IEEE Power & Energy Magazine - November/December 2021 - 30

Early tools did not incorporate chronological
grid operations and instead only
evaluated single peak load periods.
transmission, and load. By applying statistical tools to evaluate
these uncertainties, grid planners can calculate the risk
of shortfalls. Risk metrics can then be used to determine
how much investment our power grids require, how much
new generation should be built, and which can retire.
Many of the RA methods and metrics the industry uses today
originated in the second half of the 20th century. Early tools did
not incorporate chronological grid operations and instead only
evaluated single peak load periods. They often assumed static
load requirements and did not consider the energy limitations
of most resources.
However, growing shares of VRE and energy storage,
in creased load flexibility, and fossil retirement trends are all precipitating
a change in the way RA analysis should be conducted.
To modernize the old framework of RA analysis, new data,
methods, and metrics to better characterize evolving risks
are needed to address the following two driving factors: the
importance of chronological grid operations, and the recognition
of correlated events.
1) Chronological grid operations: Conventional RA analysis
often evaluates only individual peak load hours
and does not consider the full year of operation. This
approach has two assumptions. First, it presupposes the
highest risk period occurs only during peak load periods.
Second, it assumes that there are no sequential operating
characteristics of resources that can limit availability.
The two assumptions can no longer hold because
of the following new development:
* Integration of a large amount of VRE: As VRE increases
its share of the power system's supply, there
is increased dependence on the underlying weather.
Assessing RA with this changing generation mix
requires specific attention to hourly, seasonal, and
interannual resource variability.
Austin
Houston
San
Antonio
31 January 2021, Texas, USA NOAA-20
Day Night Band
(a)
* Deployment of the energy-limited resources: The
capacity of energy-limited resources (ELRs), such
as energy storage systems and demand response
programs, is growing rapidly. ELRs can shift load
from the peak load period to the off-peak periods,
and therefore reduce the reliability risk in the peak
load hours while increasing the risk in others. Accurately
understanding their reliability contribution
requires chronological modeling and scheduling of
the system.
* Emergence of the hybrid resources: Hybrid resources
consisting of different types of generation resources
are becoming increasingly favored among plant developers
due to the high degree of flexibility and the
high-quality grid services they can provide. But they
also raise questions about prioritization of RA uses
relative to other potential services and revenue streams
that may have conflicting use cases.
Austin
Houston
San
Antonio
16 February 2021, Texas, USA NOAA-20
Day Night Band
(b)
figure 1. ERCOT rolling blackouts viewed from space,
February 2021. (Source: NOAA; used with permission.)
30
ieee power & energy magazine
* Increased load-side flexibility: New levels of flexibility
in loads-whether via demand response, electrification
of transportation, time of use rates, or
real-time pricing-allow power system balancing to
occur on both sides of the supply- demand balance.
2) Correlated events: While historical RA analysis focused
on probabilities of discrete independent mechanical
or electrical failures (modeled with randomly
occurring forced outages), correlated events should
now be recognized as a driving factor of reliability.
* Weather fundamentals: There is also growing recognition
that RA risks are driven by other correlated
weather events. Weather data must also consider
geographic diversity, periods of sustained low wind
november/december 2021
IEEE Power & Energy Magazine - November/December 2021

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - November/December 2021
