Research On Suspension Control And Signal Processing For Rotor Mass Varying Magnetic Bearing

AMB(Active Magnetic Bearing) is a new kind of high performance bearing, the rotor is suspended by magnetic force. There is no contact and friction between rotor and bearing. Compared with the traditional bearing, AMB has many advantages, such as no lubrication request, no loss, low maintenance cost, long service life, etc. Due to the elimination of the friction, the rotation rate of AMB will be greatly improved, and its applications are more extensive.Stability of suspension is the key to the successful operation of AMB, the quality of control algorithm determines the suspension performance directly.In this paper, based on the single DOF radial AMB, the suspension control algorithm is studied. The primary research contents are summarized as follows:PID control algorithm is studied. The simplified mathematical model is established. The model is linearized near the balance point by the method of Taylor expansion. The open-loop transfer function of linear model is deduced and analyzed. The PID control algorithm is designed. The control parameters are calculated by the pole assignment method. The influences of suspension mass changing, brought by AMB's load changing, on system model parameters are discussed, and the influence on the PID control performances are showed by simulink.Self-turning control algorithm is studied. The AMB system has very strong nonlinearity essentially, but the Taylor expansion linearization method can only guarantee the stability of the system near the balance point. When far from balance point, the system's performances will become worse even instability. The bearing load usually varies largely, that means the rotor quality is changeable in work. In the circumatances, the conventional PID control can not maintain good performance. Aiming at this problem, the exact linearization method is proposed to make up for the shortcomings of Taylor expansion, to insure that the system maintain good performance when load changes. Based on the exact linearization, aiming at suspension mass varying, the self-turning control method is designed to guarantee the system maintain good performance when load changing.Gap differential algorithm is studied. It is very significant to design gap differential algorithm to improve the performance of the control system. According to the frequency characteristics of gap and noise signals, the basic performance indicators of differential algorithm are analyzed. Accordance with these indicators, how to choose the time constants of inertial element subtraction method and the rapid factor of piecewise linearzation tracking-differential algorithm is calculated and analyzed. Application of tracking-differentiator in control system is simulated, the system performances are better than DFDM, and the effectiveness and the capability of restraining noise are validated.