This study provides a solution for robust and early detection of Oscillatory Failure Cases in aircraft control surface servo loops. Extending the results from literature, an improved model-based monitoring approach ba...
详细信息
This study provides a solution for robust and early detection of Oscillatory Failure Cases in aircraft control surface servo loops. Extending the results from literature, an improved model-based monitoring approach based on a nonlinear actuator model is presented including novel band-pass filters, sliding windows, and an enhanced signal processing method for faster detection of high-frequency OFC. To specifically cope with the effects of actuator model uncertainties, filter time delay, and noise, an integrated method combining model-based and data-drivenmethods employing neural networks, distance-complexity correlation, and information entropy is proposed. Verification of this novel scheme is carried out using an industrial benchmark provided by Airbus. The promising results indicate that the proposed scheme is able to detect any OFC ensuring a very low false alarm rate, even in the presence of uncertainties. Copyright (c) 2024 The Authors.
暂无评论