遗传算法反演地球等离子体层离子密度分布

本文介绍了采用一维遗传算法从地球等离子体层极紫外图像反演地球等离子体层He⁺密度的原理.首先采用通量管近似和磁偶极近似将三维问题转化为一维问题.通过引入权矩阵,将极紫外光强积分离散为求和函数,再采用一维实数编码遗传算法反演得到磁赤道面等离子体层He⁺密度,最后通过磁力线追迹得到三维密度分布.算法采用动态全球核心等离子体模式模拟的密度和光强分布作为初始输入参数,并通过遗传算法得到相应密度分布.反演结果表明,等离子体层密度相对误差在8%以内,光强相对误差趋于0,算法有效可行.本文研究为中国探月二期工程中月基极紫外图像反演奠定了基础.

Inversion of the Earth's plasmaspheric density distribution from EUV images with genetic algorithm

The principles for inversion of plasmaspheric He+ density from the extreme ultraviolet images of the Earth′s plasmasphere with one-dimensional genetic algorithm are introduced.The three-dimensional problem is transformed to one-dimensional case through flux tube approximation and dipole magnetic field approximation.With the weight matrix,the extreme ultraviolet intensity integration equation is discretized into linear equation systems,then the one-dimensional real-coded genetic algorithm is used to calculate the equatorial plane plasmaspheric He+ density,and finally the three-dimensional density is obtained by magnetic field line tracing.The density and intensity simulated by the dynamic global core plasma model are used as the primary input parameters,and the corresponding density distribution is determined through genetic iterations.The results show that the relative density error is less than 8% and the relative intensity error tends to be zero,which proves that our algorithm is effective and feasible.Investigations in this work will provide basis for the inversion of the moon-based EUV images in the Second Phase of Chinese Lunar Exploration Program.