基于对数最小二乘法的卫星健康评估指标赋权方法  

Weighting Method for Satellite Health Assessment Indicators Based on Logarithmic Least Squares Method

作　　者：郑倩云 虞业泺 张钰伟 刘虎 姜烨[1,2] 禚磊 ZHENG Qianyun;YU Yeluo;ZHANG Yuwei;LIU Hu;JIANG Ye;ZHUO Lei(Innovation Institute for Microsatellite,Chinese Academy of Sciences,Shanghai 201203,China;Shanghai Engineering Center for Microsatellite,Shanghai 201203,China)

机构地区：[1]中国科学院微小卫星创新研究院,上海201203 [2]上海微小卫星工程中心,上海201203

出　　处：《航天器工程》2025年第2期97-103,共7页Spacecraft Engineering

摘　　要：针对卫星运行状态评估过程中异常指标不突出,评估结果无差异的情况,提出了一种对数最小二乘法结合传统算法的权重分配算法,先利用传统算法获取权重矩阵,再通过对数最小二乘法对矩阵进行对数转换,获得转换后的权重,最后通过标准化权重方法得到最终的权重分配结果。该方法能够准确捕捉到差异性较小的关键数据集对卫星健康状态的影响,显著提高了对整个系统中最重要组件的识别能力,使得卫星状态评估更敏感、更精确。验证结果表明,该方法具备快速、清晰的异常定位能力,为迅速识别系统或器部件的异常提供有力支持。研究强调了对关键数据的准确评估,为提高评估系统的效能和准确性做出了贡献。这一创新方法在卫星状态评估中为更全面地理解系统性能和异常提供了新的途径。In response to the issue of anomalous indicators not being prominently reflected and evaluation results lacking differentiation during satellite operation status assessment,a weight distribution algorithm combining logarithmic least squares method with traditional algorithms is proposed.First,the traditional algorithm is used to obtain the weight matrix,and then logarithmic transformation is applied to the matrix using the logarithmic least squares method to obtain the converted weights.Finally,the standardized weight method is used to determine the final weight distribution result.This method accurately captures the impact of key datasets with minor differences on the satellite health status,significantly improving the ability to identify the most critical components of the entire system,making the satellite status evaluation more sensitive and precise.Verification results demonstrate that this method possesses fast and clear anomaly localization capabilities,providing robust support for the rapid identification of system or component anomalies.The accurate evaluation of key data is emphasized in this study,contributing to the enhanced effectiveness and accuracy of the evaluation system.This innovative method provides a new approach for more comprehensively understanding system performance and anomalies in satellite status assessment.

关 键 词：卫星 综合评估 权重分配 对数最小二乘法 

分 类 号：V528[航空宇航科学与技术—人机与环境工程]