Meta-modeling of the curing process of thermoset matrix composites by means of a FEM-ANN approach

Thermal curing is a common practice to manufacture high temperature thermosetting matrix composites, in order to improve the mechanical properties of the final product. The cycle design i.e. the definition and optimization of the temperature-time curve is a key issue for a competitive production. In this framework, a suitable model describing the composite temperature and degree of cure variations versus the imposed thermal cycle is highly desirable. An effective procedure, based on the coupling of a finite element thermochemical model of the process and an artificial neural network, is herein proposed and tested. Obtained outcomes highlight the remarkable capabilities of the implemented procedure in terms of reliability of temperature and degree of cure predictions.