Neural Network Techniques for Fault Identification of Aircraft Sensors and Actuators

Aviation technics and technology


Tiumentsev Y. V.*, Kozlov D. S.**

Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia



This paper focuses on an algorithm of fault identification based on an appropriate combination of identification task for an aircraft simulation model together with classification task for a failure event. Neural network based techniques are suggested to solve identification and classification tasks as applies to the fault diagnostics problem. The results of simulation for sensor and actuator faults with regard to a fighter aircraft are presented to demonstrate efficiency of the proposed approach.


failure diagnostics, nonlinear autoregressive network, identification problem, classification problem, aircraft

Download — informational site MAI

Copyright © 2000-2021 by MAI