< prev

Page 1Page 2Page 3Page 4Page 5Page 6Page 7Page 8Page 9Page 10Page 11Page 12Page 13Page 14

Page 3 of 14
next >

Majalah Ilmiah UNIKOM

Vol.8, No. 2

205

H a l a m a n

Muhammad Aria

is the most general case and we call it

Model I.

2. Antecedents are type-2 fuzzy sets, and

consequents are crisp number. This is

special case or Model I and we call it

model II.

3. Antecedents are type-1 Fuzzy sets and

consequents are type-1 fuzzy sets. This

is another special case of Model I and

we call it Model III.

We use Model I to design interval type-2

TSK Fuzzy system in this paper. A schematic

diagram of the proposed T2TSK structure is

i

m

Rule Base

In a first-order type-2 TSK Model I with a

mn

denoted as

The membership grades

are interval sets to, which denoted as

Fuzzification

This process is transforming the crisp in-

put to a type-II fuzzy variable. The primary

membership functions for each antecedent

are interval type-2 fuzzy systems described

by Gaussian primary membership function

with uncertain means, denoted as

There are two kinds of type-2 sets. First is a

gaussian type-2 fuzzy set, which the

membership grade of every domain point is

a Gaussian type-1 set contained in [0,1].

Second is an interval type-2 fuzzy set which

the membership grade of every domain

point is a crisp set whose domian is some

interval contained in [0,1]. Figure 2. shows

gaussian interval type-2 fuzzy membership

function with uncertain means.

The upper membership function is defind as

Figure 2. Gaussian interval type-2 fuzzy

membership function with uncertain means