DATA DRIVEN MODELLING OF PEEK MODULUS FOR ADVANCED PROCESS SIMULATION

Abstract

The development of large-scale thermoplastic composite parts using conventional trial and error methods is prohibitively expensive due to the complex interaction of part geometry, layup and temperature history with residual stresses and subsequent deformation. Process simulation can be the key to making thermoplastic materials accessible, but is ultimately limited by the accuracy of the underlying material models. In this work, data fusion of mechanical and thermal test results is employed to correlate the modulus of neat PEEK to the crystallinity. Based on this database, a modulus model is developed as a function of the state variables of temperature and crystallinity.