摘要
根据密封舱改性渣土的三轴压缩实验数据,建立了基于优化方法的Duncan-Chang双曲线本构模型参数反演方法。为了解决参数识别反问题解的不稳定性问题,在Gauss-Newton优化迭代算法的基础上,提出了正则化参数反演方法。实际应用结果表明,所提出的反演方法具有较快的收敛速度和较强的鲁棒性,预测的实验模型变形与实验值吻合较好。
The parameter inversion procedure for the Duncan-Chang constitutive model of conditioned soils was proposed based on the data obtained from the compression tree-axial test in the laboratory.In order to overcome instability of parameter identification,the regularization procedure was applied based on the Gauss-Newton optimization algorithm.The investigation showed that the proposed inversion procedure had higher convergent speed and better robustness,and the forecasted deformations agreed well with the observed data.
出处
《山东大学学报（理学版）》
CAS
CSCD
北大核心
2010年第7期24-27,33,共5页
Journal of Shandong University(Natural Science)
基金
国家重点基础研究发展规划项目(2007CB714006)
国家自然科学基金资助项目(90815023)
关键词
参数反演
本构模型
渣土
优化方法
parameter inversion
constitutive model
conditioned soil
optimization algorithm
