GIMC-based Fault Detection and Its Application to Magnetic Suspension System

This paper deals with a fault detection for a magnetic suspension system by using Generalized Internal Model Control (GIMC) structure. To design robust fault detection filters, two fault detection design problems are formulated as multiple objective optimization problems by minimizing the effects of disturbances and maximizing the fault sensitivity involving an LTI system with disturbance and fault signals. The fault detection filters designed by solving each optimization problems are implemented with the magnetic suspension system to verify its validity. A filter designed via the problem 1 has good transient performance, but the output signal of the filter is affected by the disturbance signals. Another filter which is designed via the problem 2, however, has good robustness for disturbance signals. Moreover, experimental results show that both filters have enough fault detection properties compared with a conventional detection filter.