gCLxcKKAJmbACaihfr7QajzX6AsZRlzTBM0AxvT0

# Fuzzy Time Control Modeling Of Discrete Event Systems

## (Jurnal Teknik Informatika)

ABSTRAK

Linguistic modeling of complex irregular systems is helpful for the generation of decision making controls. In the various existing Fuzzy models, proposed by Mamdani, Sugeno, and Tsukamoto, the concepts of the set of membership functions and different Fuzzy logic rules to reason about data were addressed. The time control issues were not discussed in these models. In this paper, a new model is proposed with initial membership functions of the fuzzy model and the linguistic fuzzy rules with time control membership function of the binary valued outputs instead of crisp values. The system is named fuzzy logic time control system (FLTCS) with the proposed timing approach and implemented with discrete event system DEVS.

Index Terms—Irregular systems, Fuzzy models, Time control
issues, Time control membership function, Fuzzy logic time
control systems and discrete event system.

I. INTRODUCTION
The fuzzy logic and fuzzy set theory deal with the nonprobabilistic uncertainties issues suggested by zadeh in
1965 [1]. These concepts are evolved in various disciplines, such as calculus of fuzzy if-then rules, fuzzy reasoning, fuzzy inference systems, fuzzy modeling, fuzzy graphs and fuzzy topology. Fuzzy models are used in various disciplines such as automatic control, decision making expert systems, consumer electronics and computer vision etc [2],[3],[4].

The fuzzy control is based on the theory of fuzzy sets and fuzzy logic. Mamdani’s pioneering work motivated to persue the research field of control systems in fuzzy modeling [5].

Previously number of fuzzy inference system and defuzzification techniques were reported [6] and [7]. These
systems/techniques had less computational overhead and were useful to obtain crisp output [6] and [7]. Indeed existing fuzzy models have addressed the way to reason using membership function and fuzzy rule [8], [9], [10] and [11], but did not take into account the time dependency of output(s) in control systems. The crisp output value was based on linguistic rules applied in inference engine and defuzzification techniques
[12].