Operational Defense of Power System Cascading Outages

Similar documents
Risk of Cascading Outages

Cascading Outages in Power Systems. Rui Yao

Reliability of Bulk Power Systems (cont d)

Power System Security. S. Chakrabarti

6. The August 14 Blackout Compared With Previous Major North American Outages

Emergency response system for electric power systems

FAST DETECTION AND MITIGATION OF CASCADING OUTAGES IN THE POWER SYSTEM. A Dissertation CHENGZONG PANG

Section 5. TADS Data Reporting Instruction Manual DRAFT Section 5 and Associated Appendices With Proposed Event Type Numbers

Sample average approximation and cascading failures of power grids. Daniel Bienstock, Guy Grebla, Tingjun Chen, Columbia University GraphEx 15

POWER SYSTEM STABILITY AND CONTROL

11.1 Power System Stability Overview

UNITED STATES OF AMERICA BEFORE THE FEDERAL ENERGY REGULATORY COMMISSION. Interconnection for Wind Energy ) Docket No. RM

Chapter 9: Transient Stability

PUB NLH 185 Island Interconnected System Supply Issues and Power Outages Page 1 of 9

PMU-Based Power System Real-Time Stability Monitoring. Chen-Ching Liu Boeing Distinguished Professor Director, ESI Center

Identifying Plausible Harmful N-k Contingencies: A Practical Approach based on Dynamic Simulations

International Workshop on Wind Energy Development Cairo, Egypt. ERCOT Wind Experience

Power System Stability and Control. Dr. B. Kalyan Kumar, Department of Electrical Engineering, Indian Institute of Technology Madras, Chennai, India

EXPECTATIONS FOR LOSS OF SUPPLY IN THE NEW ZEALAND POWER SYSTEM

Renewables and the Smart Grid. Trip Doggett President & CEO Electric Reliability Council of Texas

Wind power and management of the electric system. EWEA Wind Power Forecasting 2015 Leuven, BELGIUM - 02/10/2015

Transmission and Dispatching Operations Manual

Power Grid Vulnerability to Geographically Correlated Failures - Analysis and Control Implications

Multi Hazard Evaluation of a High Voltage Transmission Network. John Eidinger 1 and Leon Kempner 2

Blackouts in electric power transmission systems

THE DETECTION, PREVENTION AND MITIGATION OF CASCADING OUTAGES IN THE POWER SYSTEM. A Dissertation HONGBIAO SONG

URTDSM Initiative in India and Controlled Islanding using PMU Measurements

Module 6 : Preventive, Emergency and Restorative Control. Lecture 27 : Normal and Alert State in a Power System. Objectives

EE2351 POWER SYSTEM OPERATION AND CONTROL UNIT I THE POWER SYSTEM AN OVERVIEW AND MODELLING PART A

Performance Evaluation and Review of System Protection Scheme Design with the help of Synchrophasor Measurements

NORDIC AND BALTIC GRID DISTURBANCE STATISTICS 2015

Power System Seminar Presentation Wind Forecasting and Dispatch 7 th July, Wind Power Forecasting tools and methodologies

Transmission and Dispatching Operations Manual

Electric Distribution Storm Hardening Initiatives. Paul V. Stergiou Distribution Engineering October 14 th, 2015

1998 System Disturbances

Research on the Effect of Substation Main Wiring on the Power System Reliability Yuxin Zhao 1, a, Tao Liu 2,b and Yanxin Shi 3,c

Power System Analysis Prof. A. K. Sinha Department of Electrical Engineering Indian Institute of Technology, Kharagpur

Case Study on Fault Analysis Using PMU

SINGLE OBJECTIVE RISK- BASED TRANSMISSION EXPANSION

Criticality in a cascading failure blackout model

ECE 522 Power Systems Analysis II 3.3 Voltage Stability

Cascading Failures in Power Grids - Analysis and Algorithms

EE 6501 POWER SYSTEMS UNIT I INTRODUCTION

A STATIC AND DYNAMIC TECHNIQUE CONTINGENCY RANKING ANALYSIS IN VOLTAGE STABILITY ASSESSMENT

Peterborough Distribution Inc Ashburnham Drive, PO Box 4125, Station Main Peterborough ON K9J 6Z5

DESIGN OF AN ADAPTIVE DYNAMIC LOAD SHEDDING ALGORITHM USING NEURAL NETWORK IN THE STEELMAKING COGENERATION FACILITY *

Role of Synchronized Measurements In Operation of Smart Grids

The Study of Overhead Line Fault Probability Model Based on Fuzzy Theory

Norm Hann Performance Manager, Performance Management Department Hydro One Networks Inc.

ECE 476. Exam #2. Tuesday, November 15, Minutes

California Independent System Operator (CAISO) Challenges and Solutions

Mitigation of Distributed Generation Impact on Protective Devices in a Distribution Network by Superconducting Fault Current Limiter *

ANALYSES OF POWER SYSTEM VULNERABILITY AND TOTAL TRANSFER CAPABILITY XINGBIN YU

ECE 422/522 Power System Operations & Planning/ Power Systems Analysis II 3 Load Modeling

Distributed quasi-newton method and its application to the optimal reactive power flow problem

Harmonic Modeling of Networks

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 2, MAY Definition and Classification of Power System Stability

Bulk Power System Reliability Assessment Considering Protection System Hidden Failures

ISO NE Experiences with Internal and External Model Impacts on the State Estimator

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 29, NO. 4, JULY Stochastic Analysis of Cascading-Failure Dynamics in Power Grids

Wide Area Measurement Applications for Improvement of Power System Protection

W I N T E R STORM HAZARD DESCRIPTION

B.E. / B.Tech. Degree Examination, April / May 2010 Sixth Semester. Electrical and Electronics Engineering. EE 1352 Power System Analysis

Study of Time Correlation Between Lightning Data Recorded by LLS and Relay Protection

KINGS COLLEGE OF ENGINEERING Punalkulam

Real-Life Observations of Power System Dynamic Phenomena Some Interesting Aspects

1. Explain the various methods of methods of grounding. In power system, grounding or earthing means connecting frame of electrical equipment (non-cur

Stochastic Unit Commitment with Topology Control Recourse for Renewables Integration

Adaptive Distance Relaying Scheme for Power Swing Tripping Prevention

Applications of GIS in Electrical Power System

OPALCO Outage Analysis. December 2017 Board Meeting

NOWADAYS, power systems are strongly changing all

THIRD GENERATION SYSTEM PROTECTION SCHEME PROJECT FOR ZEYA HYDRO POWER PLANT

1st Annual Southwest Ohio Snow Conference April 8, 2010 Abner F. Johnson, Office of Maintenance - RWIS Coordinator

Sensitivity-Based Line Outage Angle Factors

Risk Analysis for Assessment of Vegetation Impact on Outages in Electric Power Systems. T. DOKIC, P.-C. CHEN, M. KEZUNOVIC Texas A&M University USA

SCHOOL EMERGENCY MANAGEMENT PLAN FOR NATURAL DISASTERS

ECE 422/522 Power System Operations & Planning/Power Systems Analysis II : 8 - Voltage Stability

Assessment of Interruption Costs in Electric Power Systems using the Weibull-Markov Model

Dynamic Decomposition for Monitoring and Decision Making in Electric Power Systems

Grid Issues and Challenges Addressed by High Temperature Superconductor (HTS) Technology

Severe Ice Storm Risks in Ontario

New Method of Capacitors Failure Detection and Location in Shunt Capacitor Banks

BVES Annual Reliability Public Presentation 2016 Performance

Adaptive under frequency load shedding using synchrophasor measurement

Evaluating the Impact of Modeling Assumptions for Cascading Failure Simulation

Assessment, Planning and Control of Voltage and Reactive Power in Active Distribution Networks

Micro-grid to System Synchronization Based on Pre-Insertion Impedance Method (Version 1.0) By Peter Zhou University of Alberta Jan 30 th, 2015

ECEN 667 Power System Stability Lecture 20: Oscillations, Small Signal Stability Analysis

1. Introduction. Keywords Transient Stability Analysis, Power System, Swing Equation, Three-Phase Fault, Fault Clearing Time

Proper Security Criteria Determination in a Power System with High Penetration of Renewable Resources

Structural Vulnerability Assessment of Electric Power Grids

Common Cause Failure (CCF)

August A report for the National Electricity Market

An MINLP approach to forming secure islands in electricity networks

National Wildland Significant Fire Potential Outlook

Superconducting Fault Current Limiters

Mid-Atlantic Severe Weather Event of 23 June 2015

3VT4 Molded Case Circuit Breakers up to 1000 A

Low Frequency Transients

Transcription:

Operational Defense of Power System Cascading Outages James D. McCalley Professor, Iowa State University Cascading Failures & Blackouts April, 008 IEEE PES T&D Conference and Exposition Overview. Summary of previous blackouts. Blackout attributes. Approaches to reduce frequency/severity of high consequence events. Emergency Response System. Triggering events. Simulator attributes. Final comments

. :0 Conesville Unit, MW. : Greenwood Unit, 8 MW. : Eastlake Unit, MW, overexcitation Weakening conditions. :0 Stuart Atlanta kv (brush fire) Triggering. :0 Harding-Chamberlain kv (tree) event. : Hanna-Juniper kv (tree). : Star-South Canton kv (tree) SLOW 8. :-:0 8 KV lines around Akron tripped PROGRESSION. :0 Sammis-Star kv (zone ) AUGUST, 00. :0 Conesville Unit, MW. : Greenwood Unit, 8 MW. : Eastlake Unit, MW, (overexcitation) WHAT HAPPENED ON AUGUST, 00??? Weakening conditions. :0 Stuart Atlanta kv (brush fire) Triggering. :0 Harding-Chamberlain kv (tree) event. : Hanna-Juniper kv (tree). : Star-South Canton kv (tree) SLOW 8. :-:0 8 KV lines around Akron tripped (overload PROGRESSION and failed). :0 Sammis-Star kv (zone, tree) 0. :08:8 Galion-Ohio Central-Muskingum kv (zone ). :0:0 East Lima-Fostoria Central kv (zone ). :0:-:0: Kinder Morgan (rating: 00 MW; loaded to 00 MW). :0 Harding-Fox kv. :0:0 :0: 0 generators along Lake Erie in north Ohio, MW. :0: West-East Michigan Kv (zone ) FAST. :0:8 Midland Cogeneration Venture, MW (reduction of 00MW) PROGRESSION. :0:8 Transmission system separates northwest of Detroit 8. :0:8 Perry-Ashtabula-Erie West kv (zone ) (cascade). :0:0 :0: lines disconnect between Pennsylvania & New York 0. :0: lines disconnect and gens trip in north Ohio,88MW. :0: :0: lines disconnect in north Ontario, New Jersey, isolates NE part of Eastern Interconnection, unit trips, 80 mw. :0: :0: New York splits east-to-west. New England and Maritimes separate from New York and remain intact. (power swing+ufls). :0:0 :: Ontario separates from NY w. of Niagara Falls & w. of St. Law. SW Connecticut separates from NY,blackout.(relay operation,ulfs)

IMPACT Location Date MW Lost Duration People affected Approximate cost US-NE // 0000 hours 0 million US-NE // 000 hours million 00 million France //8 0000 0 hours West Coast //8 0 million Sweden //8 > 000. hours. million Brazil /8/8 Brazil 8/8/8 Hydro Quebec /8/88 800 US-West // 00 Brazil // 80 US-West //. million Brazil // US-West //. million US-West // 00 small number US-West 8/0/ 08. million billion dollars MAPP, NW Ontario //8 0 hours 0. million San Francisco /8/8 00 8 hours million Brazil // 000 hours million Brazil // 000 India //00 000 hours 0 million 0 million Rome //00 0. million US-NE 8//00 000 - days 0 million - billion Denmark/Sweden //00 00. hours million Italy /8/00 000. hours million Croatia //00 0 mwh. million Greece //00 000 hours million Moscow/Russia /-/00 00 > hours million European Blackout /00 > 00 Hour million PRE-EVENT CONDITIONS Location Date Weather Loading Topology US-NE // mild normal normal US-NE // Stormy normal weakened ( major tie feeder, major gen out) France //8 Heavy Normal West Coast //8 windy normal normal Sweden //8 Brazil /8/8 Brazil 8/8/8 Hydro Quebec /8/88 Freezing rain normal normal US-West // mild normal normal Brazil // US-West // cold Heavy normal Brazil // US-West // Hot 8C Heavy Normal US-West // Hot 8C Stressed Normal US-West 8/0/ Hot 8C normal weakened (three 00 KV line sections out of service MAPP, NW Ontario //8 stormy heavy normal San Francisco /8/8 normal normal normal Brazil // Brazil // India //00 Rome //00 Hot heavy weakened US-NE 8//00 heavy Weakened ( gens out of service) Denmark/Sweden //00 heavy Weakened ( nuclear unit out for maintenance) Italy /8/00 heavy Weakened (trip of Swiss 80 KV line Mettlen-Lavorgo) Croatia //00 wind,cold,ice, rain normal weakened Greece //00 Hot Heavy weakened( 0KV, a MW & 00MW unit out) Moscow/Russia /-/00 Hot Heavy Weakened (loss of a cogen plant) European Blackout //00 normal weakened as no of power plants shut down

TRIGGERING EVENTS US-NE // Faulty Relay setting N- US-NE // Lightening N- France //8 West Coast //8 00 KV Tr tower failed due to high winds N- Sweden //8 Disconector Failed N- Brazil /8/8 Xmer shutdown due to overload, and load increase N- Brazil 8/8/8 phase to grd short ckt+ in-advertent protection operation N- Hydro Quebec /8/88 Ice causes flashover N- US-West // Earthquake many Brazil // human error D.C. bipoles blocked US-West // Single phase to gnd fault, relay misop. N- (inadvertent of additional KV ckt) Brazil // human error+inadvertent prot. operation N- US-West // Tree Flashover followed by relay misop. N- US-West // Tree Flashover N- US-West 8/0/ Tree Flashover N- MAPP, NW Ontario //8 lightening N- San Francisco /8/8 human error no of lines Brazil // Bus Fault Multiple lines (> N-) Brazil // Inadvertent protection operation Many India //00 Rome //00 high load demand US-NE 8//00 Brush fire on a line (outage) N- Nuclear Plant trips (technical problem), double Denmark/Sweden //00 busbar fault N- Italy /8/00 Tree Flashover N- Croatia //00 Breaker failure N- Greece //00 Load Increasing N- Moscow/Russia /-/00 Load Increasing/Xmer bursting European Blackout //00 human error many PRE-COLLAPSE EVENTS Location Date Generation trip Transmission trip Time between initiating and secondary, precollapse events US-NE // no Four 0KV lines few minutes US-NE // Yes Yes occurred in a sequence between 0 to minutes after initial event France //8 yes > 0 minutes West Coast //8 No Yes Fast Sweden //8 No Yes 0 seconds Brazil /8/8 Xmer yes -0 minutes Brazil 8/8/8 No yes Hydro Quebec /8/88 Transformer yes - seconds US-West // Yes Yes Fast Brazil // yes yes US-West // No Yes 0- seconds Brazil // Xmer Yes US-West // yes yes 0 seconds US-West // No yes fast US-West 8/0/ yes ( generators) yes - minutes MAPP, NW Ontario //8 No yes minutes San Francisco /8/8 yes yes seconds Brazil // No Yes > 0 seconds Brazil // No Yes India //00 Rome //00 No No US-NE 8//00 yes yes more than hours Denmark/Sweden //00 yes yes minutes Italy /8/00 No Yes minutes Croatia //00 Yes Yes 0 seconds Greece //00 Yes No 0 minutes Moscow/Russia /-/00 No Yes > hours European Blackout //00 yes Yes 0 minutes 8

Location Date Causes of secondary, pre-collapse events US-NE // Proper protection operation (as designed) (overload protection) US-NE // Lightening, Proper protection operation (overload+gen protection) NATURE OF COLLAPSE France //8 Proper protection operation (overload protection,out of step relays) West Coast //8 Primary and secondary protection & communication failure Sweden //8 Proper protection operation (overload protection), underfreq LS failure Brazil /8/8 Simultaneous tripping of ckts and Xfmer Brazil 8/8/8 Protection failure (SPS setting) Hydro Quebec /8/88 Communication failure followed by load shedding protection failure US-West // Earthquake Brazil // Inefficient Protection, loss of synchronism US-West // Proper protection operation (overload protection) Brazil // Proper protection operation US-West // Proper protection operation (gen protection), relay misoperation US-West // Relay Misoperation US-West 8/0/ Trees, protection (relay) failure MAPP, NW Ontario //8 Lightening trip another kv line followed by proper ovrload protection San Francisco /8/8 No local protection, topology,delayed remote protection Brazil // Proper protection operation ( overload protection ) Brazil // Inadvertent Protection operation India //00 Rome //00 High Load, low generation, reduction in import US-NE 8//00 Proper protection operation Denmark/Sweden //00 Switching device breaks,proper protection operation (generator and overload protection) Italy /8/00 Unsuccessful reclosing, Tress, loss of synchronism, dynamic intercaction leading to voltage collapse Croatia //00 Protection failure Greece //00 Proper protection operation Moscow/Russia //00 lines from HV substation tripped due to faults and overloading European Blackout //00 Proper protection operation COLLAPSE TIME & NO. OF SUCCESSIVE EVENTS Location Date Collapse time #successive events US-NE // minutes Many US-NE // hour Many France //8 > 0 minutes Many West Coast //8 few minutes Many Sweden //8 > minute Many Brazil /8/8 > 0 minutes Topology Brazil 8/8/8 Topology Hydro Quebec /8/88 < minute Many US-West // minute Brazil // many US-West // substation topology Brazil // Topology US-West // seconds Several US-West // > minute Prevented by fast op. action US-West 8/0/ > minutes Many MAPP, NW Ontario //8 > minutes substation topology San Francisco /8/8 seconds many Brazil // 0 seconds substation topology Brazil // Topology India //00 Rome //00 US-NE 8//00 > hour Many Denmark/Sweden //00 minutes Many Italy /8/00 minutes Many Croatia //00 few seconds many Greece //00 minutes few Moscow/Russia /-/00 hours Many European Blackout //00 0 minutes Many 0

Summary of blackout attributes Impact: of largest blackouts occurred in last 0 years # of blackouts > 000 MW doubles every 0 years Pre-event conditions: Extreme weather Extreme conditions Weakened topology Triggering events: Various kinds of N- or N-k (k>) with fault + nearby protection failure Summary of blackout attributes Pre-collapse events: 0% involved generation, % involved transmission 0% had significant time between initiating & precollapse events 0% involved proper protection action Nature of collapse: Successive tripping of components and/or Voltage collapse Collapse time and # of events: 0% were slow 0% involved many cascaded (dependent) events

Scenario for 0% of blackouts. Weakened conditions: Heavy load, and/or one or more gen or cct outage possibly followed by readjustments. Initiating event: One or several components trip because of fault and/or other reasons;. Steady-state progression (slow succession): a. System stressing: heavy loading on lines, xfmrs, gens b. Successive events: Other components trip one by one with fairly large inter-event time intervals. Transient progression in fast succession: a. Major parts of system goes under-frequency and/or under-voltage. b. Components begin tripping quickly c. Uncontrolled islanding and wide spread blackout. Preventive/corrective action paradigm Probability>P (Class B events) Initiating event identification and probability calculation Emergency response system (ERS) Probability<P (Class C, D events) Highprobability events Assessment Lowprobability events Assessment Violation detection Determine preventive action and implement it Failure detection Determine corrective action and store it

Operational approach to blackout mitigation Preventive control is for high -probability events. Corrective control is for not-high (N-k) probability events. Blackouts typically result from the latter. Corrective control: operational solution to blackouts. Actions determined on-line, in anticipation of events, to be actuated through operator upon event occurrence Need new EMS Function ERS: decision support to unfolding events Operational approach to blackout mitigation The Emergency Response System: Continuously identifies catastrophic event sequences together with actions operators can take to mitigate them.. Intelligent selection of triggering events Uses current or forecasted conditions Based on mid-term (hours) simulation tool Must have protective relaying modeled Stores results for fast retrieval should event occur 8

High-Risk Triggering Events. Functional group tripping Proper relay tripping, may trip multiple components. Fault plus breaker failure to trip Breaker stuck or protection fail to send signal to open Two neighboring functional groups tripped. Inadvertent tripping of or more components Inadvertent tripping of backup breaker to a primary fault. Common mode events Common right of way, common tower.. Any of above together with independent outage of any other single component Contingency Selection Definition: A functional group is a group of components that operate & fail together as a result of breaker locations within the topology that interconnects them. Graph search identifies five types of initiating events:. Functional group tripping, ~ P Proper relay tripping, may trip multiple components. Fault plus breaker failure to trip, ~ P Breaker stuck or protection fail to send the signal to open Two neighboring functional groups tripped. Inadvertent tripping of two or more components, ~ P Inadvertent tripping of backup breaker to a primary fault. Common mode events, ~ P Common right of way, common tower.. Any of the above together with independent outage of any other single component in a selected set, ~ P ~ P 8

0 0 0 0 0 0 8 0 0 0 8 08 8 8 8 0 0 0 0 8 8 8 0 0 0 8 8 0 8 8 0 0 0 0 8 8 0 0 0 8 8 8 Contingency Selection L0 TG0 G0 BUS-0 L0 BUS-0 T0 LOAD0 BUS-0 L0 L0 BUS-0 T0 L0 L08 L0 BUS-0 BUS-0 LOAD0 T0 L0 L L L0 L0 L0 BUS-0 L0 BUS-0 L L L G0 TG0 BUS-0 Number of contingencies of type N-k resulting from a single fault (Order P) for Bus System k No. L0 BUS-0 L0 L8 L BUS-08 0 T0 BUS-0 G0 L0 BUS-0 L0 TG0 0 0 LOAD0 BUS-0 BUS-0 0 0 0 0 0 0 0 08 L0 L0 L0 8 L TG0 G0 L0 L08 L0 L8 L L G0 BUS-0 BUS-0 0 TG0 BUS-0 BUS-0 0 LOAD0 L0 8 0 L0 L 8 L L0 L0 LOAD0 8 0 L L L BUS-08 BUS-0 0 8 TG0 BUS-0 L0 G0 LOAD0 L0 L0 L0 Number of contingencies of type N-k resulting from a fault/breaker failure (order P ) for Bus System k No. Number of contingencies of type N-k resulting from ITC (Order P ) for Bus System k No. 8 Testing on Large EMS Model Use graph-search to identify functional groups, and order P and P contingencies Number of components in the system Type Bus Line Xfmr Gen Shunt No. 80 Number of contingencies of type N-k resulting from a single fault (Order P) k 8 0 No. 0 8 0 0 Number of contingencies of type N-k resulting from a fault/breaker failure (order P ) k 0 No. 0 8 0 0 0 Searching Time to Identify Contingencies (.Ghz Pentium) System Scale (No. of Buses),000 0,000 Time (Second) 0.0 0..08. 0 0

8 0 8 0 8 0 0 0 8 Simulator. Seamless interface with & simulation on node-breaker model for proper identification of initiating & successive events. Model full range of dynamics: Fast dynamics, including generator, excitation, governor Slow dynamics, including AGC, boiler, thermal loads. Model condition-actuated protection action that trips element Generator: field winding overexcitation, loss of field, loss of synchronism, overflux, overvoltage, underfrequency, and undervoltage Transmission: impedance, zone, out-of-step. Identifies islanded condition and continues simulation in each. Saves & restarts from conditions at any time. Failure detection and prevention. SPEED is essential: Adaptive time step using theta implicit integration method Intelligent Jacobian updating Sparsity-based coding and multi-frontal solver for Ax=b Deploying on an IBM BlueGene supercomputer Simulator is written in C++ ONE-LINE DIAGRAM OF DET TEST SYSTEM BUS-0 BUS-0 G0 (initiating generator trip at t=00s) Load ramping 0% from t=00s to t=00s L0 L0 L0 G0 G0 L0 L0 TG0 BUS-0 0 BUS-0 BUS-0 LOAD0 TG0 0 0 0 0 0 T0 BUS-0 0 0 0 0 8 0 0 BUS-0 L0 L0 L0 8 BUS-0 0 0 0 08 08 L0 0 0 0 0 L0 0 L0 8 L0 L08 L0 L8 L 8 L L0 L08 L0 T0 G0 BUS-0 0 BUS-0 8 TG0 BUS-0 LOAD0 L0 BUS-0 8 8 L0 L0 LOAD0 T0 L0 L L L0 L0 L L 0 0 8 0 0 L L BUS-08 L BUS-0 8 L LOAD0 8 0 L 0 BUS-0 0 L TG0 8 BUS-0 T0 TG0 8 BUS-0 G0 L0 BUS-0 G0 L0 8 L0 L0 LOAD0 BUS-0 L0 0 BUS-0 0 8 8 0 L8 L 8 L 0 8 0 0 BUS-08 8 LOAD0 L0 TG0 G0 L0 BUS-0

Initiating contingency: G0 Without Generator Protection First Control Action at 08. Second Control Action at 0.0 Final Comments on Operational Approach to Blackout Mitigation Number of major blackouts doubles every 0 years Various approaches to reduce frequency, mitigate severity Operators are last line of defense; they need better tools Preparing operators for rare events is fundamental to operating engineering systems having catastrophic potential; it has precedent in air traffic control, nuclear, & process control. Described approach is a generalization of already-existing event-based special protection systems, except here response continuously developed on-line actuation is done through a human