State-Space Self-Tuning Control for Singular Nonlinear Stochastic Hybrid Systems Containing Saturating Actuators

• Main Authors: Ying-Chuan Peng, 彭瀛全
• Other Authors: Jason S. H. Tsai
• Format: Others
• Language: en_US
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/46753114497339895378

碩士 === 國立成功大學 === 電機工程學系碩博士班 === 91 === A state-space self-tuning control scheme for adaptive digital control of continuous-time singular nonlinear stochastic systems, which have unknown system parameters, measurement noises, deterministic noises, and inaccessible system states is first proposed in this thesis. The singular nonlinear system is mixed with chaos and singular system in unstable impulsive mode; therefore, it is more challenging for control. First, an adjustable auto-regressive moving average (ARMA)-based noise model with estimated states is constructed for state-space self-tuning control of singular nonlinear stochastic systems, then the optimal tracker is proposed. Under this proposed optimal tracker, the stability can also be shown for the hybrid system with the deterministic noise. Next, the conditioning dual-rate digital redesign scheme is developed, which contains a conditioning fast sampling rate digital controller for reducing the bump-transfer effects and a slow sampling rate digital redesign optimal tracker. Finally, a steady-state observer is based on some reasonable initial values for the system identification. As a result, it can be used to design an optimal tracker with saturation actuators. Illustrative examples are given to demonstrate the effectiveness of design methodologies.