We propose a new approach to the prediction of multiaxial fatigue with proportional and non-proportional loading based on a recently developed three-dimensional small-strain microplasticity-based constitutive theory. The core idea of the theory is to incorporate pre-full-yield microplastic deformations in a computationally efficient way. Fatigue life is then correlated to the accumulated (micro)plastic work, which is obtained from the total plastic dissipation. The constitutive parameters required are calibrated using just a monotonic stress-strain curve determined from a simple compression test experiment together with a low cycle uniaxial fatigue experiment. The resulting three-dimensional constitutive model has also been implemented into the Abaqus/Explicit finite element program through a vectorized user-material subroutine interface with a fully-implicit, unconditionally-stable and robust time integration scheme. Using the suitably-calibrated constitutive model, a series of uniaxial and multiaxial stress- and strain-based, constant-amplitude fatigue finite element simulations have been conducted to compare with the physical experiment data from the literature. It is shown that the developed theory and its finite-element implementation (which have fewer parameters than cycle counting based methods) are able to better predict the experimental fatigue life under multiaxial proportional or non-proportional loading conditions.

我们提出了一种新的方法来预测基于比例和非比例载荷的多轴疲劳，这是基于最近开发的三维小应变基于微塑性的本构理论。该理论的核心思想是以计算有效的方式合并预完全屈服的微塑性变形。然后，将疲劳寿命与累积的（微）塑性功相关联，这是通过总的塑性耗散获得的。所需的本构参数仅使用从简单的压缩测试实验和低周期单轴疲劳实验确定的单调应力-应变曲线进行校准。通过具有完全隐式，无条件稳定且鲁棒的时间积分方案的矢量化用户材料子例程接口，所得的三维本构模型也已实现到Abaqus / Explicit有限元程序中。使用适当校准的本构模型，进行了一系列基于单轴和多轴应力和应变的恒定振幅疲劳有限元模拟，以与文献中的物理实验数据进行比较。结果表明，改进的理论及其有限元实现（与基于循环计数的方法相比，其参数更少）能够更好地预测多轴比例或非比例载荷条件下的实验疲劳寿命。使用适当校准的本构模型，进行了一系列基于单轴和多轴应力和应变的恒定振幅疲劳有限元模拟，以与文献中的物理实验数据进行比较。结果表明，改进的理论及其有限元实现（与基于循环计数的方法相比，其参数更少）能够更好地预测多轴比例或非比例载荷条件下的实验疲劳寿命。使用适当校准的本构模型，进行了一系列基于单轴和多轴应力和应变的，恒定振幅疲劳有限元模拟，以与文献中的物理实验数据进行比较。结果表明，改进的理论及其有限元实现（与基于循环计数的方法相比，其参数更少）能够更好地预测多轴比例或非比例载荷条件下的实验疲劳寿命。