Proposal of a model based fault identification neural technique for more-electric aircraft flight control EM actuators

There are many different ways to detect incipient failures of electromechanical actuators (EMA) of primary flight command provoked by progressive wear. With the development of a prognostic algorithm it’s possible to identify the precursors of an electromechanical actuator failure, to gain an early alert and so get a proper maintenance and a servomechanism replacement. The present work aims to go beyond prognostic algorithms strictly technology-oriented and based on accurate analysis of the cause and effect relationships because if on one hand they show great effectiveness for some specific applications, instead they mostly fail for different applications and technologies. Through the development of a simulation test bench the authors have demonstrated a robust method to early identify incoming failures and reduce the possibility of false alarms or non-predicted problems. Authors took into account friction, backlash, coil short circuit and rotor static eccentricity failures and defined a model-based fault detection neural technique to assess data gained through Fast Fourier Transform (FFT) analysis of the components under normal stress conditions.