基于变邻域分布估计算法的数传资源配置优化

Optimal allocation of ground station data transmission resources based on variable neighborhood estimation of distribution algorithms

分析了地面站数传资源分配中的影响因素,建立了问题的约束优化模型,提出了变邻域搜索与分布估计相结合的优化算法.算法在概率模型和种群个体两个层次分别设计了邻域结构,以提高算法的全局搜索能力和局部优化能力,并避免优化过程陷入局部极值.构建了变邻域分布估计算法的优化流程,推导证明了算法的收敛性.利用仿真算例分别对算法性能及控制参数选择进行了分析,实验结果表明,算法能够有效地解决数传资源配置优化问题,且求解精度较高.

Three constrained optimization models were built based on the analysis of influence factors in the problem of optimal allocation of ground station data transmission resources,and a hybrid method which combines estimation of distribution algorithms with variable neighborhood searching（VNS-EDAs） was proposed.Neighborhood structures that can avoid premature convergence were designed in both probability distribution level and population individual level to improve the capability of global exploitation and local exploration.The process of VNS-EDAs was presented,and the convergence of VNS-EDAs was proved with finite population.Simulation examples were designed to analysis the validity and parameters selection of VNS-EDAs.Experiment results show the VNS-EDAs can solve the problem of optimal allocation of ground station data transmission resources effectively,and has the advantage of precious solution.