The role of Kalman filtering in an inverse analysis of elasto-plastic material
The role of Kalman filtering in an inverse analysis of elasto-plastic material
复制标题
卡尔曼滤波在弹塑性材料反演分析中的作用
DOI：
10.2183/pjab.78.250
复制
发表时间：
2002
期刊：
影响因子：
--
通讯作者：
A. Murakami
中科院分区：
文献类型：
--
作者：
A. Murakami
本文提出了一种在控制边界条件（如荷载历史）的情况下识别土弹塑性本构模型参数的方法，并指出了卡尔曼滤波时间更新方案在逆分析中的作用。对于弹塑性材料，体的变形取决于加载历史、本构模型参数和初始应力，而只有当前应力状态决定弹性材料的变形。通过弹塑性模型的有限元计算，饱和土沉积物在不同的外部作用模式下，如路堤荷载和开挖，提供了不同的沉降曲线。然而，在没有精确的加载历史知识的情况下，很难确定弹塑性材料的某些参数。为了克服这一困难，研究了利用卡尔曼滤波的反问题公式，同时将瞬态加载历史纳入状态方程的确定性控制输入。
This paper presents an approach for identifying the parameters in elasto-plastic constitutive model for soil while controling the boundary conditions, e.g., the history of loading, and points out the role of the time update scheme of Kalman filtering in an inverse analysis. For elasto-plastic material, the deformation of the body depends on the loading history, the parameters of the constitutive model, and the initial stresses, whereas only the current stress state determines the deformation of elastic material. It is shown here that a saturated soil deposit provides various settlement curves under different patterns of external action, such as embankment loading and excavation, via finite element computation using an elasto-plastic model. Without precise knowledge of the loading history, however, it is difficult to identify some parameters of the elasto-plastic material. The formulation of an inverse problem, using the Kalman filter, is examined to overcome this difficulty, while incorporating the transient loading history into the deterministic control input of the state equation.