Rule mapping fuzzy controller design

✍ Scribed by Ching-Chang Wong; Cheng-Shun Fan

• Publisher: Elsevier Science
• Year: 1999
• Tongue: English
• Weight: 226 KB
• Volume: 108
• Category: Article
• ISSN: 0165-0114
• DOI: 10.1016/s0165-0114(97)00305-9

✦ Synopsis

In this paper, we propose a rule mapping fuzzy controller design method based on genetic algorithms to automatically generate the control rules and the membership functions of the fuzzy controller such that the controlled system will have a good control performance. Three performance evaluations (rise time, overshoot and integral absolute error) are considered in a proposed "tness function with weights and the concept of membership function is applied to express the grade of goodness in each performance measurement. It is more #exible than the conventional method for di!erent system speci"cations because the appropriate fuzzy controller can be e$ciently selected by the proposed method to meet desired performance. An inverted pendulum control problem is selected to illustrate the e$ciency of the proposed method. Finally, we consider a self-regulating rule mapping fuzzy controller where some parameters are selected by the proposed method. It shows that the performance can be further improved.