Control Design and Analysis of Uncertain Systems via Robust Sliding Mode and Complex Interval Matrix Methods

• Main Authors: Chun-Mo Lee, 李君謨
• Other Authors: Jyh-Ching Juang
• Format: Others
• Language: en_US
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/49834705179149102142

博士 === 國立成功大學 === 電機工程學系碩博士班 === 93 === 　In the dissertation two topics are investigated regarding to the robust control theory. One topic is the design of robust controllers for sliding mode control systems. A linear matrix inequality technique is developed for the design of both sliding mode and reaching phase control laws. It is shown that by solving a set of LMIs, the switching surface can be designed such that the dynamics in the sliding mode achieve robust stability and bounded L2 gain performance with respect to matched and unmatched uncertainties in the presence of disturbances. Furthermore, a smooth reaching phase control law is presented to eliminate the undesirable chattering effect and maintain the robustness property all the time. Finally, a smooth observer based control law is developed to maintain the robust performance of a system in the case that not all of the states are accessible. In the second topic, a robustness analysis formulation, based on the structured singular value techniques, is proposed to study the stability problem of a complex interval matrix. Accordingly, explicit criteria for testing the Hurwitz stability of complex interval matrices, real interval matrices, complex interval polynomials and real interval polynomials are all constructed regarding to the proposed formulation. Besides, an application of complex interval matrices to the design of a control system is presented. Several simulations are given to illustrate the aforementioned design and analysis methods.