Data-Driven Equivalent Space Aero-Engine Sensor Robust Fault Diagnosis

Abstract

Aeroengine control system is the key to ensure the stable and efficient operation of the engine within the flight envelope. Due to long-term work in harsh environments such as high temperature and high pressure, strong vibration, and electromagnetic interference, sensors are the most prone to failure components in the control system, which has a huge impact on control performance. In this paper, the sensor in the aero-engine control system is taken as the research object, and the fault diagnosis method based on data-driven is carried out. Aiming at the dynamic process of small deviation of aero-engine, first construct the method of solving the equivalent space of the system through engine data simplification and reduce the order, then design the equivalent vector through the robust performance index, and obtain the residual error generation for sensor fault detection Finally, the equivalent space is properly transformed, and the sensor fault reconstruction scheme is designed through the robust performance index. Simulation results show that the designed robust fault estimation and reconstruction scheme is effective for single sensor faults, and the fault diagnosis scheme has certain robustness.