Abstract
The performance of different residual strength theories for composites subjected to fatigue loading is discussed in this article. Especially, the probabilistic aspects of strength degradation predictions are emphasized, with respect to the reliability levels defined using different approaches. A statistical model, using the `Strength-Life Equal Rank Assumption' (SLERA), is introduced and is shown to yield robust and accurate reliability bounds for residual strength after fatigue; this holds true even when used in conjunction with simple degradation concepts. Theoretical predictions are corroborated satisfactorily by experimental results from a comprehensive test program on [0]4 glass/epoxy composite laminates used in manufacturing of large, contemporary wind turbine rotor blades. It is shown that the use of complicated strength degradation theories requiring large experimental data sets for implementation, yields more accurate predictions than simple models requiring limited experimental effort. Nevertheless, the linear strength degradation model by Broutman and Sahu [1] performs satisfactorily, offering consistent predictions on the safe side and constitutes a strong candidate for engineering design applications.
Get full access to this article
View all access options for this article.
