Multivariable self-organizing fuzzy logic control using dynamic performance index and linguistic compensators

As far as fuzzy logic based multivariable control systems are concerned, it is not always an easy task to express control strategies in the form of related multi-situations to multi-actions control rules. Decoupled control is one possible and attractive strategy to simplify this problem. However, the control performance of the decoupled controller relies greatly on ‘a prior’ knowledge of the system dynamics to build suitable compensators. This paper aims at introducing a new model-independent decoupled control architecture with the ability of on-line learning, which ensures a fast tracking performance. In this architecture, the dominating controller is developed using a new model-free Self-Organizing Fuzzy Logic Control (SOFLC) architecture whereby the Performance Index table is ‘dynamic’, of a free structure, and starting from no knowledge. Furthermore, a switching mode scheme, with a compensating action triggered by the interaction between the channels, is proposed to improve the tracking performance of the closed-loop system. A series of simulations are carried out on a two-input and two-output biomedical process, with the conclusion that the proposed control mechanism has the ability to deal with varying system dynamics and noise and is tolerant to the choice of the compensator gains effectively.