BALL SCREW SURFACE FAULTS MODELING AND DETECTION FOR ELECTROMECHANICAL FLIGHT CONTROL ACTUATORS

Ball screw faults are the major contributors to the known safety limitations of electromechanical actuators (EMA), which hinder their wide adoption as the main technology for primary flight controls in the aerospace sector. To overcome this issue, a promising approach is represented by the construction of a Prognostic and Health Management (PHM) framework around a simple architecture EMA to monitor its health state. A high-fidelity model of the entire servosystem represents a valuable tool for generating synthetic databases with a variety of simulated defects to train PHM routines. This paper introduced a detailed dynamic model of an EMA, designed to facilitate model-based PHM routines, composed of a three-phase motor description and a high-fidelity multibody model of the ball screw, capable of representing the full dynamics of each internal sphere independently. This allowed the description of a localized defect model implemented on the nut raceway to simulate superficial defects of various entities. The impact of such faults on the system's response is explored by analyzing the NA-MEMD based HHT spectra of ball screw kinematic signals and internal contact forces on the various spheres. The results verified the efficacy of the proposed model and open avenues for more comprehensive analyses and future investigations on defect detectability and isolation.