Experimental Characterization and Modeling of Nonlinear Effects in Electromechanical Actuators for Prognostic Capabilities

With the advent of modern power electronics and the advancement of aircraft system technologies—along with the need for reduced operating costs and emissions—the new concept of More Electric Aircraft (MEA) has gained traction in the aerospace industry. This paradigm aims to replace all non-propulsive systems onboard the aircraft with electrically powered alternatives. In this context, both primary and secondary flight control systems must meet the demanding performance and reliability standards traditionally achieved by high-power hydraulic actuators, adopting instead a new approach based on Electromechanical Actuators (EMAs). The work presented in this thesis aims to implement design modifications on an experimental EMA test bench, enabling it to accurately reproduce various operating conditions typical of these servomechanisms. Representative phenomena such as gear backlash and actuation friction are introduced into the test setup in a controlled manner, by means of the linear displacement of two precision micro-movers. In addition, mutual validation with a simulation model was considered essential for gaining insight into the behavior of such systems. The test bench assembled can be broadly divided into two functional sections. The first is responsible for generating high-torque, low-speed power, using a conventional synchronous electric motor-reduction gearbox assembly. An incremental encoder is connected to this assembly via a gear interface that allows adjustable backlash and closes the position feedback control loop. The second section is dedicated to generating a friction torque on the actuation line through a brake-disc mechanism. This allows for the regulation of the distance between the brake pad and the disc, while the torque produced is measured via a load-cell lever system. An experimental campaign was conducted to accurately measure and map the key monitored variables and compare them with the outputs of the system’s simulation model. The variables analyzed were sampled during system operation, and the observed trends confirmed both the non-linear nature of these phenomena and a good degree of accuracy of the model under nominal conditions. The updated test bench–model framework lays the groundwork for future model-based prognostic capabilities, potentially enhancing both the understanding and reliability of modern EMA systems.