IEEE Power & Energy Magazine - November/December 2019 - 63

Since the introduction of electricity markets, reliability services markets have existed to provide individual price and
quantity schedules for different services needed to maintain
the reliability of the bulk power system. Other reliability services have rules for how and which costs can be recovered.
The number of reliability services has grown, with new services being added by market operators due to new challenges
being introduced by VRE. The design of these reliability
services markets (also referred to as ancillary services) differs across the globe. U.S. markets align them closely with
energy markets using co-optimization and marginal cost
pricing for every market interval, whereas other areas use
contracting and bilateral trading for many services. Markets
may exist for contingency reserve, regulating reserve, frequency response, voltage support, and black-start services.
What has been generally consistent across different markets
is that, although these service markets crucially provide the
incentives to ensure a reliable electric system, they are an
afterthought based on the small amount of revenue actually
earned in them. An important question for any changes to
reliability services market design is whether or not that will
continue to hold or if these markets will become an increasing revenue source for those resources that provide services
as much as they provide energy.
With predominantly low energy prices, increases in
reliability service requirements, and displacement of the
resources that historically provided these services due
to the lower energy prices, revenue from reliability service markets may become a greater proportion of overall
november/december 2019

Comparison Across ISOs
900
800

0.75

700
600
0.5

500
400
300

0.25

200

Intervals (%)

Reliability Services of the Future

revenue sources. However, if the service is abundant and
can be provided cheaply, this may not be the case, no
matter how crucial that service is. For example, a single
bolt on an airplane is very crucial, and its absence could
be catastrophic. However, the bolt is still valued at just a
few cents. If the services become scarce or more expensive
to produce, this can change the paradigm. Again, the key
will be setting shortage price levels and triggering points
to send the appropriate price signals or allowing consumers to better direct prices. At high VRE penetration levels,
certain reliability services become more valuable: inertia,
fast frequency response (FFR), frequency regulation, and
reserves to cover against renewable forecast error. Some of
these were in plentiful supply with yesterday's generation
portfolio. Now, we need to make sure the market includes
compensation for scarcity providers of reliability services
going forward.
Reliability service markets have different definitions,
requirements, and eligibility rules. Some definitions are
somewhat archaic, others are based on existing characteristics other than explicit needs, and still others may aggregate multiple services as one to make things simpler. The
first step may be to ensure that the definitions and incentives
are targeted toward the actual service and contribution provided. As they say, if you build it (a market), they (suppliers)
will come-and if you ask for something that is not exactly
what you meant, you will get it. The ERCOT Board of Directors recently approved a redefinition of the set of ancillary
services products to align with the grid's current and future
needs while also staying agnostic to the technology of the

Occurrences of Prices Spikes

studies have defined additional value that DR can provide
to a grid, including load shifting (the movement of energy
consumption from times of high demand to times of surplus generation) and load shimmy (dynamic adjustment
of demand to alleviate short-run ramps and disturbances).
Market design should adapt to exploit the full scope of
value from the demand side. CAISO is considering the
development of a load-shift product, whereby incentives
are provided to behind-the-meter technologies to consume excess renewable energy and supply it back when
it is needed.
There are questions about the market timelines as well.
Some regions are looking at multiple-day-forward markets to better capture the optimal use of fuel from heavy
natural gas use and energy storage. Other regions may see
a decrease in resources that need substantial time to staff
and synchronize to the grid, thus reducing the need for a
day-ahead market. As VRE forecasting errors tend to reduce
significantly as we approach real time, would this reduce the
benefits of day-ahead markets, which increasingly deviate
from real-time conditions? Europe has observed shifts away
from day-ahead markets toward intraday and real-time markets. Do ahead markets need to be more responsive to the
forecasting time frames for VRE?

100
0

0
2014

2015

2016

2017

Year
Zone
ERCOT South
MISO Louisiana
CAISO SP15

NYISO Long Island
SPP South

figure 4. The 5-min energy price spikes across U.S. markets. The dark colors indicate more than US$1,000/MWh;
the light colors indicate more than US$500/MWh. ISO:
independent system operator; MISO: Midcontinent ISO.
ieee power & energy magazine

IEEE Power & Energy Magazine - November/December 2019

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - November/December 2019