Environmental Engineering Reference
In-Depth Information
In the case of the SDE, only random errors make a contribution to the error
measure. For both the RMSE and the SDE large forecast errors make a stronger
contribution to the measure than small forecast errors.
To illustrate the way these measures are typically presented, consider the
application of the basic persistence model to two cases: (1) a single wind farm and
(2) 15 geographically dispersed wind farms. The single wind farm has a high
capacity factor and the total capacity of the fifteen geographically dispersed wind
farms is about 20 times that of the single wind farm. Using the normalised measured
power the MAE, RMSE and SDE for the basic persistence forecast model are cal-
culated for a test period of six months and plotted in Figure 6.3 against look-ahead
Single wind farm
0.45
0.40
0.35
0.30
0.25
0.20
0.15
MAE
RMSE
SDE
0.10
0.05
0.00
0
6
12
18
24
30
36
42
48
(a)
Look-ahead time (h)
15 geographically dispersed wind farms
0.45
0.40
0.35
0.30
0.25
0.20
0.15
MAE
RMSE
SDE
0.10
0.05
0.00
0
6
12
18
24
30
36
42
48
(b)
Look-ahead time (h)
Figure 6.3
Error measures in per unit of installed capacity versus look-ahead
time for (a) a single wind farm and (b) 15 geographically dispersed
wind farms, calculated over a six-month test period using a basic
persistence model. Note that in the above plots the SDE and RMSE are
indistinguishable from each other
Search WWH ::
Custom Search