Incorporation of a general strain-to-failure fracture criterion into a stress-based plasticity model through a time-to-failure softening mechanism.

摘要

Many classic and contemporary fracture models are based on some variant of strain-to-failure with linear accumulation of damage. These models are categorized as strain-to-failure models, even if the damage weighting function is stress-based. Recent experimental investigations suggest that strain-to-failure fracture models are a natural choice when modeling metals. Notably, the third stress invariant (J3) dependence of strain-to-failure has been shown to be nonnegligible. In response to the metal-fracture literature proposing a multitude of new strain-to-failure fracture models with little demonstration of predictiveness in large-scale general-loading simulations, this research implements a strain-to-failure framework into a generalized plasticity model, Kayenta, tested in conjunction with three representative fracture models: constant equivalent-strain-to-failure, Johnson-Cook strain-to-failure theory, and Xue-Wierzbicki strain-to-failure theory. These models constitute a sampling of J2, J3, strain-rate, and temperature dependence that greatly extend the softening options available in Kayenta. As Kayenta is portable and already available in multiple host codes, this research allows analysts to rapidly gauge which failure theory is best suited to their applications, thus potentially allowing one of these theories to emerge as more broadly valid in general loading problems. This fracture framework is designed to operate in the realm of time-to-failure so as to function seamlessly with the current softening implementation in Kayenta and lay the foundation for mixed-response fracture behavior to transition between ductile to brittle fracture models dynamically as the stress state evolves.