梅杰, 张博文, 张春云, 彭海峰, 崔苗. 幂强化弹塑性材料平面应变反问题[J]. 工程力学, 2020, 37(1): 248-256. DOI: 10.6052/j.issn.1000-4750.2019.01.0086
引用本文: 梅杰, 张博文, 张春云, 彭海峰, 崔苗. 幂强化弹塑性材料平面应变反问题[J]. 工程力学, 2020, 37(1): 248-256. DOI: 10.6052/j.issn.1000-4750.2019.01.0086
MEI Jie, ZHANG Bo-wen, ZHANG Chun-yun, PENG Hai-feng, CUI Miao. THE INVERSE PROBLEM OF PLANE STRAIN IN POWER-HARDENING ELASTICOPLASTICITY MATERIALS[J]. Engineering Mechanics, 2020, 37(1): 248-256. DOI: 10.6052/j.issn.1000-4750.2019.01.0086
Citation: MEI Jie, ZHANG Bo-wen, ZHANG Chun-yun, PENG Hai-feng, CUI Miao. THE INVERSE PROBLEM OF PLANE STRAIN IN POWER-HARDENING ELASTICOPLASTICITY MATERIALS[J]. Engineering Mechanics, 2020, 37(1): 248-256. DOI: 10.6052/j.issn.1000-4750.2019.01.0086

幂强化弹塑性材料平面应变反问题

THE INVERSE PROBLEM OF PLANE STRAIN IN POWER-HARDENING ELASTICOPLASTICITY MATERIALS

  • 摘要: 幂强化弹塑性材料在工程领域诸如金属管材制备、岩土工程分析中都具有广泛的应用。幂强化弹塑性材料的本构参数(例如弹性模量)和结构的边界条件(例如位移)往往不容易确定。在这种情况下,反问题为确定这些参数提供了一种新思路。将ABAQUS二次开发的子程序和复变量求导法结合,用于求解基于幂强化弹塑性材料的平面应变力学反问题:以传统的用户单元子程序为框架,将程序中实数变量转换为复数,建立了复数用户单元;采用复变量求导法确定测点位移对反演参数的灵敏度矩阵;结合最小二乘法和高斯消去法对反问题进行迭代求解。给出应用算例讨论了复变量求导法对正问题计算精度影响、算法在反问题求解过程中的精度,以及反演初值、测量误差对反演结果的影响。

     

    Abstract: Power-hardening elastoplastic materials have a wide range of engineering applications, such as metal pipe manufacturing and geotechnical analysis. The constitutive parameters of power-hardening elastoplastic materials (such as Young's modulus) and the boundary conditions of a structure (such as displacements) are often difficult to be determined. Under this circumstance, the inverse problem provides a new approach for determining these parameters. In the present work, ABAQUS UEL (user element subroutines) and the CVDM (complex variable-differentiation method) are combined to solve the inverse problem of plane strain mechanics based on power-hardening elastoplastic materials. Firstly, the traditional user element subroutine is used as the framework to convert a real variable in the subroutine into a complex variable, and the complex user element is established. Then the complex variable-differentiation method is used to determine the sensitivity matrix of the displacements at measurement point with respect to the inverse parameters. Finally, the inverse problem is solved iteratively by the Least-squares method and Gaussian elimination method. Numerical examples are given to discuss the influence of the CVDM on the accuracy of the direct problem calculation, and the accuracy of the present algorithm. The influence of initial value and measurement errors on inversion results are also investigated.

     

/

返回文章
返回