利用多目标参数优化推断ICF热斑状态的径向分布  

Determination of radial profile of ICF hot spot's state by multi-objective parameters optimization

作　　者：董建军[1] 邓博[1] 曹柱荣[1] 丁永坤[1] 江少恩[1] 

机构地区：[1]中国工程物理研究院,绵阳621900

出　　处：《核聚变与等离子体物理》2014年第4期299-306,共8页Nuclear Fusion and Plasma Physics

基　　金：国家惯性约束聚变课题资助项目(GFZX02030102.1)

摘　　要：提出了一种以多目标参数优化确定热斑的温度和密度径向分布的方法。用热斑中心位置和燃料界面位置对应的电子温度、靶剩余烧蚀层峰值密度、剩余烧蚀层质量与初始烧蚀层质量的比值以及燃料与烧蚀层分界面位置构成五维变量空间参数空间的每个点描述一组温度和密度分布。将实验测量的X射线图像的归一化强度分布与理论模型计算的强度分布的方差作为两个目标函数;另一个目标函数是实验测量的热斑的平均温度与理论模型计算的平均温度的方差。通过多目标遗传算法搜索五维参数空间获得最优参数,从而获得最优的温度和密度分布。对KB显微镜配合X射线胶片测量的实验结果进行了计算,给出了热斑的最佳温度和密度的径向分布。由测试算例发现,温度分布对目标函数的依赖较强。A method using multi-objective parameters optimization is presented to determine the radial profile of hot spot temperature and density. And a parameter space which contain five variables： the temperatures at center and the interface of fuel and remain ablator, the maximum model density of remain ablator, the mass ratio of remain ablator to initial ablator and the position of interface between fuel and the remain ablator, is used to described the hot spot radial temperature and density. Two objective functions are set as the variances of normalized intensity profile from experiment X-ray images and the theory calculation. Another objective function is set as the variance of experiment average temperature of hot spot and the average temperature calculated by theoretical model. The optimized parameters are obtained by multi-objective genetic algorithm searching for the five dimention parameter space, thereby the optimized radial temperature and density protiles can be determined. The radial temperature and density profils of hot spot by experiment data measured by KB microscope cooperating with X-ray film are presented. It is observed that the temperature profile is strongly correlated to the objective functions.

关 键 词：惯性约束聚变 内爆压缩 热斑 多目标优化 遗传算法 KB显微镜 

分 类 号：O434.13[机械工程—光学工程]