自动化学报2012，Vol.38Issue(1)：120-127,8.DOI:10.3724/SP.J.1004.2012.00120
Kriging代理模型在对地观测卫星系统优化中的应用
Application of Kriging Surrogate Model to Optimization of Earth Observation Satellite System
摘要
Abstract
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.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Key words
Optimization of earth observation satellite system/Kriging surrogate model/optimized value of surrogate model/maximized expected improvement/improved generalized pattern search引用本文复制引用
刘晓路,陈盈果,贺仁杰,陈英武..Kriging代理模型在对地观测卫星系统优化中的应用[J].自动化学报,2012,38(1):120-127,8.基金项目
国家自然科学基金(70171156,70971131,70801062)资助 （70171156,70971131,70801062）
