小卫星内部测点剪裁优化方法  

Measurement point optimization for sensor layout inside small satellites

作　　者：岳中文 孙浩 刘闯[2] 彭瑾坤 何玲[2] 邱汉平[2] 武文华[1,3] YUE Zhongwen;SUN Hao;LIU Chuang;PENG Jinkun;HE Ling;QIU Hanping;WU Wenhua(School of Aeronautics and Astronautics,Dalian University of Technology,Dalian 116024,China;Beijing Institute of Spacecraft Environment Engineering,Beijing 100029,China;Ningbo Research Institute of Dalian University of Technology,Ningbo 315000,China)

机构地区：[1]大连理工大学力学与航空航天学院,大连116024 [2]北京卫星环境工程研究所,北京100029 [3]大连理工大学宁波研究院,宁波315000

出　　处：《航天器环境工程》2024年第5期633-643,共11页Spacecraft Environment Engineering

基　　金：国家重点研发计划项目(编号:2021YFA1003501);辽宁省兴辽英才计划项目(编号:XLYC2002108);大连市支持高层次人才创新创业项目(编号:2021RD16)。

摘　　要：小卫星在动力学测试过程中的内外结构传感器布点直接影响测试效率和试验周期。在充分考虑小卫星结构内/外测点相关性的前提下,基于智能算法提出一种传感器布点优化方法,旨在剪裁掉内部测点的同时确保外部测点之间的独立性。首先,通过对卫星结构的扫频分析,构建内/外测点的频响函数矩阵;然后,采用主成分分析和K-means聚类分析方法,将外部测点划分成不同的类别库。以此为基础,以内/外测点频响函数相关系数最大化为优化目标,运用遗传算法在不同的类别库中选择合适的测点进行传感器布置优化;进一步采用修正信息熵的方法来实现卫星结构内部测点传感器数量的剪裁优化。某小卫星的测点剪裁实际应用验证了该方法能够实现大幅度的内部测点剪裁。The layouts of sensors located both inside and outside the small satellite structure during dynamic test directly affect the test efficiency and period.After fully considering the correlation between internal and external measurement points,an intelligent algorithm-based sensor layout optimization method was proposed.It was intended to reduce the number of internal measurement points while ensuring the independence of external ones.Initially,a frequency response function matrix of internal and external measurement points was constructed through frequency-sweeping analysis of the satellite structure.Subsequently,the external measurement points were categorized into different groups using principal component analysis and K-means clustering analysis.Based on this,the optimization objective was determined to maximize the correlation coefficient of frequency response functions between internal and external measurement points.Genetic algorithms were utilized to select appropriate measurement points from the obtained categories to optimize the sensor layout.Furthermore,a modified information entropy approach was applied to optimize the number of internal measurement sensors in the satellite structure.Through practical applications in a specific small satellite,the proposed method was verified to be able to achieve a good number reduction in internal sensors.

关 键 词：传感器布置优化 频响函数 遗传算法 聚类分析 主成分分析 修正信息熵 

分 类 号：V416.1[航空宇航科学与技术—航空宇航推进理论与工程]