金属靶板侵彻数值模拟对比研究

数值模拟是研究靶板侵彻问题的重要手段。为对相关软件的选取提供参考，采用当前广泛使用的3种有限元软件（LS-DYNA、ABAQUS、PAM-CRASH）对同一个靶板侵彻实验进行了数值模拟，并比较了各模拟软件的优缺点。研究结果表明:3种软件的模拟结果与实验结果均基本一致；对平头形弹体侵彻的模拟效果普遍优于对半球形弹体的模拟效果；在弹体剩余速度、冲塞块速度2个方面，3种有限元软件的数值模拟结果与实验结果大致吻合。其中，ABAQUS和PAM-CRASH对上述特征量的模拟结果更接近实验结果，平均相对误差普遍小于15%，且PAM-CRASH多数模拟结果误差都小于10%；而在弹体变形量上，3种软件的模拟结果均与实验结果相差较大。此外，LS-DYNA在模拟时的鲁棒性较好，模拟结果较稳定，PAM-CRASH对参数设置最敏感，ABAQUS在计算时间和效果上较平衡。

Comparative study of numerical simulations of projectile penetration into metal targets

Numerical simulation is an important method to study the penetration into targets. Distinct results may be acquired if different kinds of software are adopted. In this paper, three kinds of commonly-used softwares (LS-DYNA, ABAQUS and PAM-CRASH) were adopted to simulate the same series of penetration experiments. The simulation results were analyzed and the advantages and disadvantages of each kind of software were compared. The purpose of this work is to help the researchers and engineers to select the most suitable software. The results in this paper demonstrate that the calculation results of the three kinds of software are basically consistent with the experimental results. The simulation results of the flat-headed projectile are generally better than those of the hemispherical projectile. The residual velocity of the projectile and the plug velocity were simulated well using all of the three kinds of software. The calculated velocities by ABAQUS and PAM-CRASH are closer to the experimental results, and the average relative errors are generally lower than 15%. Moreover, the errors from most calculation results by PAM-CRASH are even less than 10%. However, for the deformation of the projectile, the calculation results of the three kinds of software are quite different from the experimental ones. There are other differences in the performance of the three kinds of software. For example, the ballistic limits calculated by ABAQUS and LS-DYNA are higher than the experimental results, while those by PAM-CRASH are lower. ABAQUS is most likely to report errors but is balanced by calculation time and effects. LS-DYNA has a low error rate and good robustness. The change of model parameters (such as mesh density, friction coefficient, contact stiffness, viscous damping coefficient, etc.) has little effect on its calculation results. PAM-CRASH is greatly affected by model parameters. The conclusion of this paper is based on the following conditions: the target material is Weldox 460E steel, the projectile is ARNE tool steel, and the impact velocity range is 180－450 m/s. However, it is also of guiding significance for the penetration problem simulation of other materials in this velocity range.