IEEE Power & Energy Magazine - November/December 2015 - 64

Flexibility-A New Product
or Business as Usual?
To meet the changing demand at various timescales,
resources must be able to adjust output, sometimes quite
rapidly. For example, during the morning load pickup, the
supply resources must continuously increase their energy
output. The large pool of generators dispatched by the system operator can inherently provide significant flexibility. At
present, in large systems with a fast real-time balancing market, there are typically many resources in the dispatch that
can contribute toward balancing and ramping at a low cost.
The value of integrating wind and solar into the dispatch
process has also been demonstrated in several markets (see
"Knowledge Is Power" in "For Further Reading").
When faced with increased variable generation, the need
for the system to be able to quickly ramp up or down may
increase. This raises the question of whether market design
evolution is required to ensure that enough flexibility will
be available in the future. While the resources are certainly
being incentivized to provide more energy when needed, it is
unclear whether market designs are incentivizing resources to
provide faster ramping capability and other flexibility attributes. In theory, prices that change frequently, reflect intertemporal optimization of resources schedules, and are not
limited in the up or down direction should reward the ability
of flexible resources to respond rapidly to system conditions.
However, the profitability of flexible resources can be reduced
if settlements are averaged over long intervals (e.g., an hour)
and are subject to price caps and floors and because market
software fails to recognize that steeper than expected ramps
can occur for which preparation is needed.
Providing incentives for ramping flexibility continues to
be a topic of debate. Is it a new product, or does it flow naturally from the normal energy and system services process?
In theory, it should be more profitable for resources to allow
the system operator to dispatch all resources, and the system
operator would then have full access to their flexibility.
However, there can be nontechnical, institutional constraints limiting the flexibility that can be extracted from existing resources. For example, a self-scheduled resource operates
without regard to the optimal dispatch that is calculated and
carried out by the system operator. Any flexibility that this
resource may possess is therefore generally not available to
the market. The reason as to why many flexible resources
choose to self-schedule rather than bid may often be due to
long-term contracts, or the entities may have other "out of
market" considerations that influence their behaviors. Similarly, environmental regulations on hydro flows, air emission
restrictions, run-time constraints, and transmission constraints
can all reduce the effective flexibility of the system. Renewables can be part of the solution rather than the problem if they
are allowed to provide flexibility within dispatch.
Two U.S. independent system operators (ISOs), California ISO (CAISO) and Midcontinent ISO (MISO), are
augmenting their markets with ramp products to schedule
64

ieee power & energy magazine

additional amounts of flexible capacity to meet potential
ramp needs and avoid transient price spikes that occur due
to ramp events. We will call such products "flexiramp" (consistent with CAISO), although they are also known as "ramp
capability" (in MISO). An up-flexiramp (down-flexiramp)
product is defined as the designation of capacity in one time
interval to meet higher (lower) than expected net energy
demands in subsequent intervals, at which time that capacity
can be dispatched to meet those energy demands on a leastcost basis. Presently, the CAISO market only has a real-time
upwards ramp constraint in their market that compensates
capacity that helps meet that constraint, while both markets
have "up and down" products under development.
Flexiramp differs from existing spinning reserve because
spinning reserve is usually reserved for defined contingency
events (such as the unexpected failure of a generator) and
is typically not allowed to be used for other events. In contrast, flexiramp is reserved in one period to meet possible
ramps in future periods and will be dispatched on an economic basis when needed. Thus, flexiramp will be called
on much more often to produce energy, and the payment for
providing flexiramp should cover the "lost opportunity cost"
of reserving the ramping capability rather than selling it in
the energy market.
Not all markets believe such a product is necessary.
PJM, a large U.S. ISO, is content with the ability of its
noncontingent operating reserves to help the system meet
unexpected ramps. While a traditional contingency event
can occur in a fraction of a second, the system-level ramping from wind and solar generally occurs over minutes and
hours. Systems with sufficient resources in their real-time
dispatch may find that their normal dispatch process provides ample ramping flexibility. Some have also argued
that truly unexpected events, such as from a significant
wind or solar forecast error, should actually qualify as a
contingency and permit use of contingency reserve products such as nonspinning reserve. So at this point, except
for MISO and CAISO, the future development of dedicated
flexiramp products remains unseen.

Planning and Investment
Both planning and operations are critical to a reliable and
economic power system, but they happen on very different time frames. The operational time frame is now. The
operator must use whatever is currently available. The optimization problem is how to operate the existing fleet in a
cost-effective way, given reliability constraints and costs. If
done correctly, the pricing signals in the day-ahead and realtime energy markets will provide the incentives for market
participants to offer their energy and flexibility for use by the
system operator.
In contrast, the planning time frame is up to several years
into the future, long enough to make and implement decisions about what to build. If planning is done well, then the
operators will have the resources and capabilities they need
november/december 2015

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