Prediction of low cycle fatigue life of short fibre composites at elevated temperatures using surrogate modelling

The performance of any approximation scheme is known to be largely dependent on the type of surrogate models and its corresponding input variables. Elastic modulus prediction of short fibre composites with polyester and vinylester is presented using the surrogate framework supported by multiple spatially distributed surrogate models of different types. Poisson's ratio is included as an additional variable to predict low cycle fatigue damage of short-fibre composites at elevated temperatures. The validation of the surrogate model is established through a comparison with analytical results. Furthermore, the surrogate model is coupled with an optimization algorithm to solve a number of inverse problems which is of significant interest to the end users in practice.