IEEE Power & Energy Magazine - September/October 2017 - 40

200
MW

30%

50%

350
MW

400
MW

S1 (π = 15%)

50%

250
MW

350
MW

S2 (π = 15%)

50%

250
MW

300
MW

S3 (π = 35%)

50%

200
MW

S4 (π = 35%)

2025-2028

2029-2032

250
MW

200
MW
70%

200
MW

50 MW
400 MWh
2017-2020

(a)

2021-2024

(b)

figure 11. (a) The system under study showing candidate assets. (b) A scenario tree showing the evolution of peak
-demand across four scenarios (S1-S4) over four-year periods.

Build Line

No
Investment

No
Investment S1 £70.0 Million

Build Line

No
Investment

No
Investment S2 £70.0 Million

No
Investment

Build Line

No
Investment S3 £60.4 Million

No
Investment

No
Investment S4

£0.0 Million

figure 12. The optimal investment plan for each of the four
scenarios shown in Figure 11 when examined in isolation.

been developed to describe potential trajectories of the system's
demand evolution, as shown in Figure 11.
Scenario 1 represents a high-growth situation and is assign-
-ed 15% probability. Scenarios 2 and 3 capture more moderate growth cases, while scenario 4 describes the case in which
no additional load is added to the system, occurring with 35%
probability. The planner has the following two options for reinforcing the system:
✔✔ Install a new 200-MW line at a present capital cost of
£70 million. We assume that the commissioning of the
new line takes four years to complete due to permitting and other asset deployment constraints.
✔✔ Immediately deploy an ES plant with a 50-MW power
rating and 125-MWh energy capacity; this plant has a
present capital cost of £30 million.
Figure 12 presents the optimal investment plans obtained
with a deterministic approach, in which the planner considers
40

ieee power & energy magazine

each of the individual scenarios in isolation. In scenarios 1
and 2, the optimal strategy is to build a new line from the
start, while in scenario 3 the investment is deferred to the
second stage. Scenario 4 entails no demand increase and
hence no investment is required.
It is of paramount importance to note that in Figure 12,
although the option to build ES is available, it is not used
in any individual scenario as it does not present additional
benefit under perfect foresight. Figure 13, on the other hand,
shows the optimal investment strategy when the planner
does not know which of the four scenarios will actually materialize and hence needs to consider uncertainty in making
investment decisions.
Two cases are shown, one in which investment decisions
are limited to new lines only [Figure 13(a)] and the other that
also allows the possibility of investing in ES [Figure 13(b)].
In the first version, it is necessary to build a line immediately to ensure that scenarios 1 and 2 do not lead to demand
curtailment. However, when investment in ES is possible, a
different strategy emerges: the first-stage investment decision is deferred to later stages and is made conditionally
based on the scenario's realization. The planner adopts a
"wait-and-see" approach in the beginning and subsequently
deploys both an ES plant and a line if demand begins to
increase. The ES plant is used to enable interim operation
until the line is commissioned. The expected net present
values of the total investment cost for the two versions are
£70 million and £50.85 million, respectively, making the
option value of ES £19.15 million. This value arises from
the ability of ES to allow the decision to invest in a new
line to be deferred until later stages, if demand increases
sufficiently; this case also ensures that, under scenario 4, no
unnecessary investment is undertaken.
september/october 2017

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