Abstract
In industrial processes, faults may occur in any part of the system, such as actuators, sensors, and system components. Consideration of these uncertainties in control applications is important in research and practice. In this paper, the compensation of time-varying multiplicative/additive faults in both actuator and sensor for nonlinear systems with unknown nonlinear dynamics and disturbances is proposed. Actuator faults may result from one or more physical phenomena such as freezing or lock-in-place, float, hard-over-failure and loss of effectiveness. Also, sensor faults may result from one or more physical phenomena such as bias, drift, loss of accuracy, loss of effective-ness, and freezing. By the online estimation of upper bounds of the unknown disturbances, nonlinear dynamics, actuators and sensors faults, a descriptor-based sliding mode observer is implemented to estimate the states of the given system. The overall system stability is ensured using a robust fault tolerant with sliding mode control scheme. Two examples are studied, and simulation results are given to show the effectiveness of the proposed design method.