【摘要】 基于人工神经网络和反问题理论 ,将不同应力路径下砂土的三轴实验结果作为样本来训练网络 ,获得了砂土的神经网络本构模型 ,并计算了某工程基础的沉降。分析结果表明 ,随着围压增大 ,沉降减小 ,因此模型能够定量的反映出应力路径及围压对地基沉降的影响 ,从而为实际工程进行不同应力路径的数值模拟和仿真提供了可能。更多还原

【Abstract】 The network is trained by using the samples obtained from triaxial tests to set up the neural network constitutive model.Then caculate the settlement of foundation.It is shown as the surrounding stress increases,the settlement of foundation is reduced.Therefore,the network model can quantitatively reflects the influence of stress path and surrounding stresses on settlement of foundation,which makes it possible for numerical simulation of actual engineering under different stress path.更多还原