IEEE Power & Energy Magazine - November/December 2015 - 63

Meanwhile, if large capital resources are still needed for
long-term adequacy and are earning less in the energy market, they may demand more of the share within the capacity
markets. The result could be a major shift in the revenue mix
for power plants, with less revenue from selling energy and
more from providing capacity and system services.
As renewable penetration increases, where will markets
provide value and for what services? Markets are already
commonly used for system services such as regulation and
reserves; the costs of these services are so closely linked with
energy that this seemed logical. For other services, it may not
be as straightforward to assess whether the market option is
viable or if interconnection standards and requirements are
the better solution. In some cases, a well-functioning market
is not possible, such as when the market would be highly
concentrated or when the service has the quality of a public
good. As shown in Table 1, there may be many reasons why
market mechanisms are not an efficient way to provide a particular system service.
However, if the market is reasonably competitive and
not too costly to implement and administer, then it may be a
more efficient way to secure the service. Extending the reach
of markets to new services often makes sense if the service
is presently costly and supplies can be provided by diverse
sources (perhaps in surprising and innovative ways). As we
learn to better identify the evolving need for additional system services and the evolving characteristics of resources
that might provide that need, we can make logical decisions
about whether services should be obtained through new
market products, location-specific or cost-based procurements, or with grid codes and interconnection standards.
For instance, no power system in North America currently
has a market product for autonomous primary frequency
response. This is a characteristic that was historically provided by the default turbine governor settings of conventional
resources, so there was no perceived need to procure this service. However, for a variety of reasons, including the locking
of governors as the service is not rewarded, some areas have
seen a decline in frequency response over the past decade. As
a result, there is a new reliability standard to ensure that North
America balancing areas have sufficient primary frequency
response, and some markets are now discussing whether frequency response may become a market product.
Markets in New Zealand, Ireland, and Australia have
already introduced different versions of frequency responsive reserve products with good success. The New Zealand
market procures a fast instantaneous reserve product that
must be fully activated within 1 s following a frequency
event and sustained for at least 60 s. Similarly, the Australian
National Electricity Market (NEM) has a contingency service that requires automatic sensing of the frequency event
and providing full response within 6 s. In North America,
the Electric Reliability Council of Texas (ERCOT, the system operator in the state of Texas) requires a frequency
response from wind and is proposing a frequency response
november/december 2015

Today

Future?

Energy
Capacity
System Services

figure 1. A conceptual view of the possible shift from
energy to system services. (Courtesy of EPRI.)

market and a market for demand-side resources that curtail
during significant frequency deviations.
As another example, we are not aware of any system that
has implemented a market or incentive-based reward for generators that provide inertia. Some regions impose requirements;
for instance, Hydro Quebec mandates inertia provision as a
condition of interconnection (including for wind, as a very fast
frequency response in the inertial timeframe using the wind
turbine control system and the kinetic energy of the turbine
drivetrain). Also, some energy markets enforce implicit inertia
and other complex requirements with nomograms (an approximation of a complex function), such as the Southern California
Import Transmission nomogram in California. Other markets,
like Australia, ERCOT, and Ireland, may establish an explicit
inertia service in the next few years. The explicit product may
motivate the units to remain online and provide inertia even
when energy prices provide insufficient incentive to do so.
While increasing amounts of variable generation are certainly a factor in this move toward more market products for
system services, they are not the only driver. Advancements
in software, communications, control systems, and experience with electricity markets may also lead to evolving
designs and mechanisms to procure the services with market products. At the same time, some believe the markets are
getting too complex, with the evolution and introduction of
too many products as being unnecessary. This debate will
continue as new technologies participate.

table 1. Situations and examples where a market
mechanism may not be an efficient solution.
Reasons Why a Market Product
May Not Be Justified

Example

Too complex to design or too specific to
certain local areas

Local voltage
support

System inherently has more than sufficient
amounts of the service

Inertia

Costs for the service are small, so market
product may be overkill

Black start

ieee power & energy magazine

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - November/December 2015