Computational intelligence methods for the fatigue life modeling of composite materials

Novel computational methods such as artificial neural networks, adaptive neuro-fuzzy inference systems and genetic programming are used in this chapter for the modeling of the nonlinear behavior of composite laminates subjected to constant amplitude loading. The examined computational methods are stochastic nonlinear regression tools, and can therefore be used to model the fatigue behavior of any material, provided that sufficient data are available for training. They are material independent methods that simply follow the trend of the available data, in each case giving the best estimate of their behavior. Application on a wide range of experimental data gathered after fatigue testing glass/epoxy and glass/polyester laminates proved that their modeling ability compares favorably with, and is to some extent superior to, other modeling techniques