Detection of Helicopter Rotor System Simulated Faults Using Neural Networks

Abstract

Simulated fault data from a mathematical model of a damaged rotor system is used to develop a neural network-based approach for rotor system damage detection. The mathematical model of the damaged rotor is a comprehensive rotorcraft aeroelastic analysis based on a finite element approach in space and time. Selected helicopter rotor faults are simulated through changes in inertial, damping and stiffness properties of the damaged blade. A feed-forward neural network with back-propagation learning is trained using both 'ideal' and 'noisy' simulated data. Testing of the trained neural network shows that it can detect and identify damage in the rotor system from simulated blade response and vibratory hub loads data. (Author)