基于粒子群和有限元的材料本构模型参数的确定

Parameter Identification of Material Constitutive Model Based on Particle Swarm Algorithm and Finite Element Method

本文提出了一种基于粒子群算法和有限元方法的弹塑性本构方程参数的确定方法,并在此基础上开发了参数自动优化系统CMPI(Constitutive Material Parameter Identification)。通过使最小二乘形式的目标函数达到极小值,实现参数的动态寻优;模型参数作为粒子群优化模块的输入,采用应变(应力)控制方法,调用材料非线性有限元模块计算特定参数时的应力(应变)响应,并计算与实验数据的误差,从而实现优化过程。CMPI目前适用于A-F类模型(Chaboche、Ohno-Wang、Jiang-Sehitoglu)的参数优化,以此为基础,不难扩展到其他材料模型。

A new method of parameter identification for elasto-plastic constitutive model was presented in this paper by employing particle swarm algorithm and finite element method.Then automatic optimization system CMPI（Constitutive Material Parameter Identification）was developed on the basis of this method.By making the objective function which has the least-square forms to achieve the minimum,dynamic parameter optimization can be realized.In the CMPI system,the model parameters act as input data of the particle swarm optimization module.In the meantime,the nonlinear FEM module was applied to calculate the stress（or strain）response under the strain（stress）controlled and the errors compared with the experimental results.At present,CMPI only suits for some models which have the form of Armstrong-Frederick model.But it is easy to be extended to other models.