Comparison of two types of nonlinear controllers for magnetic bearing system stabilization: An experimental approach

In this paper, the performance of a dynamic output feedback sliding mode controller (SMC) has been compared experimentally with that of a fuzzy logic controller (FLC) for the stabilization of a MBC500 magnetic bearing system. The sliding mode controller with a state observer is designed for the unstable nonminimum phase magnetic bearing system based on the reduced order model of the system. The FLC uses error and rate of change of error in the position of the rotor as inputs and produces an output voltage to control the current of the amplifiers driving the magnetic bearing system. Both types of nonlinear control methods can handle the system resonant modes and maintain the magnetic bearing system stability without the use of any notch filters. The performances of the designed two types of nonlinear controllers are compared via real time experiment. In the experiment, the controllers have been coded in C and implemented in real time via a Digital Signal Processor (DSP). Experimental results have demonstrated that both nonlinear controllers can maintain the magnetic bearing system stability when the system is subjected to small to medium size disturbances. Furthermore, it has been shown that SMC can provide faster dynamic responses with much smaller control signals than FLC. In addition, the SMC is more robust than FLC as it can maintain system stability when the magnetic bearing system is subjected to large disturbances.