Annotation

Annotation: Modern and future aircraft are characterized by an increased degree of electrification of onboard systems and the use of electromechanical systems to ensure critical functions of the aircraft flight control system. Widespread implementation of onboard electromechanical systems requires their safe and efficient operation, for which purpose it is assumed that the aircraft is equipped with an early diagnostics system for faults. Electromechanical systems of an aircraft, the closed control loop of which is implemented based on feedback signals for the speed or movement of the output link, are subject to high requirements for the quality of regulation, including under the action of mechanical load. Deterioration of the characteristics of an electromechanical system is associated with degradation processes that develop in individual components and can be determined based on data obtained both in the passive mode during a typical flight and in the active mode when processing specified control signals between flights in ground conditions. Modern artificial intelligence methods allow their effective use in creating algorithms for early diagnostics of electromechanical systems for classification and forecasting of technical condition using big data containing valuable information about work processes associated with the development of faults. To obtain such data, the main scenarios of operation of electromechanical systems were studied in the work, allowing to obtain representative samples. To ensure an automated data collection process using signals transmitted via standard interfaces of information interaction of the integrated control system with end devices, as well as from additionally used sensors, data accumulation methods were developed and studied using the results of modeling, testing and operation. Based on the obtained data, it is possible to assess the degradation of static and dynamic characteristics of electromechanical systems and determine their significance in classifying the technical condition for localizing malfunctions and performing condition-based maintenance. For this purpose, methods for processing the obtained data were developed and studied and the practical application of the results for subsequent on-board integration of the diagnostic system was assessed.