6.4. EXAMPLE: TEMPERATURE CONTROL 213
0.4-0 .6 -0 .4 -0 .2 0 0.2 0.4 0.6 0.8
0.5
0.6
0.7
0.8
0.9
1
1.1
Figure 6.13. Integral gain versus pattern vector for ìnet2î
Figure 6.14. Variation in proportional gain to the training pattern
Figure 6.15. Variation in integral gain to the training pattern
We now plot the resulting error surfaces shown in Figures 6.16 and 6.17.
wv=-1:0.1:1; bv=-2.5:.25:2.5;
es11=errsurf(dp,tprop,wv,bv,ítansigí);
plotes(wv,bv,es11,[75, 25]);
es12=errsurf(p,tint,wv,bv,ítansigí);
plotes(wv,bv,es12,[75, 25]);
