Use of Artificial Neural Networks for Real Time Determination of Loads on In-Service Components

 

Dr S Noroozi, Dr R Amali, Dr J Vinney and Dr P Sewell

Faculty of Computing, Engineering and Mathematical Sciences, University of the West of England, Bristol, BS16 1QY, UK
Tel: 0117 3442628, E-mail: siamak.noroozi@uwe.ac.uk

Abstract

The monitoring of loads on in-service components or structures is a necessity to determine the overall structural integrity and life expectancy. Although it is routine in some cases, there are many cases where this task is difficult or until now an impossibility (i.e. the aerodynamic load fluctuations on aircraft components). 

Research at the University of the West of England (UWE) aims to develop an Inverse Problem Engine (IPE), which can be configured to solve any inverse problem. In this study an IPE is used to determine the magnitude and location of loads on a simplified model aircraft wing. 

This hybrid system, once trained, has been found to give an acceptable solution to any problem bound by the training envelope, including those situations not experienced by the system. This paper concludes by discussing the potential for using this approach as a real time tool for continuous monitoring of structural loads. 
. 


To read the full article,
please log in:
If you haven't registered already,
you can do so for free:
 
© Copyright 2000-2004 Professor F.R. Hall & Dr I. Oraifige, University of Wolverhampton.
 
close this window