摘要
结构优化是对地观测卫星系统(Earth observation satellite system,EOSS)性能提高的关键,但其覆盖性能难以解析计算.为实现EOSS优化,提出了仿真优化的求解思路:构建Kriging代理模型对仿真数据进行拟合,采用代理模型最优和最大化期望提高相结合的机制选择更新点,并定义单位距离的函数改进对更新点进行过滤;提出了改进广义模式搜索算法求解代理模型,搜索步采用遗传算法和序列二次规划算法实现,筛选步采用不完全动态筛选.最后,通过仿真实例和对比实验验证了本文方法的有效性.
Optimized system configuration is the key point to improve performance of earth observation satellite system （EOSS）. However, its performance can not be calculated analytically. To solve the problem of EOSS optimization, we propose a simulation based optimization method, in which Kriging surrogate model is built to approximate simulation data. Points with optimized values or maximal expected improvement are selected to update our surrogate model. And a measure named objective improvement versus distance is defined to filtrate the selected points. To get the optimized solution of the surrogate model, we construct an improved generalized pattern search algorithm. In the search step, genetic algorithm and sequential quadratic programming are used to find potential update points. In the poll step, dynamic incompletion poll is carried out to find points with greater value. Finally, through a series of test cases and contrastive experiments, the results prove that our method is effective.
出处
《自动化学报》
EI
CSCD
北大核心
2012年第1期120-127,共8页
Acta Automatica Sinica
基金
国家自然科学基金(70171156
70971131
70801062)资助~~
关键词
对地观测卫星系统优化
Kriging代理模型
代理模型最优
最大化期望提高
改进广义模式搜索
Optimization of earth observation satellite system, Kriging surrogate model, optimized value of surrogate model, maximized expected improvement, improved generalized pattern search
