Iterative Learning for Nonlinear Systems with Variable Pass Lengths: Fault Estimation in the Presence of Actuator Faults

This research addresses the faults estimation (FE) of the actuator faults in nonlinear dynamic systems with varying pass lengths. Additionally, the adverse impact of the bounded disturbances and measurement noise on the systems is also considered. A nonlinear iterative learning observer (NILO) is designed to achieve fault estimation. To tackle information loss and redundancy caused by variable pass lengths, a novel enhanced weighted averaging operator for the NILO is proposed. This approach enables fast tracking of fault signals and enhances system robustness. The method ensures the final boundedness of tracking error. Finally, the simulation results are given to verify that the proposed approach is validated by numerical simulation studies.