Information Technology Reference
In-Depth Information
12
49-rule FLC with Gaussian MFs
Approximated 4-rule FLC with Gaussian
MFs
10
8
6
4
2
0
Time (sec)
Fig. 20 Comparison of THD pro
le of source current with 49-rule FLC and approximated 4-rule
FLC using Gaussian MFs
Table 6 Comparison of computational memory requirement
Sl. no.
Module
Parameter
49-Rule
FLC
4-Rule
approximated
FLC
1.
Fuzzi
cation
No. of input variables
2
2
Number of MF for each input
7
2
Memory units required
14 * m
4 * m
2.
Fuzzy infer-
ence
and knowledge
base
Number of rules
49
4
Number of antecedent
7
4
Number of consequent
7
4
Memory units required
686
32
3.
Defuzzi
cation
Number of output variable
1
1
MF for output variable
δ
I
max
7
3
Memory units required
7 * m
3 * m
4.
Approximation
Number of comparators
-
-
Number of multipliers
9
-
Number of adders
3
-
Number of terms with non zero
coef
cients
-
4
Memory units required
16
-
Total memory requirement
21 *
m + 686
7*m+48
where
'
m
'
represents unit memory required for each membership function
4.6 Comparative Analysis of Computational Memory
A comparison of memory requirement of 49-rule and proposed approximated FLC
is presented in Table
6
. For comparative analysis of computational memory
requirement, an FLC is functionally divided into following three sections:
Search WWH ::
Custom Search