Fault detection for full-closed loop servo drive systems based on interval observer

Most existing fault detection techniques for servo drive systems are data-driven, which requires the acquisition of a large amount of data from the system in operation, and the method is ineffective for fault detection when the fault data are not easily available. Aiming at the faults of servo drive systems, this paper proposes a fault detection method based on the interval observer, in which the model-driven approach was used. And it does not require to know a large amount of data from the system in operation. Firstly, the servo drive system is modelled and the interval observer is constructed. Secondly, whether there is a fault in the system is determined by judging the positivity and negativity of the residual signals. Finally, the fault simulation is carried out to verify the validity of the interval observer used in the fault detection of the servo drive system.