IEEE Power & Energy Magazine - July/August 2020 - 44

Resilience Level

equal to US$10. In the second option, the consumer pays
US$50 for a network service that fails more often and with
larger amounts of ENS each time, totaling an associated
expected cost equal to US$50. Note that, in both cases, the
total amount paid by the consumer for the network and the
EENS is the same. In this context, the consumer is said to be
✔ risk neutral if the consumer is indifferent to these two
options
✔ risk averse if the consumer prefers the first option
over the second one
✔ risk seeking if the consumer prefers the second option
over the first one.
The reader can easily deduce that risk neutrality is intrinsically a part of traditional probabilistic analysis because
the aforementioned two options seem equally attractive. In
reality, however, consumers are typically risk averse and,
arguably, prefer a more stable and predictable outcome from
the electricity network, even if this may (slightly) increase
cost, as mentioned previously. In fact, as in many other
industries, consumers may be willing to pay the price of
insurance policies that eliminate (or at least mitigate) the
losses associated with some rare but high-impact scenarios
that may occur. In this context, risk-averse electricity consumers (which we argue represent the majority) may prefer
to be hedged against the consequences of HILP events on
their electricity supply and pay for the corresponding cost
increase even if these events occur rarely or may not happen at all.
Apart from very extreme cases such as earthquakes,
which may be life-threatening and for which higher risk
aversion may be justified for different reasons in any case,

Predisturbance
Resilient State

evidence of such a consumer attitude can be seen more and
more often even for relatively smaller-impact events. For
example, heat waves in Australia in January 2019 led to sporadic, rolling load-shedding events in several areas, including central Melbourne. The relatively short outages were
considered outrageous by many consumers, even though
they experienced only a minor overall adverse impact.
In addition to consumer attitudes, governments must
consider the welfare of their citizens and, understandably,
may want to take a risk-averse approach in dealing with
HILP events for political reasons, irrespective of the classical economics associated with traditional power system
planning methodologies. That is why specific regulatory and
market mechanism responses were called for following the
South Australia black system event of September 2016, and
more are expected in response to the bushfires that occurred
again throughout Australia in January 2020. In any case,
once again, the main message here is that the risk-averse
approaches and metrics that should be contemplated for
HILP events are typically not present in current system planning practices.

Recognizing the Outage-and-Restoration
Evolution: The Need for Time Domain
Modeling
One important aspect of resilience is its time-varying nature.
The concept of resilience includes the phases before and during a severe event as well as after the event, when the system
recovers. In this context, Figure 2 shows the time-varying,
multiphase resilience trapezoid, which clearly highlights the
phases of a power system when exposed to extreme events,

Phase 1

Phase 2

Phase 3

Disturbance
Progress

Postdisturbance
Degraded State

Restorative
State

Postrestoration
State

Ro

Resilience Trapezoid

Rpd
to
Time
Sequence
Type of
Actions

toe

toe

tee

Event Hits the
Network
Preventive

tr

End of Event

Corrective

Emergency
Coordination

T

Restoration Is
Initiated

Time

End of
Restoration

Restorative

Adaptive

figure 2. A time-varying, multiphase resilience trapezoid. (Source: Panteli et al.)
44

ieee power & energy magazine

july/august 2020
IEEE Power & Energy Magazine - July/August 2020

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - July/August 2020
