当前位置:主页 > 科技论文 > 地质论文 >

地震液化灾害自适应步长计算方法及控制参数研究

发布时间:2018-08-07 19:07
【摘要】:饱和砂土液化是岩土地震工程和土动力学研究领域的重要课题。在动力液化数值计算中,计算精度和计算效率一直是衡量数值方法的重要指标。在水土二相耦合弹塑性计算的数值平台上开发了自适应步长方法,通过位移误差、孔压误差和混合误差的评估体系建立了时间步长自动调整的策略及相关控制参数。通过控制参数的影响性分析,确定了主要控制参数为误差允许值和孔压误差比例系数,辅助控制参数为初始时间步长、步长调整的下限和上限。对处于液化场地的地铁车站模型进行了动力自适应步长数值计算,获得了上浮位移和超孔压比的发展过程,预测出液化上浮的灾害。同时也对比了固定步长法和自适应步长法的精度和效率,发现采用自适应步长法可极大的节约计算成本并不失计算精度。
[Abstract]:Saturated sand liquefaction is an important subject in the field of geotechnical seismic engineering and soil dynamics. In the numerical calculation of dynamic liquefaction, the calculation accuracy and efficiency are always the important indexes to measure the numerical method. The adaptive step method is developed on the numerical platform of soil and water two-phase coupling elastoplastic calculation. The automatic adjustment strategy of time step and related control parameters are established through the evaluation system of displacement error, pore pressure error and mixed error. Through the influence analysis of the control parameters, the main control parameters are error allowable value and pore pressure error proportional coefficient, and the auxiliary control parameters are the initial time step, the lower limit and the upper limit of step size adjustment. The dynamic adaptive step numerical calculation of subway station model in liquefaction site is carried out, and the development process of floating displacement and super-pore pressure ratio is obtained, and the disaster of liquefaction floating is predicted. At the same time, the accuracy and efficiency of the fixed step size method and the adaptive step size method are compared. It is found that the adaptive step size method can greatly save the calculation cost and not lose the calculation accuracy.
【作者单位】: 济南大学土木建筑学院;大连理工大学海岸与近海工程国家重点实验室;
【基金】:国家自然科学基金项目(41402261)
【分类号】:TU435

【参考文献】

相关期刊论文 前10条

1 张西文;唐小微;n&\×冀,

本文编号:2171063


资料下载
论文发表

本文链接:https://www.wllwen.com/kejilunwen/diqiudizhi/2171063.html


Copyright(c)文论论文网All Rights Reserved | 网站地图 |

版权申明:资料由用户203b7***提供,本站仅收录摘要或目录,作者需要删除请E-mail邮箱bigeng88@qq.com