Artificial Neural Network model for prediction of fatigue lives of composite materials
Loading...
Date
2006-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand)
Abstract
The application of composites as engineering materials has become state of art. Fatigue is one of the most complicated problems for fiber composites. Fatigue tests are expensive and time consuming. In all testing it is need to define a failure criterion in use. This is particularly difficult for fiber reinforced plastics.
A potential solution to this problem is offered by artificial neural networks (ANNs). ANNs are an alternative to conventional programmed computing and are based on the operation of the human brain.
In the present study an ANN model is developed for fatigue life prediction of carbon fiber reinforced plastics of two different lay up. The input to ANN model are lay up , volume fraction, monotonic properties, applied load parameters and probability of failure. The output of the ANN is logarithmic value of fatigue life cycles. The architecture selected has two hidden layers with 18 and 6 units each. The learning rate coefficient η = 0.9 and momentum factor α = 0.3 is selected.
The predictions of fatigue life by ANN are comparable to experimental values. The percentage error in more than 92 percent cases is found to be less than 10 percent. The study of the effect of various monotonic properties on fatigue life prediction by ANN is studied. The trained ANN model is applied to predict the fatigue life of new materials of other group
Description
Keywords
null