建立岩土本构模型的数值方法

根据反问题的理论,阐明了建立岩土本构模型属于反问题,提出了建模的四个基本步骤.在试验数据中蕴藏着应力应变关系的信息,直接提炼本构规律是可能的.神经网络是强有力的反演工具,借助于它建立了砂土和粘性土的非线性和弹塑性神经网络模型.研究结果表明,这种数值建模方法具有许多优越性,已成为求解边值问题的主要工具,也必将在建立岩土本构模型中发挥重要作用.

Numerical Method in Modeling the Constitutive Relations of Rock and Soil

Based on the inverse problem theory, it is explained that building the constitutive models of rock and soil is an inverse problem and four basic steps for modeling constitutive relations are proposed. The information on stress strain relations is contained in the test data, so it is possible to abstract the constitutive law directly from that. The neural network is a powerful inversion approach, by means of which the nonlinear and elastoplastic neural network models for sand and cohesive soil are built. The research results show that the numerical methods for modeling constitutive law have many advantages and have been a main means for solving boundary value problems, yet it would play an important role in modeling the constitutive law of rock and soil.