IEEE Power & Energy Magazine - November/December 2019 - 53

the system-critical part of the capacity is determined,
wind energy penetration has reached a level that can meet
e.g., 30% of the capacity)
or exceed the local demand. In such areas, concurrent shutdown of multiple wind plants due to high wind speeds can
✔ an accumulated part of the expected cutoff capacity,
cause system problems.
which provides the accumulated cutoff probability for
Ireland is such an area, where practically all wind plants
the expected temporal shortage of capacity and ramps
can produce full generation when wind speeds reach approxi✔ a table combining the information above for more demately 15 m/s and continue until the plants reach their
tailed analysis and action planning.
HSSD set point. The power measurements do not indiFigure 9 displays the three components of the syscate when such set points are reached in the wind-speed tem. Figure 9(a) shows the amplitude of the extreme wind
range below the HSSD: i.e., 20-25 m/s. Forecasting and speeds that cause a certain percentage of cutoff combined
signaling the high wind speed enables a system operator with strong down-ramps. This provides important knowlto prepare for large-scale HSSD events. Since a lack of edge to help determine 1) the maximum reserve required
power is more dangerous for power system reliability to maintain system balance, 2) the risk of a shortage that
than surplus, which can always be curtailed, a need for exceeds available up-ramp capacity, and 3) the risk for
action exists long before a deterministic forecast can pro- excess capacity and congestion in the power lines. After
vide accurate information.
Forecasting the probability of
an HSSD event enables the system
Capacity Availability
operator to verify system reliability and allocate reserves long in
55,000
advance to keep costs at a mini52,500
mum. At the Irish system operators
EirGrid and System Operator for
50,000
Northern Ireland, an HSSD warn47,500
ing system based on WEPROG's
75-member Multi-Scheme Ensem45,000
ble Prediction System (MSEPS), a
42,500
type 3 method (see Figure 1), has
14:00
15:00
16:00 17:00
18:00
19:00 20:00
21:00
22:00
been developed for this purpose.
Time
The HSSD warning system conOnline Generation With Wind
tains three components:
Online Generation With Wind and Uncertainties
✔ a probability indicator that
Load Forecast
Regulation-Up Nonspinning Online Room
a fraction of the HSSD will
experience cutoff (in cooperation with the end user, figure 8. The ERCOT online tool to use probabilistic wind forecasts.

(a)

(b)

(c)

figure 9. The three components of the HSSD warning system. (a) The probabilities of HSSD in terms of nine percentile
bands P10-P90, computed from WEPROG's 75-member MSEPS. The y-axis depicts HSSD as a percentage of total generating capacity. (b) The graph shows the accumulated probability over time and provides information about the required
amount and length of reserve allocation, with probabilities to be read from the nine percentile bands. (c) The same information in tabular format. (Source: WEPROG; used with permission.)
november/december 2019

ieee power & energy magazine

IEEE Power & Energy Magazine - November/December 2019

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - November/December 2019