International audience

Air data sensors provide essential measurements to ensure the availability of autopilot and to maintain aircraft performance, flight envelope protection and optimal aerodynamic surfaces control laws. The importance of these sensors imply the existence of embedded fault tolerance features, mainly represented by hardware redundancy. The latter is prone to fail in case of common fault of multiple sensors, especially if the faults are coherent and simultaneous. Increasing the robustness of fault detection and isolation (FDI) techniques for air data sensors to the aforementioned conditions is essential for the development of more autonomous aircraft, reducing crew workload and guaranteeing flight protections under adverse conditions. This paper reviews recent works on Air Data System (ADS) FDI, assessing proposed model, data and signal-driven approaches. We finally argue in favor of data-driven and hybrid approaches for the development of virtual sensors and semi-supervised anomaly detectors, offering an overview of ways forward.

This paper is concerned with an active fault diagnosis method and is developed based on an adaptive fault detection approach using$\chi^{2}$-statistics [1]. The system under consideration is linear discrete time-variant with sensor faults in the framework of mixed uncertainties (stochastic noises and bounded uncertainties for parameter matrices). The proposed method enhances the fault detection performance and provides the ability of localization and estimation of the detected fault. It provides also an on-line fault diagnosis with no delay in time instant with the computation time depending only on the computer performance.

In this paper, an improvement of the bounded-error fault detection and identification method based on input-output polynomials of ([2]) is proposed. It is based on integro-differential polynomials used to estimate the fault values. The standard input-output polynomials are obtained from differential algebra elimination theory and can be used both for diagnosability analysis and fault estimation. Unfortunately, they may involve derivatives of high order whose estimation is a hard problem when system outputs are uncertain. Distribution theory allows us to transform them into integro-differential polynomials that involve lower order derivatives of the model outputs. In this paper, this method, extended to the set-membership (SM) framework, is used with the focus of achieving fault detection and identification. The original method and the new method are applied to a coupled water-tank model and compared. It is shown that the new method significantly improves the fault detection and identification results.

Through parameter estimation schemes, one could be able to detect, localize and identify the occurring fault via simple computation. Yet, certain faults may not be discovered even be mistaken in a normal condition with unknown noises by trend checking or state monitoring. A more informative way when a correct model is present to analyses the data via parameter estimation. In this paper, we propose by using interval analysis a diagnosis scheme, from which we can extract the guaranteed diagnostic results to inform the supervisor so that appropriate actions could be taken. Sending them the results in a guaranteed way to tell the diagnostician which kind of fault exist is firstly taken care in diagnosis context. Our original fault detection and localization procedure has been firstly proposed in an interval analysis context for the constant fault in parameters. Moreover, another new technique in parameter estimation is the distance check, which speed up the estimation procedure. Some drawbacks have been discussed in the end.

International audience