This paper discusses the efficiency of an algorithm based on the Asymptotic Numerical Method (ANM) to solve large strain plasticity problems. In the framework of ANM, the non-smooth constitutive law has to be replaced by a smooth one in order to be able to represent the solution path in the form of Taylor series. For this purpose, we propose to generalize the approach used in the small strain case. To achieve this, we introduce a regularized stress–strain relation that considers smoothed elastic–inelastic transitions. This regularized law is a relationship between two scalars, namely the yield function and the plastic multiplier. Classical uniaxial traction benchmarks permit to assess the procedure and to adjust the parameters of the algorithm: regularization parameters, truncation orders and parametrization.

本文讨论了基于渐近数值方法 (ANM) 的算法解决大应变塑性问题的效率。在 ANM 的框架中，非光滑本构律必须被一个光滑的本构律代替，以便能够以泰勒级数的形式表示解路径。为此，我们建议推广小应变情况下使用的方法。为了实现这一点，我们引入了考虑平滑弹性-非弹性转变的正则化应力-应变关系。该正则化定律是两个标量之间的关系，即屈服函数和塑性乘数。经典的单轴牵引基准允许评估程序并调整算法的参数：正则化参数、截断阶数和参数化。