvertent connection to the Internet
� Power plant: 2 250MW, gas fired turbine, combined cycle, 5 years old, 2 operators, and typical multiscreen layout: � "A: do you worry about cyber threats? � Operator: No, we are completely disconnected from the net. � A: That's great! This is a peaking unit, how do you know how much power to make? � Operator: The office receives an order from the ISO, then sends it over to us. We get the message here on this screen. � A: Is that message coming in over the internet? � Operator: Yes, we can see all the ISO to company traffic. Oh, that's not good, is it?"
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

"... And so, extrapolating from the best figures available, we see that current trends, unless dramatically reversed, will inevitably lead to a situation in which the sky will fall."

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Infrastructure Security

We are "Bullet Proof"

The Truth

"The Sky is Falling"

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Power Grid Vulnerabilities
� Physical:
� Over 450,000 miles of 100kV or higher (215,000 miles of 230kV or higher) transmission lines, and many more thousands of miles of lowervoltage lines � Natural disasters or a wellorganized group of terrorists can take out portions of the grid as they have done in the U.S., Colombia, and other countries � Effects typically confined to the local region.

� OpenSource Information:
� Analysts have estimated that public sources could be used to gain at least 80% of information needed to plot an attack
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Smart Grid Vulnerabilities
� Cyber:
� Existing control systems were designed for use with proprietary, standalone communications networks � Numerous types of equipment and protocols are used � More than 90% of successful cyber attacks take advantage of known vulnerabilities and misconfigured operating systems, servers, and network devices � Possible effects of attacks:
1) 2) 3) 4) Loss of load Loss of information Economic loss Equipment damage
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Stuxnet & Digital Systems: SCADA, EMS, ICS
� Iran's nuclear agency trying to stop computer worm Stuxnet worm can take over systems that control industrial plants By NASSER KARIMI updated 9/25/2010 1:50:43 PM ET TEHRAN, Iran Iran's nuclear agency is trying to combat a complex computer worm that has affected industrial sites throughout the country and is capable of taking over power plants, Stuxnet can take over systems that control the inner workings of industrial plants. Experts in Germany discovered the worm in July, and it has since shown up in a number of attacks primarily in Iran, Indonesia, India and the U.S. The ISNA report said the malware had spread throughout Iran, but did not name specific sites affected. The destructive Stuxnet worm has surprised experts because it is the first one specifically created to take over industrial control systems, rather than just steal or manipulate data. The U.S. is also tracking the worm, and the DHS has specialized teams that can respond quickly to cyber emergencies at industrial facilities across the country.
http://www.msnbc.msn.com/id/39357629/ns/technology_and_science tech_and_gadgets#
� 2011 No part of this presentation may be reproduced in any form without prior authorization.





� �



Threat Evolution: Malicious Code
Seconds Class III Human response: impossible Automated response: unlikely Proactive blocking: possible Class II Human response: difficult/impossible Automated response: possible Class I Human response: possible "Flash" Threats

Contagion Timeframe

"Warhol" Threats

Minutes

Hours

Blended Threats e-mail Worms

Days Macro Viruses Weeks or months File Viruses

Early 1990s

Mid 1990s

Late 1990s

2000

2003

Time

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

What Can They Do and How Can They Do It?
Information Leakage Integrity Violation Denial of Service Illegitimate Use

Eavesdropping Traffic Analysis EM/RF Interception Indiscretions by Personnel Media Scavenging

Penetration Masquerade Bypassing Controls Authorization Violation Physical Intrusion

Planting Trojan Horse Trapdoor Service Spoofing Theft

Information Leakage Integrity Violation Theft Replay Resource Exhaustion Integrity Violation

Intercept/Alter Repudiation

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Programs Initiated and Developed at EPRI
1999-2001
Y2K2000-present

Overview of focused research areas (19982003):
2002-present 2001-present

EPRI/DoD Complex Interactive Networks (CIN/SI)
Underpinnings of Interdependent Critical National Infrastructures Tools that enable secure, robust & reliable operation of interdependent infrastructures with distributed intelligence & selfhealing

Enterprise Information Security (EIS)
1. 2. 3. 4. 5.

Infrastructure Security Initiative (ISI)
Response to 9/11 Tragedies 1. Strategic Spare Parts Inventory 2. Vulnerability Assessments 3. Red Teaming 4. Secure Communications

Consortium for Electric Infrastructure to Support a Digital Society (CEIDS)
1. 2. 3.

Information Sharing Intrusion/Tamper Detection Comm. Protocol Security Risk Mgmt. Enhancement High Speed Encryption

4.

Self Healing Grid IntelliGridTM Integrated Electric Communications System Architecture Fast Simulation and Modeling

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Enterprise Information Security (EIS) program
Information Networks for OnLine Trade, Security & Control
OASIS
ICCP UCA CIM
Transmission Reservation Congestion Management Transaction Information System Trade Data Net

API
Ancillary Services

ISN
ICCP UCA
PRM

TTC

RSDD

Security Data Net

PSAPAC

TRELSS

DTCR

DSA

VSA

TRACE

CC-RTU
Control Data Net

ICCP UCA

Integrated Substation Diagnostics FACTS Controllers

RCM

MMW Stabilizer Tuning

WAMS

Event Recording and Diagnostics

EIS Focus

Dynamic Data Net
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Lessons learned, e.g.: Redundancy Lowers Impact of Threats
� Two Separate Control Rooms � 500 miles apart � Dual EMS systems at each location + Training/testing EMS � Diversified communications networks

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Utility Telecommunications
� Electric power utilities usually own and operate at least parts of their own telecommunications systems � Consist of backbone fiber optic or microwave connecting major substations, with spurs to smaller sites � Media:
� � � � � � Fiber optic cables Digital microwave Analog microwave Multiple Address Radio (MAS) Spread Spectrum Radio VSAT satellite � � � � � Power Line Carrier Copper Cable Leased Lines and/or Facilities Trunked Mobile Radio Cellular Digital Packet Data (CDPD) � Special systems (Itron, CellNet)

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Context: Better Situational Awareness and Automation
� Increasing Dependence on ICT, Computation and Communications. � Increasing Complexity: System integration, increased complexity: call for new approaches to simplify the operation of complex infrastructure and make them more robust to attacks and interruptions. � Centralization and Decentralization of Control: The vulnerabilities of centralized control seem to demand smaller, local system configurations. Resilience rely upon the ability to bridge topdown and bottomup decision making in real time.
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Data and Measurements
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Detecting Precursors: Classification of fault signatures
L o s s o f N e a rb y G e n e ra to r E v e n ts 6 0 .0 5
6 0 .0 4 L o s s o f R e m o te G e n e r a to r E v e n ts

60 Frequency (Hz)
Frequency (Hz)

6 0 .0 2 60 5 9 .9 8 5 9 .9 6 5 9 .9 4

5 9 .9 5

5 9 .9

5 9 .8 5

0

5

10

15 20 T im e (s e c )

25

30

5 9 .9 2

0

5

10

15 20 T im e (s e c )

25

30

L o s s o f L o a d E v e n ts 6 0 .0 3 6 0 .0 4 6 0 .0 2 Frequency (Hz) Frequency (Hz)

L in e T rip E v e n ts

6 0 .0 1

6 0 .0 2

60

60

5 9 .9 9

5 9 .9 8

0

5

10

15 20 T im e (s e c )

25

30

5 9 .9 8

0

5

10

15 20 T im e (s e c )

25

30

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Disturbance Feature Extraction
Disturbance Loss of nearby generation Loss of remote generation Loss of load Frequency change Negative Negative Positive Frequency derivative Steep Moderate Moderate Steep Small Line flow change Large Negligible Detectable Large oscillations

Line trip close to Negligible DRD Oscillations Negligible

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Precursor Detection for Situational Awareness
� Enhancing Reliability and Security of Network Operation via quantification of the system state and its "direction/ speed/momentum" toward a major failure � Making Network Availability (quick restoration) a key requirement � Introducing Quality of Service as an additional constraint

Combinations not leading to System Failure
SC: Combinations leading to System Failure

� Ultimately, enabling operators to act more efficiently and with greater Which trajectories lead to confidence in difficult (sometimes unclear, unexpected or even conflicting) catastrophic failures? circumstances
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

MultiScale Time Hierarchy of Power Systems
ACTION / OPERATION Wave effects (fast dynamics, lightning caused overvoltages) Switching overvoltages Fault protection Electromagnetic effects in machine windings Stability Stability Augmentation Electromechanical effects of oscillations in motors & generators Tie line load frequency control Economic load dispatch Thermodynamic changes from boiler control action (slow dynamics) System structure monitoring (what is energized & what is not) System state measurement and estimation System security monitoring Load Management, load forecasting, generation scheduling Maintenance scheduling Expansion planning Power plant site selection, design, construction, environmental impact, etc. TIME FRAME Microseconds to milliseconds Milliseconds 100 milliseconds or a few cycles Milliseconds to seconds 60 cycles or 1 second Seconds Milliseconds to minutes 1 to 10 seconds; ongoing 10 seconds to 1 hour; ongoing Seconds to hours Steady state; on-going Steady state; on-going Steady state; on-going 1 hour to 1 day or longer; ongoing. Months to 1 year; ongoing. Years; ongoing 10 years or longer

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Precursor Detection for Situational Awareness Fast modeling highconfidence lookahead simulation
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Saving systems from collapse in Multihazard environments: The Case of the Missing Wing (198397)

NASA/MDA/WU IFCS: NASA Ames Research Center, NASA Dryden, Boeing Phantom Works, and Washington University in St. Louis.
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Critical System Dynamics and Resilience Capabilities
� Anticipation of disruptive events � Lookahead simulation capability � Fast isolation and sectionalization � Adaptive islanding � Selfhealing and restoration
resilience, noun, 1824: The capability of a strained body to recover its size and shape after deformation caused especially by compressive stress; An ability to recover from or adjust easily to misfortune or change

Resilience enables "Robustness": A system, organism or design may be said to be "robust" if it is capable of coping well with variations (internal or external and sometimes unpredictable) in its operating environment with minimal damage, alteration or loss of functionality.
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Complex Dynamical Systems
Systems have multiple "modes" during which specific operational and control actions/reactions take place Enable complex systems to become smarter and adaptive to stressors ... detect precursors, predict, and adapt to disturbances
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Complex Dynamical Systems
Normal mode: economic dispatch, load frequency control, maintenance, forecasting, etc.; � Alert mode: red flags, precursor detection, reconfiguration and response; � Emergency/Disturbance mode: stability, viability, and integrity instability, load shedding, etc.; � Restorative mode: rescheduling, resynchronization, load restoration, etc.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

EPRI/DOD Complex Interactive Network/Systems Initiative (1998-2002)

Self-healing Grid and Network-centric Objective Force

Complex interactive networks:
� Energy infrastructure: Electric power grids, water, oil and gas pipelines � Telecommunications: Information, communications and satellite networks � Transportation and distribution networks � Energy markets, banking and finance
108 professors and over 240 graduate students in 28 U.S. universities were funded: Over 420 publications, and 24 technologies extracted, in the 3-year initiative

Goal: Develop tools that enable secure, robust and reliable operation of interdependent infrastructures with distributed intelligence and self-healing abilities
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Background: The SelfHealing Grid
Dependability/ Robustness/ Self-Healing (min-hours)
Hidden Failure Monitoring Agents Vulnerability Assessment Agents Event identification Agents Planning Agents Reconfiguration Agents Knowledge/Decision Exchange Restoration Agents

(sec)

Event/Alarm Filtering Agents

Triggering EventsModel Update Agents

Plans/Decisions Command Check Interpretation Agents Consistency Frequency Stability Agents Controls

Inputs

Events/ Alarms

Fault Isolation Agents

Autonomy/ Fast Control (msec)

Protection Agents

Inhibitor Signal Controls Power System

Generation Agents

EPRI/DoD CIN/S Initiative
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Spectrum: Deterrence Mitigation Restoration
Intelligent Adaptive Islanding: remedial action in WECC
Northern Electrical Island
British Columbia & Alberta

Southern Electrical Island

Northwest

Montana Idaho Wyoming

PACI

PDCI

California & Southern Nevada

Utah& Northern Nevada

Colorado

Arizona New Mexico

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

The SelfHealing Grid Intelligent Adaptive Islanding
35 33 32 31 30 80 79 78 72 v v 76

74 66 75

230 kV 345 kV 345 kV 500 kV

77

82 81 86 84 85 156 167 165 158 159 157 161 162 5 11

83

112 114 155 44 45 160 115

6

166

163 118 119 107 110 103 102 56 48 47 49 108 109 104 63 154 151 140 141 149 150 43 42 50 153 37 64 143 142 146 147 138 139 14 3 4 12 13 8 17 18

7
9

145 136 152 57 16

19

15

8/10/2011

32
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

The SelfHealing Grid
60.0 59.5 59.0
60.0 59.8

59.6

58.5 58.0 57.5
0.0 0.7 1.4 2.1 Time in Seconds 2.8 3.5

59.4 59.2

59.0
0.0 0.7 1.4 2.1 Time in Seconds 2.8 3.5

Past Scheme

New Scheme

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Critical Infrastructure Security & Protection

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Prioritization: Security Index
General
1. 2. 3. 4. Corporate culture (adherence to procedures, visible promotion of better security, management security knowledge) Security program (up-to-date, complete, managed, and includes vulnerability and risk assessments) Employees (compliance with policies and procedures, background checks, training) Emergency and threat-response capability (organized, trained, manned, drilled)

Physical
1. 2. 3. 4. Requirements for facilities (critical list, inventory, intrusion detections, deficiency list) Requirements for equipment (critical list, inventory, deficiency list) Requirements for lines of communications (critical list, inventory, deficiency list) Protection of sensitive information

Cyber and IT
1. 2. 3. 4. Protection of wired networks (architecture analysis, intrusion detection) Protection of wireless networks (architecture analysis, intrusion detection, penetration testing) Firewall assessments Process control system security assessments (SCADA, EMS, DCS)
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Assessment & Prioritization: A Composite Spider Diagram to Display Security Indices

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Example of In Depth Analysis: Critical Contingency Situations
Critical Root Causes in the Proba/Voltage Impact State space (Region Cause: all, Affected Region: all)

1500.0575 Impact (kV)

Most significant root cause

1000.0575

500.057498

0.0574983 0.000001

0.00001

0.0001

0.001

0.01

0.1

1

Logarithmic Probability (direct)

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Initial Conditions on August 14, 2003
STAR-345 kV BUS Aug 8-14, 2003
360

Fri 8

Sat 9

Sun 10

Mon 11

Tue 12

Wed 13

Thur 14

355

350

kV

345

340

335

330 1 25 49 73 Hour 97 121 145

Star 345 kV Bus Voltages (Aug 8-14, 2003)

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Situation Awareness Tool (SAT)

Source: NERC
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Situation Awareness Tool (SAT)

Source: NERC
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

What are we working on at the U of Minnesota ?
� � � � � Integrating PHEVs into the grid Grid agents as smart and distributed computer Fast power grid simulation and risk assessment More Secure and Smarter Grid Security of cyberphysical infrastructure

University of Minnesota Center for Smart Grid Technologies (2003present) Dept. of Electrical & Computer Engineering Faculty: Professors Massoud Amin and Bruce Wollenberg PhD Candidates/Research Assistants: Anthony Giacomoni, Jesse Gantz, Laurie Miller, and Sara Mullen (PhD 9/9) PI: M. Amin (support from EPRI, NSF, Honeywell, SNL, ORNL, and UofM funding)
Center for Smart Grid Technologies
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Objectives
� Our strategic goal is to better understand the true dynamics of complex interdependent energy/communications/economic networks in order to enable stronger, greener, more secure and smarter power grids. � The objective of this project is to model, design and develop reconfigurable and distributed smart energy systems supported by secure sensing/wireless communication network overlay and faultresilient realtime controls.
Center for Smart Grid Technologies
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Integrated Sensing, Protection and Control
Information and Communication Systems Threats or Disturbances

Power System Electricity Markets

Protection and Control Systems

Human Agents

In many complex networks, the human participants themselves are both the most susceptible to failure and the most adaptable in the management of recovery.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Smart Grid Interdependencies
Security, Efficiency, and Resilience

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Smart Grid as a Distributed Computer
... the State Estimator

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Transmission Limits
� High dimensional problem
� Large interconnection models (1/5 of the North American system) require ~40,000 buses & ~50,000 lines, and ~3,000 generators with ~120 control areas � Each line has a capacity limit � N1 Contingency Criteria: The system must withstand the loss of any one line or generator (~53,000 contingencies)
� 53,000 x 50,000 = 2,650,000,000 possible constraints

� Reliable operation requires an operating point that satisfy these 2.65 billion constraints!
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

State Estimation:
Z = h(X) + V
where: Z = The measurement vector X = The state vector V = The measurement error vector h(X) = Nonlinear observation function, the set of electrical equations relating MW and MVAR values to bus voltages and angles Min. J(X) = [ Z h(X) ]T R1 [ Z h(X) ] R = The measurement error covariance matrix Extended to Advanced Topology Estimator: � determine unknown substation switch settings from voltages, power flows, and current measurements

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Distributed State Estimation & System Identification Lookahead Simulation
� Process of using a set of over determined, noisy measurements to estimate the system state
� V and are the state variables in our problem � We use measurements that are functions of V and to form our estimate

� One of the first milestones in moving towards a Smart Grid is demonstrating that distributed state estimation is workable

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Central computer control system

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Using the grid agents to do power system state estimation calculation

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Relevant equations
� PQ node (typical load node):
Pk Vk Vm (Gkm cos km Bkm sin km )
m 1 n

Qk Vk Vm (Gkm sin km Bkm cos km )
m 1

n

� PV node (typical generator node) has same P equation and trivial V equation � Line Flow Equations
Pij GijVi 2 ViV j (Gij cos ij Bij sin ij ) Qij BijVi 2 VV j (Gij sin ij Bij cos ij ) Vi 2 Bcap i
ij

� In our case, each bus has six measurements so we write 6 equations
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Algorithm
� 1.Choose Slack Bus � 2.Make initial guess for voltages and phases � 3.Determine measurement error

error ( zmeasi

f i (V , ))

� 4.Calculate Jacobian as a function of V, � 5.Compute correction z1 f1 (V , ) T 1 1 T 1 x ( J R J ) J R z2 f 2 (V , ) � � � � 6.Determine if a component of the correction vector exceeds 7.If so, apply correction, repeat steps 36 8.Share state variables with adjacent nodes 9.Repeat
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Exchange of data

Marcus Braun

Gives rapid reliable algorithm convergence
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Smart Grid Protection Schemes & Communication Requirements
Type of relay
Data Volume (kb/s)
Present Future

Latency
Primary Secondary (ms) (s)

Over current protection Differential protection Distance protection Load shedding Adaptive multi terminal Adaptive out of step

160 70 140 370 200 1100

2500 1100 2200 4400 3300 13000

48 48 48

0.31 0.31 0.31

0.060.1 (s) 48 0.31

Depends on the disturbance

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

EndtoEnd Power Delivery Operation & Planning
Power Plants Fuel Supply System Transmission System Distribution System

Renewable Plants Fuel Source/Storage Energy Storage
Controllers Sensors

End-uses & DR

Data Communication Data Communication Wide Area Control Wide Area Control
Dynamic Power Plant Models

M ZIP

Dynamic Load Models

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

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Smart Grid: Tsunami of Data Developing
New devices in the home enabled by the smart meter
800 TB

Annual Rate of Data Intake

600 TB

OMS Upgrade RTU Upgrade

400 TB

Mobile Data Goes Live You are here.
200 TB

Programmable Communicating Thermostat Come Online AMI Deployment Distribution Management Rollout GIS System Deployment

Time Distribution Automation

Substation Automation System Workforce Management Project

Tremendous amount of data coming from the field in the near future paradigm shift for how utilities operate and maintain the grid
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

New Challenges for a Smart Grid
� Need to integrate:
� � � � � Largescale stochastic (uncertain) renewable generation Electric energy storage Distributed generation Plugin hybrid electric vehicles Demand response (smart meters)

� Need to deploy and integrate:
� New Synchronized measurement technologies � New sensors � New System Integrity Protection Schemes (SIPS)

� Critical Security Controls
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

New Challenges for a Smart Grid
� Cybersecurity and Interoperability
� NIST's Mandate: Energy Independence and Security Act (EISA) of 2007, Title XIII, Section 1305. Smart Grid Interoperability Framework � The Framework: � common architecture � flexible, uniform, technologyneutral � aligns policy, business, and technology approaches � includes protocols and standards for information management � Data exchange within the Smart Grid and between devices and technologies � NIST has offered a Smart Grid architecture; priorities for interoperability standards, including cybersecurity: NISTIR 7628

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Our team's Smart Grid Research

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Local System Communication Overlay

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Intelligent Agents and Functionalities

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Centralized or Decentralized Control?

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Centralized or Decentralized Control?

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

"Computers are incredibly fast, accurate, and stupid; humans are incredibly slow, inaccurate and brilliant; together they are powerful beyond imagination."
Albert Einstein

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

I35W bridge

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

To improve the future and avoid a repetition of the past: Sensors built in to the I 35W bridge at less than 0.5% total cost by TLI alumni

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Enabling the Future
Infrastructure integration of microgrids, diverse generation and storage resources into a secure system of a smart self-healing grid

Source: Interview with Massoud Amin, "Upgrading the grid," Nature, vol. 454, pp. 570�573, 30 July 2008
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Source: M. Amin, IEEE Smart Grid Newsletter chairman http://smartgrid.ieee.org/publications/smartgridnewsletter
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

R&D challenges
� Develop a theoretical framework, modeling and simulation tools for interdependencies and their fundamental characteristics, to provide:
� An understanding of true dynamics and impact on coupled infrastructure robustness and reliability. � An understanding of emergent behaviors, and analysis of multi-scale and complexity issues and trends in the future growth and operations. � Real-time state estimation and visualization of infrastructures-- flexible and rapidly adaptable modeling and estimation

� Integrated assessment, monitoring, and early warning:
� Vulnerability assessment, risk analysis and management � Underlying causes, distributions, and dynamics of and necessary/sufficient conditions for cascading breakdowns (metrics). � Data mining and early signature detection � Infrastructure databases.

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Selected Areas in Applied Mathematics Dynamical Systems and Controls
� Modeling: Idealized models, consisting of static graphtheoretic models, and interactive dynamic models, such as interconnected differentialalgebraic systems; Hybrid Models. Robust Control: Design of selfhealing systems requires the extension of the theory of robust control in several ways beyond its present focus on the relatively narrow problem of feedback control. Complex Systems: Theoretical underpinnings of complex interactive systems. Dynamic Interaction in Interdependent Layered Networks: Characterization of uncertainty in large distributed networks: Multiresolutional techniques where various levels of aggregation can coexist. Disturbance Propagation in Networks: Prediction and detection of the onset of failures both in local and global network levels. Forecasting, Handling Uncertainty and Risk: Characterizing Uncertainties and Managing Risk; Hierarchical and multiresolutional modeling and identification; Stochastic analysis of network performance; Handling Rare Events.
� 2011 No part of this presentation may be reproduced in any form without prior authorization.



� �

� �

R&D Challenges
� Sensing and Communication � Early Fault Detection and System V&V � Systems Integration and Interoperability � Security (from embedded... to endtoend)

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Strategic Goals:
� � � � � �

Enabling a Resilient, Stronger & Smarter Grid
Isolate the network Fortify the network Reduce the attacker pool Assume defenses will fail Reduce human error Sensing, Communications, Controls, Security, Energy Efficiency and Demand Response if architected correctly could assist SG development:
� Distributed Control � Grid Architectures � Cyber Security
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Selected References
Downloadable at: http://umn.edu/~amin

� Special Issue of Proceedings of the IEEE on Energy Infrastructure Defense Systems, Vol. 93, Number 5, pp. 8551059, May 2005 � Special issues of IEEE Control Systems Magazine on Control of Complex Networks, Vol. 21, No. 6, Dec. 2001 and Vol. 22, No. 1, Feb. 2002 � "Complex Interactive Networks/Systems Initiative (CIN/SI): Final Summary Report", Overview and Summary Final Report for Joint EPRI and U.S. Department of Defense University Research Initiative, EPRI, 155 pp., Mar. 2004 � "New Directions in Understanding Systemic Risk", with NAS and FRBNY Committee, National Academy of Sciences and Federal Reserve Bank of NY, Mar. 2007
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Selected References
Downloadable at: http://umn.edu/~amin
� � � "A Control and Communications Model for a Secure and Reconfigurable Distribution System," (Giacomoni, Amin, & Wollenberg), IEEE American Control Conf., June 2011 "Securing the Electricity Grid," (Amin), The Bridge, the quarterly publication of the National Academy of Engineering, Volume 40, Number 1, Spring 2010 "Preventing Blackouts," (Amin and Schewe), Scientific American, pp. 6067, May 2007 "New Directions in Understanding Systemic Risk", with NAS and FRBNY Committee, National Academy of Sciences and Federal Reserve Bank of NY, Mar. 2007 "Powering the 21st Century: We can and must modernize the grid," IEEE Power & Energy Magazine, pp. 9395, March/April 2005 Special Issue of Proceedings of the IEEE on Energy Infrastructure Defense Systems, Vol. 93, Number 5, pp. 8551059, May 2005 "Complex Interactive Networks/Systems Initiative (CIN/SI): Final Summary Report", Overview and Summary Final Report for Joint EPRI and U.S. Department of Defense University Research Initiative, EPRI, 155 pp., Mar. 2004 "North American Electricity Infrastructure: Are We Ready for More Perfect Storms? ," IEEE Security and Privacy, Vol. 1, no. 5, pp. 1925, Sept./Oct. 2003 "Toward SelfHealing Energy Infrastructure Systems," cover feature in IEEE Computer Applications in Power, pp. 2028, Vol. 14, No. 1, January 2001
� 2011 No part of this presentation may be reproduced in any form without prior authorization.

� � �

� �

� 2011 No part of this presentation may be reproduced in any form without prior authorization.

Subscribe
x