1300MPa级热成形钢高应变速率本构模型分析

Analysis of high strain rate constitutive model of 1300 MPa hot-stamped steel

对CR950/1300HS的静态和高应变速率下的力学性能进行测试,采用5种不同的硬化模型对静态力学性能曲线进行拟合;选取误差较小的Swift和Hockett-Sherby模型,通过引入加权系数组成混合硬化模型;选取半径为5 mm的缺口拉伸试件进行混合硬化模型验证。在混合硬化模型中引入应变速率强化项,构建应变速率相关的混合本构模型。基于车身B柱侧面碰撞工况,对比应变速率相关的混合本构模型和其他本构模型仿真分析结果与实测结果的差异,以验证模型的可靠性。结果表明,混合硬化模型的仿真分析结果与试验测试的最大力误差为2.15%,可决系数达到0.976,优于其他硬化模型;CR950/1300HS的塑性、屈服强度和抗拉强度随着应变速率的增加得到一定的提升。应变速率相关的混合本构模型所得关键点应力与实测结果误差小于5%;B柱侧面碰撞仿真时,采用混合本构模型得到的最大侵入量和侵入速度的误差分别在4.5%和3.68%以内,优于采用其他模型的结果;因此,可选用该模型作为材料输入应用于碰撞仿真分析。

The static and high strain rate mechanical properties of CR950/1300HS were tested,and five different hardening models were selected to fit the static mechanical property curves.The Swift and Hockett-Sherby models with small errors were selected,and weighting coefficients were introduced to form the mixed hardening model.The notched tensile specimen with radius of 5 mm was selected to verify the mixed hardening model.The strain rate hardening term was introduced into the mixed hardening model to construct the strain rate dependent mixed constitutive model.Based on the side impact condition of the car body B-pillar,the differences between the simulation results strain rate related hybrid constitutive model and other constitutive models and the measured results of were compared to verify the reliability of the model.The results show that the maximum force deviation between the simulation analysis results of the mixed hardening model and the test results is 2.15%,and the determination coefficient reaches 0.976,which is better than other models.With the increase of strain rate,the plasticity,yield strength and tensile strength of CR950/1300HS increase to a certain extent.The errors between the stress of key points obtained by the strain rate-dependent mixed constitutive model and the measured results is less than 5%.In the simulation of B-pillar side impact,the errors of the maximum intrusion amount and intrusion velocity obtained by the mixed constitutive model are within 4.5%and 3.68%,respectively,which are better than that obtained by other models.Therefore,this model can be selected as the material input for crash simulation analysis.