Predicting the Fatigue Life of Different Composite Materials Using Artificial Neural Networks

Abstract

Artificial Neural Networks (ANN) have been recently used in modeling the mechanical behavior of fiber-reinforced composite materials including fatigue behavior. The use of ANN in predicting fatigue failure in composites would be of great value if one could predict the failure of materials other than those used for training the network. This would allow developers of new materials to estimate in advance the fatigue properties of their material. In this work, experimental fatigue data obtained for certain fiber-reinforced composite materials is used to predict the cyclic behavior of a composite made of a different material. The effect of the neural network architecture and the training function used were also investigated. In general, ANN provided accurate fatigue life prediction for materials not used in training the network when compared to experimentally measured results.

Using Artificial Neural Networks to Predict the Fatigue Life of Different Composite Materials Including the Stress Ratio Effect

“…It should be noted that, in this study, since the composite might be subjected to compressive loads, compressive laminate properties had to be included as input parameters to the network. This was not the case in [18] when the material was only subjected to a tension-tension fatigue load. Since the range of fatigue life varied between 3 and 10,000,000 cycles, training the networks to learn such a wide range will produce unacceptable and unbalanced modeling performance.…”

“…In a recent study, Al-Assadi et al [18] used ANN to predict the fatigue life for composite materials other that those used for training. Seven different materials were used in training the network for the purpose of predicting the fatigue behavior of an eighth material.…”

“…Figures 2 and 3 show typical variations of the RMSE obtained using FNN for two of the materials considered as a function of the number of hidden neurons. For all cases considered [29], a number of hidden neurons between 6 and 12 lead to the most accurate fatigue life prediction. Figures 4,5,6,7,8,9 and 10 show the relation between the maximum applied stress and the number of cycles to failure for some of the materials considered.…”

Fatigue Life Prediction of Different Fiber-Reinforced Composites Using Polynomial Classifiers

“…Using a first order PC, a RMSE of the order of 50% was obtained [12]. For this case, the PC predicted a nearly constant value for the fatigue life irrespective of the maximum applied stress and the fiber orientation angle.…”

Comparative study between the PNL method and a MN in modelling fatigue of composite materials

“…Another approach demonstrated in recent studies 4–6 is the use of artificial neural networks (ANN). This form of modelling has different characteristics compared to the models mentioned earlier.…”