Data-Driven Fault-Tolerant Control for Discrete-Time Systems based on LMI

Abstract

Actuator failure of control systems is one of the main factors causing instability and performance degradation. Over the past few decades, fault-tolerant control strategies have been introduced to improve the performance of control systems against failure. The current fault-tolerant control approaches are mostly model-based, however, model identification is time-costly and costly. In this paper, a data-driven fault-tolerant control method for discrete-time systems with actuator faults is designed. Using the input and output data collected directly from the system, the data-driven method develops the fault-tolerant control strategy for discrete-time systems with unknown or complex model. Moreover, the stability condition of the strategy is derived based on linear matrix inequality (LMI). Finally, a simulation is presented to demonstrate the effectiveness of the data-driven fault-tolerant control strategy, using an aircraft dynamic system model with actuator failure.