| Summary: | Fuzzy controllers (FCs) are supposed to work in situations where there exists a large uncertainty or unknown variation in plant parameters and structures. Thus, it is necessary to introduce adaptive mechanisms in such control systems to tune the controllers to match current process characteristics. Of the various tunable parameters in a FC, the input and output scaling factors have the highest priority due to their global effect on the control performance. However, the working mechanisms of these scaling factors to the performance of fuzzy control system are yet to be founded due to the lack of an analytical structure for the general fuzzy control systems. In this thesis, for a general fuzzy control system, a systematic investigation on the importance of the scaling factors is carried out. Some quantitative and qualitative results about the relationships between the scaling factors and the system steady-state and dynamic properties are obtained. By applying the Lyapunov stability theory, this thesis analyzes the relationship between the scaling factors and the system stability for a class of uncertain nonlinear processes, where a simple and realistic sufficient condition is established in the sense that the system trajectory is bounded. The methodology and results established open up many opportunities for further research and encourage wide applications. Based on the obtained analysis results, a knowledge-based self-tuning fuzzy control scheme is proposed, where the scaling factors are tuned on-line based on the system performance indices and the current process states, and a stability monitor is introduced to guarantee that the states of the controlled process are bounded. The relevant design issues are also discussed.
|
|---|
