IEEE Power & Energy Magazine - September/October 2019 - 75

Total Operation Cost (US$Million)

advance the VER integration literature. As mentioned later to be thoroughly assessed both for their physical impacts on
in this article, EPECS simulation has recently been demon- grid reliability and for the economic implications of their prostrated in a full-scale case study in the ISO New England curement and provision.
The electricity grid can currently accommodate relatively
System Operational Analysis and Renewable Energy Study.
Second, as VERs are increasingly integrated into the grid, modest levels of VERs. However, as VERs gain penetrathey motivate the need for an expanded set of essential grid tion into the energy mix, the generation fleet slowly loses its
services. Perhaps the most prominent of these is the provision dispatchability. In such a case, fast-ramping natural gas and
of three different types of operating reserves that respond to hydroelectric power plants take on a prominent grid-balancing
grid conditions from as fast as several seconds in the case or role. At even higher VER penetration, the natural gas plants
regulation reserves and to up to 30 min in the case of second- become crowded out, and dispatchable DSRs become the only
remaining option for grid balancing. Such a scenario constitutes
ary contingency reserves.
✔ Contingency (or emergency event) reserves have long a fundamental transformation of the grid's architecture. At a
been an essential grid service to cover the unsched- technical level, DSRs take on a greater fraction of the control
uled outage of the system's largest source or generator. inputs for grid balancing, line congestion management, and
✔ Similarly, regulation reserves are procured as a grid voltage control. At an economic level, the market structures that
service to enable automatic generation control that incentivize these DSRs remain an open area of research.
For example, a recent EPECS study contrasted the technoecomaintains the interchange schedule and frequency.
nomic
benefits of demand-side management using a transactive
✔ Beyond these traditional grid services, VERs require
operating reserves during normal operating condi- energy scheme versus the virtual power plant scheme in Order
745 of the Federal Energy Regulatory Commission (FERC).
tions or simply normal operating reserves.
Because VERs are inherently intermittent and uncer- FERC is the United States' federal agency that regulates the
tain, they affect the variability and forecast error of the net transmission and wholesale sale of electricity and natural gas
load (i.e., the total power demanded minus variable genera- and the transportation of oil by pipeline in interstate commerce.
tion). Both intra- and interhour variabilities, when coupled FERC Order 745 concerns demand-response compensation in
with forecast errors, tend to cause power system imbalances. organized wholesale energy markets and implements demandRecent research has numerically shown that as the net-load response resources using virtual power plants that are incentivvariability and forecast errors increase, the required quanti- ized downwards from their predicted demand baselines.
As shown in Figure 4, the research concludes that the presties of load following, ramping, and ramping reserves must
also increase. In the ISO New England study, the increasing ence of demand baseline errors, present only in the FERC
penetration of VERs was shown, in some cases, to entirely Order 745 implementation, leads to a cascade of additional
exhaust all three types of operating
reserves and that such a condition
was closely tied to the presence
of power system imbalances. The
2.2
+35.69%
development of EPECS simulation
2
models has provided the theoreti1.8
+17.3%
cal insight to develop closed-form
37.5%
1.6
21.43%
analytical solutions for the quanti6.43%
1.4
5.84%
5.84%
ties of all three types of operating
5.84%
1.2
reserves based upon the statistical
37%
1
properties of the net load and the
37.55%
38.07%
time step of the day-ahead and
0.8
real-time energy markets.
0.6
VERs decrease the ability of
0.4
52.48%
54.27%
50.73%
existing system operations to bal0.2
ance the grid and, consequently,
0
increase the dependence on the
Transactive
5% Real-Time
10% Real-Time
available operating reserves. As
Energy
Baseline Error
Baseline Error
the grid comes to increasingly
Generation
Virtual Generation
Load Following
Regulation
rely on these resources, it is likely
that the provision of grid inertia
and voltage regulation will become figure 4. A comparison of the total production costs from the transactive energy
important grid services as well. model, an industrial model with a 5% demand baseline error, and an industrial
All of these new and increasingly model with a 10% demand baseline error. The presence of a demand baseline only
important grid services will need exists in the FERC demand-response implementation.
september/october 2019

ieee power & energy magazine

IEEE Power & Energy Magazine - September/October 2019

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - September/October 2019