Majalah Ilmiah UNIKOM

Vol.10 No. 1

67

H a l a m a n

Fuzzy as discussed in [5] – [7], as far as the

author concerns, there is no work on

development of type-2 Fuzzy is suitable for

elevator group control. Contribution of this

paper is to propose type-2 fuzzy algorithm

suitable for elevator group control and

compare the performance between the

interval type 2 Fuzzy and type 1 Fuzzy.

The paper is arranged as follows. Type-2

Fuzzy structure is provided in Section II. In

Section III we describe the elevator group

control system and the area-weight of the

control system. Fuzzy model to determine

the area-weight is presented in Section IV.

Its performance is analyzed through the

simulation in Section V. In Section VI, we

conclude with conclusion.

II. TYPE-2 FUZZY STRUCTURE

Type-2 fuzzy sets were originally

presented by Zadeh in 1975. The new

concepts were introduced by Mendel and

Liang allowing the characterization of a type-

2 fuzzy set with a superior membership

function and an inferior membership

function; these two functions can be

represented each one by a type-1 fuzzy set

membership function. The interval between

these two functions represent the footprint

of uncertainty (FOU), which is used to

characterize a type-2 fuzzy set. Type-2 fuzzy

sets allow us to handle linguistic

uncertainties, as typified by the adage

“words can mean different things to

different people”.

For type-2 TSK models, there are three

possible structure [11]:

1. Antecedents are type-2 fuzzy sets, and

consequents are type-1 fuzzy sets. This

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.

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

where

and

. The

consequent parameter

, which are type-1 fuzzy sets, has interval, is

denoted as

T h e

m e m b e r s h i p

g r a d e s

are interval sets to, which denoted as

Where

is lower membership function

and

is upper membership function.

These rules let us simultaneously account

for

uncertainty

antecedent

membership functions and consequent

parameter values.

Fuzzification

This process is transforming the crisp

input to a type-II fuzzy variable. The primary

membership functions for each antecedent