Abstract
Due to flow asymmetries created in the tip region, aircraft forebodies can experience large net side forces at high angles of attack. It has been shown that these forces can be controlled by the use of suction in the tip region. For this purpose, a control system that manipulated the flow near the tip was tested on an aircraft forebody model in a wind tunnel. Utilizing a conventional PD algorithm, the system was able to control the forebody satisfactorily when tuned for a particular Reynolds number; but the performance degraded as the Reynolds number was changed. Fuzzy logic control (FLC) was then applied and shown to be more robust than PD. However, the FLC also showed some degradation for flows at different Reynolds numbers. To address this problem, an adaptive fuzzy controller was used. This paper discusses fuzzy model reference learning control (FMRLC) as applied to the aircraft model. It is shown that FMRLC is able to quickly and automatically generate a rule base to control the model. It is then shown that he FMRLC is able to modify it’s rule base and adapt in real time when the Reynolds number of the system changes.
Nonlinear performance seeking control using Fuzzy Model Reference Learning Control and the method of Steepest Descent