An Evolutionary Objective Cluster Analysis-Based Interpretable Fuzzy Identification Method - Intelligent Computing for Sustainable Energy and Environment - page 174

As is shown, the inner of each cluster distributes uniformly and the margin is
obvious and easily to be distinguished. Thus the result of clustering is effective.
Fig. 3shows the consistency relation on Z C and Z D .

140

120

100

80

60

c=2, ita CD =0.0211

40

20

0

0 50 100 150

Number of clusters at each clustering

Fig. 3. Enhanced consistency criterion-cluster number at each clustering in Z C and
Z D

As is seen in Fig. 3., the clustering results are directly obtained without any
presetting parameters or human judgment. Therefore they have stronger robust-
ness than other Ad-Hoc algorithms.
The error between the actual output and the proposed model is shown in
Fig. 4.

4

Real output
Output of model

3

2

1

0

−1

−2

−3

0 50 100 150 200 250 300

Time(s)

Fig. 4. Comparison between the real output and the output of model

Case II. Low voltage electrical application
In this problem, the number of users and the area of villages is the input
( x 1 ,x 2 ), and the output y is the length of low voltage line. Fig. 5. shows the
unadjusted membership functions. As is shown, in each variable domain, the
fuzzy sets have more overlapping and are dicult to be distinguished.