一种基于聚类的卫星太阳电池阵衰减估计方法  

An Attenuation Estimation Method of Satellite Solar Array Based on Clustering

作　　者：曹孟达 张涛[1] 李训嘉 刘亚杰 CAO Mengda;ZHANG Tao;LI Xunjia;LIU Yajie(College of Systems Engineering,National University of Defense Technology,Changsha 410073,China)

机构地区：[1]国防科技大学系统工程学院,湖南长沙410073

出　　处：《无线电工程》2019年第5期397-403,共7页Radio Engineering

基　　金：装备技术基础基金资助项目(181GF35003)

摘　　要：针对资源卫星的太阳电池阵输出功率衰减问题,基于影响太阳电池阵输出功率的因素和卫星地面站所能接收到的相关遥测数据,提出了一种基于聚类的卫星太阳电池阵衰减评估方法,通过聚类方法寻找电池阵发电同状态点,利用周期识别消除电池阵电流周期性影响的解决方案,建立了周期补偿和功率衰减因子估计模型,并结合实际工程数据进行了衰减估计。通过把所提出的方法对某在轨资源卫星的太阳电池阵衰减进行评估,发现该卫星的太阳电池阵的年衰减率约1.425%,其结果与文献[11,15]等对太阳同步轨道卫星太阳电池阵衰减约为1.5%相一致。该方法有效克服了现有方法必须依赖长期数据才能进行衰减评估的不足,可应用于在轨卫星电源系统的健康状态管理等辅助决策,对于提供卫星的使用效益具有重要参考价值。Aiming at the problem of output power attenuation of solar arrays for resource satellites,a clustering-based attenuation assessment method for satellite solar arrays is proposed based on the factors that affect the output power of the solar arrays and the related telemetry data that can be received by satellite ground stations.The clustering method is used to find the same state of the solar array power generation.The cycle identification is used to eliminate the periodic influence of the solar array current.The cycle compensation and power attenuation factor estimation models are established.The attenuation estimation is combined with the actual engineering data.The attenuation of the solar array of an on-orbit resource satellite is assessed by using the proposed method which is found that the annual attenuation rate of the solar array of the satellite is about 1.425%.The results are consistent with the result of literatures [11,15].The attenuation of the array is approximately 1.5%.This method effectively overcomes the shortcomings that existing methods must rely on long-term data to perform attenuation assessment.It can be applied to assist decision-making such as health management of on-orbit satellite power systems,and is of important reference value for providing satellite use efficiency.

关 键 词：卫星太阳电池阵 短期数据 聚类 周期识别 衰减估计 

分 类 号：TM934[电气工程—电力电子与电力传动]