Sparse Bayesian polynomial chaos approximations of elasto-plastic material models
Sparse Bayesian polynomial chaos approximations of elasto-plastic material models
复制标题
弹塑性材料模型的稀疏贝叶斯多项式混沌近似
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
H. Matthies
中科院分区：
文献类型：
--
作者：
B. Rosic;H. Matthies
本文研究了材料不确定性参数弹塑性模型的泛函近似形式下的不确定性量子fi正离子。通过考虑解的可能稀疏性，将多项式混沌Coeffi指数的估计问题重演为线性回归形式。从经典优化的观点出发，我们选择了一条略微不同的路径，在新的基于谱的稀疏fffi算法的帮助下，以贝叶斯的方式求解该问题。
In this paper we studied the uncertainty quantification in a functional approximation form of elastoplastic models parameterised by material uncertainties. The problem of estimating the polynomial chaos coefficients is recast in a linear regression form by taking into consideration the possible sparsity of the solution. Departing from the classical optimisation point of view, we take a slightly different path by solving the problem in a Bayesian manner with the help of new spectral based sparse Kalman filter algorithms.