Search

The development of adequate diagnostic/prognostic methodologies, suitable to provide a timely and reliable evaluation of the health status of a given system on the basis of some representative parameters (measured in a direct or indirect way), is fundamentally started in engineering fields, but, especially in recent years, it is encountering more and more interest and application in many technical fields and nowadays it represents an important task in various scientific disciplines. The health status of a given dynamic system (e.g. environmental, mechatronic, structural, etc.) and the eventual incipient failures that concern it, especially if related to progressive evolutions, can be identified and quantified by means of different approaches widely described in the literature. It must be noted that, particularly in recent years, there has been a strong impulse in the development of strategies aimed to design prognostic algorithms able to identify precursors of the progressive failures affecting a system: in fact, if it is correctly identified the degradation pattern, an early warning can be triggered, leading to proper corrective actions (i.e. proper remedial or maintenance tasks, replacement of the damaged components, etc.). Since these algorithms are strictly technology-oriented, they can show great effectiveness for some specific applications, while they may fail for other applications and technologies: therefore, it is necessary to properly conceive the specific prognostic method as a function of several parameters such as the given (dynamic) system, the available sensors (physical or virtual), the considered progressive failures and the related boundary conditions. This work proposes a critical comparison between several diagnostic/prognostic strategies in order to put in evidence their strengths and the eventual shortcomings. [ABSTRACT FROM AUTHOR]

The proposal of prognostic algorithms able to identify precursors of incipient failures of primary flight command electromechanical actuators (EMA) is beneficial for the anticipation of the incoming failure: an early and correct interpretation of the failure degradation pattern, in fact, can trig an early alert of the maintenance crew, who can properly schedule the servomechanism replacement. An innovative prognostic model-based approach, able to recognize the EMA progressive degradations before his anomalous behaviors become critical, is proposed: the Fault Detection and Identification (FDI) of the considered incipient failures is performed analyzing proper system operational parameters, able to put in evidence the corresponding degradation path, by means of a numerical algorithm based on spectral analysis techniques. Subsequently, these operational parameters will be correlated with the actual EMA health condition by means of failure maps created by a reference monitoring model-based algorithm. In this work, the proposed method has been tested in case of EMA affected by combined progressive failures: in particular, partial stator single phase turn to turn short-circuit and rotor static eccentricity are considered. In order to evaluate the prognostic method, a numerical test-bench has been conceived. Results show that the method exhibit adequate robustness and a high degree of confidence in the ability to early identify an eventual malfunctioning, minimizing the risk of fake alarms or unannounced failures. [ABSTRACT FROM AUTHOR]