A neural network approach to load identification on a wing rib

S.B. Cooper, D. Di Maio

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review


Monitoring of in-service flight loads on aircraft structures has become a safety measure used in developing a better understanding of the aircraft behaviour during real time flight operation. However the increase in the complexity of these structures increases the difficulty in obtaining accurate and acceptable data for flight parameters. The use of advanced mathematical techniques (i.e. artificial neural networks - ANNs) to accurately predict the load on these structures based on measured parameters have proven to be of great advantage. The ANN is a powerful machine learning tool which has the capacity to predict the relationship between variables through an adaptive learning process. In this paper, a method for predicting the static load applied across an aluminium plate is presented and the method is based on a combination of the finite element method and ANNs. The finite element model of the plate was calibrated using ANN trained data generated from a static test. Finally, the results indicate that this technique can provide load identification across a structure once the load-response relationship of the structure has been identified from the ANN training.

Original languageEnglish
Title of host publicationProceedings of the Fourth International Conference on Soft Computing Technology in Civil, Structural and Environmental Engineering
EditorsY. Tsompanakis, J. Kruis, B.H.V. Topping
Place of PublicationStirlingshire, UK
PublisherCivil-Comp Press
Publication statusPublished - 2015
Externally publishedYes

Publication series

NameCivil-Comp Proceedings
PublisherCivil-Comp Press
ISSN (Print)1759-3433


  • Artificial neural network
  • Finite element method
  • Load identification
  • Structural monitoring


Dive into the research topics of 'A neural network approach to load identification on a wing rib'. Together they form a unique fingerprint.

Cite this