Health Management of a Typical Small Aircraft Fuel System Using an Adaptive Technique

Abstract

Faults in the aircraft fuel system will degrade its performance and may lead to the complete system failure. In commercial aircraft system, efficient diagnosis can optimize the time to return the aircraft to service, thus allowing less disruption to passenger travel. In this work, an adaptive fault diagnosis technique is developed for a typical small aircraft fuel system, which facilitates efficient learning procedure to forecast the system parameters for non-linear situations. This adaptive technique represents the integration of the Fuzzy Logic and Support Vector Machine (SVM) algorithms in the field of fault diagnosis. Using this adaptive technique health monitoring of aircraft fuel system is discussed. In an aircraft fuel tank, the fault is effectively located by assessing and contrasting the actual parameters and set point parameters related to the system for various time-frames. The fuzzy logic controller is configured with the logical rules as per the required target output. It relies on the aircraft fuel system parameters like the fuel flow rate, level of fuel in the tank, fuel temperature, and fuel pressure. From the logical rules, the control signals related to the aircraft fuel system are derived by the SVM technique. The efficiency in execution of this fault diagnosis tool-based aircraft fuel system gets authenticated in the MATLab/Simulink platform. The simulation is carried by assuming normal operating conditions of aircraft in the laboratory environment.