Experimental validation demonstrates online actuator fault detection in a water tank system, highlighting efficient identification via sliding-window ARX-Laguerre models.
In this paper, we aim to develop a diagnosis approach for the online detection and identification of actuator faults. This goal is achieved using the linear representation of the ARX–Laguerre model and its online identification on a sliding window at each time increment. The ARX–Laguerre model is obtained by projection of the input and the output of the classical ARX model on independent bases of Laguerre orthonormal functions. Laguerre functions have the property of being completely characterised by a parameter called the Laguerre pole and its optimal identification leads to a significant reduction in the number of Fourier coefficients. We are interested in the online identification of the model's parameters (Laguerre poles and Fourier coefficients) on a sliding window as well as the proposed online actuator fault estimation. This latter enables the development of an online actuator fault estimation algorithm which will be validated by application on a laboratory water tank system.
No takes yet. Share an insight, caveat, or question.
Yousfi et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: