Contents 1 Introduction
It is very difficult to construct a reduced-order model for control system design of rotor system with gyroscopic effect because gyroscopic matrix is a anti-symmetric matrix. So the reduced order model of rotor system without gyroscopic effect is used to construct the controller and the controller used in closed loop system with gyroscopic effect is desired to get good control performance. But sometimes good control performance can not be obtain for rotor system with strong gyroscopic effect. So, in this paper cholesky decomposition has been used to construct the reduced order model for control. In order to demonstrate the validity of this method, the controlled model and zero power control of a 0. 5KWh class flywheel system using magnetic bearing with gyroscopic effect are given in this paper. In this paper, the zero power control means not only the zero bias current but also the zero control current. Only electromagnetic bearings are used in the system and permanent magnetic bearing is not used. When a magnetic bearing is generating control input, the opposite magnetic bearing does not generate any control input. When the rotor keep it's equi-librium, the magnetic bearings do not use any energy or current at this time. The magnetic bearing system is different from a conventional magnetic bearing system [1]. Firstly, the one-dimensional FEM model and the reduced-order FEM model with gyroscopic effect of a 0.5KWh class flywheel system used for control are introduced in this paper. Secondly, the zero power controller and the low-pass filter are designed by using the reduced-order FEM model Here, the low-pass filter is used to avoid spillover of higher order vibration modes. Thirdly, the validity of the controller and the low-pass filter are verified by experiment. From experiment, it seems the proposed method is useful for control system design taking into account a strong gyroscopic effect.
