METHODOLOGY FOR IDENTIFYING NEAR-OPTIMAL INTERDICTION STRATEGIES FOR A POWER TRANSMISSION SYSTEM

1
METHODOLOGY FOR IDENTIFYING NEAR-OPTIMAL
INTERDICTION STRATEGIES FOR A POWER TRANSMISSION
SYSTEM

• Vicki M. Bier, Eli Robert Gratz, Naraphorn J.
  Haphuriwat, and Wairimu Magua
• Department of Industrial and Systems Engineering
• University of Wisconsin-Madison
• Kevin R. Wierzbicki
• Department of Electrical and Computer Engineering
• University of Wisconsin-Madison

2
Objectives

• The objectives of the project are to
• Develop a simple, inexpensive, and practical
  method for identifying promising interdiction
  strategies
• Compare our method and results with those of
  other proposed approaches for vulnerability
  assessment
• Study the effectiveness of protecting
  transmission lines

3
System Topology

• We use the IEEE Reliability Test
• System 1996 (RTS-96)
• Representative of typical systems
• We base our analysis on decoupled
• load (DC) flow with optimal dispatch

4
(No Transcript)
5
System Topology (continued)

• We model the RTS-96 systems
• as networks consisting of
• 24 nodes and 38 arcs for the One Area RTS-96
• 48 nodes and 79 arcs for the Two Area RTS-96

6
Schematic View of Process
Load-Flow Algorithm (Determine optimal DC power
dispatch)
Max Line Interdiction Algorithm (Interdict the
line with maximum flow, and any lines in close
geographical proximity)
Hardening Algorithm (Make the first n sets of
interdicted lines from the Max Line algorithm
invulnerable)
Terminate (after a pre-determined number of
iterations)
7
Other Approaches

• The method of Apostolakis and Lemon (2005)
  applies only to distribution networks (with
  one-directional flows)
• Salmeron et al. (2004) use a non-linear nested
  optimization method that is difficult to solve

8
Results (One Area RTS-96)
Attacked33 Load shed 56
Attacked11 Load shed 44
9
Results (Two Area RTS-96)
45
44
10
Results contd

• The Max Line interdiction strategy reasonably
  approximates the load shed by Salmeron et al.
• The transmission lines interdicted by Salmeron et
  al. differ from those interdicted by our strategy

Salmeron
MaxLine
2119 22 24 2728 30 3839 6159 62 69 7279 7778
64 30 1921 7878 23 41 52 11 7473 3435 21
11
Results (Random Interdiction)
12
Hardening

• We apply the hardening algorithm to
• simulate an upgrade of the system
• H0 represents the original interdiction
• strategy
• H1, H2, and H3 show the interdiction
• strategies obtained after three iterations
• of hardening

13
Results (One Area RTS-96)
Strategy H0 results in a loss of 56 Strategy
H3, hardening 39 of all lines, results in a loss
of 42
14
Results (Two Area RTS-96)
Strategy H0 results in a loss of 56 Strategy
H3, hardening 39 of all lines, results in a loss
of 39
15
Observations
Our results cast doubt on the claim by Salmeron
et al. By considering the largest possible
disruptions, our proposed plan will be
appropriately conservative Hardening even a
significant percentage of lines does not
dramatically diminish the load shed by an attack
Hardening seems unlikely to be cost effective!
16
Conclusions

• We developed a simple, inexpensive, and viable
  method of identifying promising attack strategies
• Our results are comparable to those of Salmeron
  et al.
• A single run of either method will not be
  sufficient to identify critical vulnerabilities
• Hardening of transmission lines is unlikely to be
  cost effective

17
Directions for Future Research

• In future research, this method could be extended
  to
• Address other components of transmission systems,
  such as transformers
• Identify strategies that may trigger cascading
  power failures
• Take into account the importance of different
  loads
• Apply to other types of systems, such as
  structures, water, and transportation

18
Acknowledgement

• This material is based upon work supported in
  part by
• The U.S. Army Research Laboratory and the U.S.
  Army Research Office under grant number
  DAAD19-01-1-0502
• The National Science Foundation under grant
  number ECS-0214369
• The Department of Homeland Security under grant
  number EMW-004-GR-0112
• Any opinions, findings, and conclusions or
  recommendations expressed
• in this material are those of the authors and do
  not necessarily reflect
• the views of the sponsors.
• The authors would like to thank Prof. Ian Dobson
  of the University of
• Wisconsin-Madison for his contributions to this
  study.