烟幕初始云团最大半径数值模拟

基于圆柱体发烟装置,采用Truegrid仿真软件建立带有"V"型刻槽的圆柱壳体网格模型,并用Autodyn软件建立发烟装置仿真模型,对烟幕初始云团爆炸分散过程爆轰压力变化规律进行数值模拟。借助于MATLAB提供的分段线性插值函数,将压力作为已知量引入爆炸分散理论模型,采用欧拉法对模型进行数值计算。开展了野外测试实验,采用摄像法和图像处理技术得到试验数据。结果表明,爆炸分散第一阶段发生时间为0～8.9×10-6s,烟幕初始云团最大半径增大为原来的3～4倍;在爆炸分散第二阶段,烟幕初始云团最大半径增大到约3 m。基于Truegrid与Autodyn混合仿真,并与爆炸分散理论模型相结合的方法,相比仅利用数值积分求解爆炸分散模型的传统理论方法,烟幕初始云团最大半径的计算误差降低了3%～8%,且烟幕初始云团最大半径比单一理论模型法更接近实验结果。采用欧拉法计算模型的收敛性优于四阶龙格-库塔法。

Numerical Simulation of Maximum Radius of Initial Cloud Cluster of Smoke Screen

Based on the cylindrical smoking device, the Truegrid simulation software was used to establish a cylindrical shell mesh model with a"V"groove. The simulation model of the smoking device was established by Autodyn software, and the variation law of the detonation pressure during the initial cloud explosion and dispersion process of the smoke screen was numerically simulated. With the help of the piecewise linear interpolation function provided by MATLAB, the pressure as a known quantity was introduced into the explosion dispersion theory model, and the model was numerically calculated by the Euler method. Field test experiments were carried out, and experimental data were obtained using camera method and image processing technology. The results show that the time occurring the first stage of explosion dispersion is from 0 to 8.9×10^-6 s, and the maximum radius of the initial smoke cloud increases by 3 to 4 times that of the original. In the second stage of the explosion dispersion, the maximum radius of the initial smoke cloud increases to about 3 m. Based on the mixed simulation of Truegrid and Autodyn and combined method with the theory of explosion dispersion theory, the calculation error of the maximum radius of the initial smoke cloud is reduced by 3%~8% compared with the traditional theoretical method of solving the explosion dispersion model by numerical integration alone. The maximum radius of the initial smoke cloud is closer to the experimental result than the single theoretical model method. The convergence of calculation model of the Euler method is better than that of the fourth-order Runge-Kutta method.