Title :
Explaining how a multi-layer perceptron predicts helicopter airframe load spectra from continuously valued flight parameter data
Author :
Vaughn, M.L. ; Franks, J.G.
Author_Institution :
Cranfield Univ., Swindon, UK
Abstract :
This study explains how a multi-layer perceptron predicts helicopter airframe load spectra from continuously valued flight parameter data for 2 example regression outputs - a high strain case and a low strain case. In each case, the input channels are discovered that determine the output activation values. The underlying mechanism that drives the MLP regression output for the two cases is determined.
Keywords :
aerospace control; aerospace simulation; helicopters; multilayer perceptrons; prediction theory; regression analysis; MLP regression output; continuously valued flight parameter data; helicopter airframe load spectra prediction; high strain case; low strain case; multilayer perceptron; Capacitive sensors; Data mining; Helicopters; Multilayer perceptrons; Neural networks; Neurons; Predictive models; Standards development; Strain measurement; Testing;
Conference_Titel :
Neural Networks, 2003. Proceedings of the International Joint Conference on
Print_ISBN :
0-7803-7898-9
DOI :
10.1109/IJCNN.2003.1223837
