Process modeling and characterization of thermoset composites for residual stress prediction

A computational process modeling framework is presented to predict performance-altering residual stress generation at the microscale. A comprehensive material characterization effort is carried out as a function of the resin temperature and curing state, resulting in a novel material database. For a prescribed cure cycle, in-situ elastic modulus evolution, chemical and thermal strains, and random fiber distribution are shown to significantly influence residual stress generation. The results also show that a full process modeling analysis that includes the complete cure cycle (instead of the standard approach of just considering post-processing cool-down) is necessary to accurately predict manufacturing-induced residual stresses.