Electromechanical actuators affected by multiple failures: a simulated-annealing-based fault identification algorithm

The identification of early evidences on monitored parameters allows preventing incoming faults. Early alerts can avoid rate of the failures and trigger proper out-of-schedule maintenance activities. For this purpose, there are many prognostic approaches. This paper takes into account a primary flight command electromechanical actuator (EMA) with multiple failures originating from progressive wear and proposes a fault detection approach that identifies symptoms of EMA degradation through a simulated annealing (SA) optimization algorithm; in particular, the present work analyses the functioning of this prognostic tool in three different fault configurations and it focuses on the consequences of multiple failures. For this purpose, we developed a test bench and obtained experimental data necessary to validate the results originated from the model. Such comparison demonstrates that this method is affordable and able to detect failures before they occur, thus reducing the occurrence of false alarms or unexpected failures. © 2016, North Atlantic University Union. All rights reserved.