基于改进遗传算法的胶结砂砾石料细观层面参数反演分析

Inversion analysis of meso-level parameters of cemented sand-gravel material based on improved genetic algorithm

针对传统室内试验无法获取胶结砂砾石料细观层面参数的问题,提出基于改进遗传算法的胶结砂砾石料细观层面参数反演分析方法。以胶结砂砾石料室内单轴压缩试验结果为反演目标值,通过有限元软件模拟建立室内试验的胶结砂砾石试件随机骨料模型,反演分析确定胶结砂砾石细观组分参数,解决了室内试验无法获取参数的问题。研究结果表明:在该试验条件下,通过反演分析确定的细观组分参数进行室内试验数值模拟结果与试验值基本一致,胶结砂砾石料的骨料弹性模量EG最终值为550 MPa,砂浆基质弹性模量Em最终值为40MPa,界面弹性模量Ei最终值为30MPa。

Aiming at the problem that the parameters of the meso layer of cemented sand gravel can not be obtained by traditional laboratory tests,an inverse analysis method of the parameters of the meso layer of cemented sand gravel based on improved genetic algorithm is proposed.Taking the results of laboratory uniaxial compression test of cemented sand and gravel as the inversion target value,the random aggregate model of cemented sand and gravel specimen for laboratory test is established through finite element software simulation,and the microscopic component parameters of cemented sand and gravel are determined through inversion analysis,which solves the problem that the parameters cannot be obtained in laboratory test.The research results show that:under this test condition,the numerical simulation results of laboratory tests on the meso component parameters determined by back analysis are basically consistent with the test values,the final value of the elastic modulus of the aggregate of the cemented sand gravel is Ec=550 MPa,the final value of the elastic modulus of the mortar matrix is Em=40 MPa,and the final value of the elastic modulus of the interface is E;=30 MPa.