IEEE Power & Energy Magazine - September/October 2017 - 38

system inertia; conversely, during
peak load conditions, the value of
Value (£/kW)
frequency response provision is
0
1,000
2,000
3,000
4,000
5,000
very low, so ES would be able to
(1) = Arbitrage Only
provide peak management service.
(2) = (1) + Balancing
However, the present arrangement
for the provision of FR based on
(3) = (2) + PV Support
long-term contracts would pre(4) = (3) + Network Support
vent ES from accessing the rev(5) = (4) + FR Provision
enues associated with providing
(6) = (5) + Capacity Market
both services. This illustrates
clearly the need for efficient realEnergy Arbitrage Balancing PV Support
time market design for ancillary
Network Support FR Capacity Payment
services that would fully align
commercial incentives with system benefits. Such a design, howfigure 7. The value of ES for providing multiple services.
ever, would need to fully account
for reliable system operations,
It is clear that the revenues from any single service would not in-cluding operational planning needs and the human-ma--
justify the investment in ES. Optimized provision of mul- chine interface.
tiple services is, therefore, the key route for ES to make a
profitable business case.
Assessing the Security of
The results suggest considerable added value from stor- ES Supply Contribution
age being colocated and operated together with a PV farm ES plants, as well as other flexible solutions such as DSR
(or, potentially, a wind farm), particularly when there is an and distributed generation, could play a role in enhancing the
active network constraint. Participation in the capacity mar- security of electrical systems; they can provide peak shaving
ket could secure additional up-front payment for ES while services, as well as maintain supply during network cononly slightly reducing its profit from other markets.
tingencies, and may thus represent a viable alternative to
Provision of FR from ES would significantly enhance its conventional network reinforcement. The emergence of new
value proposition, although this would greatly depend on the technologies is creating the need to review historical network
FR market arrangements. Clearly, the present long-term, con- planning standards to establish a level playing field that will
tract-based ancillary service framework-rather than a real- allow both conventional network solutions and new techtime market for ancillary services-would prevent ES from nologies to be considered when determining the most costsimultaneously accessing the revenues associated with pro- effective solutions for dealing with demand growth.
viding FR and distribution network management services or
To determine the capacity credit of an ES plant conenergy arbitrage. Also, the system requirements for FR (and nected to a distribution network, we adopt the concept of
hence its value) vary significantly with actual system condi- effective load-carrying capability (ELCC), a standard mettions, i.e., the level of demand and production of renewable ric defined as the amount by which the demand can increase
generation (this is associated with the issue of inertia available while maintaining the same reliability of supply, measured
in the system, provided by rotating synchronous generators).
through expected energy not served (EENS). As such, ELCC
When low-demand conditions coincide with high renew- represents the amount of demand that can be added to the
able output, the value of FR can be very high due to reduced network supported by ES resulting in the same EENS as the
original network (referred to as the "base case"); this equivalence is shown in Figure 8.
A continuous-time chronological Monte Carlo simulation
framework
is used to generate populations of fault histories,
Same
which are combined with a discrete-time steady-state model
EENS
of ES operation to quantify the EENS across a large number of years. A simple N-1 system is assumed, as shown in
ES
Figure 8, with two transformers of 1 MW each and peak
D
D + ELCC
demand of 1 MW. It is further assumed that an ES plant is
fully dedicated to security provision. The ES security contri(a)
(b)
bution is then quantified based on the ES plant's power and
energy capability but also on network characteristics such as
figure 8. The equivalence between (a) the base case
network outage times and demand shape.
-system and (b) the system with ES. D: demand.
38

ieee power & energy magazine

september/october 2017

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - September/October 2017