Variable Clustered Fuzzy Rules for Self-Tuning Scheme for Respiratory Distresses by Similarity and the Phase Plane Trajectory Concept

Abstract

This work proposes a fuzzy rule extraction methodology for a self-tuning fuzzy controller based on the fuzzy clustering method (FCM) and the similarity approach technique. The similarity technique and phase plane trajectory method are used to even lower the acquired rules. To show a potential rule extraction scheme, the self-tuning fuzzy-logic-based proportional-derivative (STFLPDC) with 49 expert fuzzy rules and 49 clustered fuzzy gain rules is used. The utility of the approach is validated using common clustering validity indices. With 49 initial clustered fuzzy rules, the suggested method addressed oxygen supply in a human respiratory model. After further reduction using the similarity strategy, 29 and 14 fuzzy rules remained. Parallel to this, a real-time benchmark application using 49, 21, and 16 extracted fuzzy gain rules is used to access the performance of a lab-based overhead crane. Finally, to investigate control performance in both models, the phase plane trajectory concept is used to generate as few as 13 fuzzy gain rules.