| Summary: | 博士 === 國立成功大學 === 機械工程學系 === 89 === Analysis and control for active magnetic bearing (AMB) systems with flexible rotor is considered in this thesis. Identification and control methods for chaotic systems are proposed. System state trajectories, power spectrum, Poincarė map, bifurcation diagram and fractal dimension are used to identify chaotic motion of the AMB systems.
For the control of chaotic system, a sliding mode hyperplane design approach is applied. The approach guarantees the asymptotical tracking of desired orbit. The converging speed of error states can be determined by the assigned sliding surfaces. Tow systems, namely, Chua’s circuit system and Rössler’s system are studied.
The nonlinear model of the AMB system with electromagnetic actuator and flexible rotor are proposed for system analysis and control design study. This nonlinear model includes: flexible rotor dynamics, nonlinear electromagnetic suspended system and contact force between auxiliary bearing and shaft. The AMB system is stabilized for a PD controller and the bifurcation diagram is used to analyze dynamic behaviour of the AMB system. The results show that some system parameters have strong inference on the behaviour of the AMB system, such as, the radial clearance between shaft and magnetic bearing, the unbalance effect of disk, the stiffness of shaft, and the bias current of control, etc. Therefore, aperiod motion effect of the AMB system can be avoided by suitable parameter manipulation.
For the control of the AMB systems, the above sliding mode control scheme is employed in this study. Numerical results show that the control algorithm will make aperiodic motion into periodic motion under control action. Besides, the vibration amplitude is decreased due to unbalanced effects and disturbances.
|
|---|
