Paper:
Views over last 60 days: 346
Improvement of Control Performance for Low-Dimensional Number of Fuzzy Labeling Using Simplified Inference
Kenichiro Hayashi, Akifumi Otsubo and Kazuhiko Shiranita
Applied Electronics Division Industrial Technology Center of Saga Prefecture 114 Yaemizo, Nabeshima-cho, Saga-shi 849-0932, Japan
Received:April 20, 1999Accepted:May 24, 1999Published:October 20, 1999
Keywords:Fuzzy control, Simplified inference, Low number of fuzzy labeling, Fuzzy control simulator, Control performance, Control map
Abstract
One of the important concepts in fuzzy control is fuzzy inference, and simplified inference, which increases the speed of the fuzzy inference, has been used to realize a high-speed fuzzy controller. In designing a fuzzy controller, a high dimension, such as 7 x 7 or 5 x 5 partitions, is frequently used for the number of fuzzy labeling. However, as the number of fuzzy labeling increases, the number of control parameters increases rapidly and tuning of the fuzzy controller becomes difficult. Therefore, a fuzzy controller is required to be partitioned into a low number of fuzzy labeling, such as 3 x 3 partitions. With this in mind, first, a method of improving control performance for a low number of fuzzy labeling using simplified inference which enables high-speed inference, is proposed in this paper. Next, the effectiveness of this improvement method is studied based on the results of several simulations where a first-order lag system with dead time, a representative model for plant characteristics, is used as the controlled system.
Cite this article as:K. Hayashi, A. Otsubo, and K. Shiranita, “Improvement of Control Performance for Low-Dimensional Number of Fuzzy Labeling Using Simplified Inference,” J. Adv. Comput. Intell. Intell. Inform., Vol.3 No.5, pp. 431-438, 1999.Data files:
