Energy Dissipation Characterization and Design Methodology for Composite Materials

摘要

Various damages, which develop in structural composites, cause a softening behavior which can cause significant load redistribution in non- critical as well as critical areas of various structural elements. The strain energy dissipation (SED) concept provides a set of consistent nonlinear constitutive relations for stress analyses and the in-plane loader (IPL) provides a way for material characterization. Damage surfaces for moderate values of dissipated energy densities (DED) were estimated to match the response of conventional (+/-theta) sub ns coupon tests, which are influenced by matrix mode damages. Softening behaviors for high values of DED which can result from fiber breakage/pull out were estimated using engineering judgement. Nonlinear finite element analyses were performed for two classes of structural elements; namely. (i) a (+/- theta) sub ns cylindrical pressure vessel, where the nonlinearities in load- deformation responses are noticeable before fiber failure. and (ii) open hole tension of two quasi-isotropic layups, where local failure and softening in the fiber direction become critical to cause catastrophic failure. The results show the usefulness of the approach for design purposes and for explaining the complex problem of the so called 'hole size' effect. Extension of the SED concept to the 3-D problem is suggested and a new approach to the problem of certification is discussed. Usefulness of the data generated by the IPL could not be assessed since they were not available for use. However, it appears that such data would be useful if they can provide full information up to high DED levels required for complete softening due to fiber breakage and pull out. Additional studies are suggested for the following phases for utilizing the full potential of the approach. Use of the test devices (IPL) and/or other tests by other testing organizations are also recommended to determine their acceptability and versatility.