| Summary: | 碩士 === 國立高雄應用科技大學 === 機械與精密工程研究所 === 93 === This thesis presents a robust disturbance attenuation method for a class of multi-degree-of-freedom systems with time-varying uncertainties and unknown persistent excitation. The control input is divided into two parts: one is obtained from the optimal LQ method or Genetic Algorithm that is responsible for primary stabilization, the other is derived from the minimum norm solution to attenuate the effect of the persistent disturbance. The states of controlled dynamics and unknown disturbance are estimated by a Waites Estimator in which the estimate gain matrix can be obtained by Pole Placement or Genetic Algorithm. Then, taking into account plant variations, an LMI stability condition is proposed to ensure the stability of the resulting closed system. It is shown that, using the proposed stability condition, the designed controller can effectively suppress the unknown persistent excitation and keep the system from the possibility of instability caused by the time-varying uncertainties. An example is given to demonstrate the use of the design method. Besides, the thesis also discusses the system response with constrained inputs. The results show that the designed control system can also achieve the robust disturbance attenuation with constrained inputs.
|
|---|
