Prediction of long-term behaviour of composite materials

Citation
Rm. Guedes et al., Prediction of long-term behaviour of composite materials, COMPUT STRU, 76(1-3), 2000, pp. 183-194
Citations number
38
Categorie Soggetti
Civil Engineering
Journal title
COMPUTERS & STRUCTURES
ISSN journal
00457949 → ACNP
Volume
76
Issue
1-3
Year of publication
2000
Pages
183 - 194
Database
ISI
SICI code
0045-7949(200006)76:1-3<183:POLBOC>2.0.ZU;2-4
Abstract
After a brief introduction, where the question of durability related to com posite materials is raised concerning the type of applications and the cons tituents of the composite material, some considerations are given about the interaction between physical and chemical ageing and its effect on the lon g-term behaviour of those materials. A comparison of different methodologie s to define models for the constitutive laws of the material is made, toget her with a reference to methodologies used for real structures in order to predict long-term behaviour. An algorithm to predict long-term laminate pro perties of fibre reinforced composites is presented. In the procedure, the classical laminate plate theory was extended to include time-related respon se of composite materials for membrane and bending load. An efficient semi- direct time integration scheme, providing a stable integration process, was derived to be included in the numerical procedure. The present formulation enables solving problems related with creep, stress relaxation and rate de pendent stress-strain behaviour for in-plane and bending loads. The results concerning studies related to the influence of water and temperature on th e long-term behaviour of fibre reinforced composite, hence on the constitut ive models, are presented. The experimental results shows that a combined t ime-temperature-moisture superposition principle (TTMSP) is not valid. This principle simply states that the effects of temperature and moisture on th e viscoelastic properties can be uncoupled. (C) 2000 Elsevier Science Ltd. All rights reserved.