Failure of carbon/epoxy composite tubes under combined axial and torsional loading 1. Experimental results and prediction of biaxial strength by the use of neural networks

Abstract

Biaxial tests have been conducted on cross-ply carbon/epoxy composite tube under combined torsion and axial tension/compression up to failure. Strength properties and distributions were evaluated with reference to the biaxial loading ratio. The scatter of the biaxial strength data was analyzed by using a Weibull distribution function. Artificial neural networks were introduced to predict failure strength by means of the error back-propagation algorithm for learning, providing a different and new approach to the representation of complicated behavior of composite materials. Further prediction is made from experimental data by the use of Tsai–Wu theory and a combined optimized tensor polynomial theory. Comparison shows that the artificial neural network has the smallest root-mean-square error of the three prediction methods.