228
G.J. Bowden . G.c. Dandy . H.R. Maier
4500
AClual
4000
-r-Predi 1«1
~ 3500
-;;
..:?
2 3000
,!:!
~ 2500
!f c:
0
U 2000
ci.
sr
"
1500
il
"
..,
" 1000
"
"'=
500
0
0
20
40
60
80
100
imc (wecks)
Figure 11.2.
Forecasts and actual concentrations of Anabaena spp. for the
validation period for modell (input determination: apriori + GA-ANN).
4500
4000
E
500
~
~ 3000
.!:!
~ 2500
" u c
0
V 2000
ci.
c.
'"
"
il
ISOO
.g
~ 1000
"'=
500
0
-
Aclunl
- r - Prcdiclcd
0
20
40
60
Time (wecks)
80
100
Figure 11.3.
Forecasts and actual concentrations of Anabaena spp. for the
validation period for model 2 (input determination: apriori + Stepwise ANN
modelling procedure).
G.J. Bowden . G.c. Dandy . H.R. Maier
4500
AClual
4000
-r-Predi 1«1
~ 3500
-;;
..:?
2 3000
,!:!
~ 2500
!f c:
0
U 2000
ci.
sr
"
1500
il
"
..,
" 1000
"
"'=
500
0
0
20
40
60
80
100
imc (wecks)
Figure 11.2.
Forecasts and actual concentrations of Anabaena spp. for the
validation period for modell (input determination: apriori + GA-ANN).
4500
4000
E
500
~
~ 3000
.!:!
~ 2500
" u c
0
V 2000
ci.
c.
'"
"
il
ISOO
.g
~ 1000
"'=
500
0
-
Aclunl
- r - Prcdiclcd
0
20
40
60
Time (wecks)
80
100
Figure 11.3.
Forecasts and actual concentrations of Anabaena spp. for the
validation period for model 2 (input determination: apriori + Stepwise ANN
modelling procedure).
