Fault Tolerant Control Based on an Observer on PI Servo Design for a High-Speed Automation Machine

The fault tolerant control (FTC) technique is widely used in many industries to provide tolerance to systems so that they can operate when a system fault occurs. This paper presents a technique for FTC based on the observer signal application, which is used for a high-speed auto core adhesion mounting machine. The utilization of the observer signal information of the linear encoder fault is employed to adjust the gain parameters to achieve the appropriate gain value while maintaining the required performance of the system. The dynamic modeling of the servo motor system design utilizing a pole placement technique was designed to support the proposed method. A scaling gain fault step size adjustment from −1% to 1% with increments of 0.2% is used to simulate the fault conditions of the linear encoder. The statistical mean value of the observer error signal is used to train the artificial neural network (ANN) model. The results showed that the control system design successfully tracked the dynamic response. Furthermore, the ANN model, with more than 98% confidence, was satisfactory in classifying the linear encoder fault condition. The gain compensation was successful in reducing position error by more than 95% compared with the system without compensated gain.