Neural networks can be used to predict failure of anisotropic materials under any loading condition, provided that sufficient expenmental data were initially available for the system to "learn" the rules. An example failure envelope for a typical fibre-reinforced material is obtained using this method.
Get full access to this article
View all access options for this article.
References
1.
Tsai, S.W. and E.M. Wu1971. "A General Theory of Strength for Amsotropic Materials,"Journal of Composite Materials, 5:58-80.
2.
Labossière, P. and N. Turkkan1992. "On the Optimization of the Tensor Polynomial Failure Theory with a Genetic Algorithm,"Transactions of the Canadian Society of mechanical Engineering , 16.251-265.
3.
Pao, Y.-Y.1989Adaptative Pattern Recognition and NeuralNetworks Addison-Wesley.
4.
Rumelhart, D.E. , G.E. Hinton and R.J. Williams1986"Learning Representations by Back-Propagating Errors,"Nature, 323.533-536.
5.
Rowlands, R.E. , D.E. Gunderson, J.C. Suhling and M.W. Johnson.1985. "Biaxial Strength of Paperboard Predicted by Hill-Type Theories,"Journal of Strain Analysis, 20:121-127
6.
Suhling, J.C. , R.E. Rowlands, M.W. Johnson and D.E. Gunderson1985"Tensorial Strength Analysis of Paperboard,"Experimental Mechanics, 2575-84.
7.
Labossière, P. and K.W. Neale1987"On the Determination of the Strength Parameters in the Tensor Polynomial Failure Criterion,"Journal of Strain Analysis, 22.155-161
8.
Labossière, P. and K.W. Neale1991"A Parametric Failure Criterion for Paperboard Material,"Journal of Strain Analysis, 266:461-466.
9.
BrainMaker. 1989. California Scientific Software, Grass Valley, CA 95945, U.S.A.
