Sviluppo di un algoritmo per la Manutenzione Predittiva nell'ambito dei sistemi generali del velivolo = Development of an algorithm for Predictive Maintenance in aircraft general systems

Reliability and predictive maintenance play a crucial role in the aerospace sector, where safety and operational efficiency depend on the ability to predict and mitigate the degradation of critical components. In this context, prognostics emerges as an innovative approach for estimating the Remaining Useful Life (RUL) of a component and for planning targeted maintenance interventions, reducing operational costs and improving system availability. This thesis first introduces the concept of prognostics, analyzing its main methodologies and potential future applications in the field. The integration of prognostic algorithms, employing artificial intelligence, into modern aircraft represents a crucial step toward the development of more efficient maintenance strategies capable of extending component useful life without compromising safety. The primary objective of this study is to develop an algorithm for estimating the RUL of an electric motor within an electromechanical actuator (EMA) used in the flight control systems of a regional aircraft. The reference model is a simplified version of the one described in "Simulation of an All-Electric Flight Control System for the Evaluation of Power Consumption", which analyzes the power requirements of electromechanical actuators used to control aircraft surfaces. The focus is specifically on the three-phase brushless synchronous motor within the EMA, investigating the degradation of resistance and inductance, key parameters in monitoring operational conditions. Due to the lack of real data and established empirical models for the degradation of these parameters, synthetic data was generated based on reasonable assumptions taken from existing literature. The developed Matlab code utilizes specific toolboxes to extract meaningful features from the data and formulate an RUL prediction for the analyzed component. The primary goal of this study is not to achieve definitive and applicable results but rather to establish a structured methodological process that can serve as a foundation for future advancements in predictive maintenance.