基于BP神经网络的星外设备温度预测
详细信息    查看全文 | 推荐本文 |
  • 英文篇名:Temperature Prediction of Equipment Out of Satellite Base on BP Neural Network
  • 作者:宁东坡 ; 徐志明
  • 英文作者:NING Dong-Po;XU Zhi-Ming;DFH Satellite Co.Ltd;
  • 关键词:BP神经网络 ; 温度预测 ; 敏感性分析
  • 英文关键词:BP neural network;;temperature prediction;;sensitivity analysis
  • 中文刊名:GCRB
  • 英文刊名:Journal of Engineering Thermophysics
  • 机构:航天东方红卫星有限公司;
  • 出版日期:2019-07-15
  • 出版单位:工程热物理学报
  • 年:2019
  • 期:v.40
  • 语种:中文;
  • 页:GCRB201907023
  • 页数:6
  • CN:07
  • ISSN:11-2091/O4
  • 分类号:151-156
摘要
卫星舱外设备由于热容较小,所处的空间热环境复杂,热控设计难度较大。本文基于神经网络算法,建立了以星外设备热控设计参数为输入、设备温度为输出的BP神经网络模型。经过训练后的神经网络模型对样本数据的温度预测误差在1%左右,对新设计的热控参数预测误差在2%左右,表明所建立的BP神经预测温度模型精度高、稳定性强。通过所建立的神经网络模型对设备的热控参数进行了敏感性分析,分析结果表明支架的长度和支架安装面的传热系数对设备一轨内的最高温度影响更大。
        The thermal design of small equipment out of satellite is a problem because of small heat capacity and complex space thermal environment.A BP neural network that thermal design parameters as input and temperature as output is build based on neural network algorithm.The prediction error of BP neural network after training is about 1%,and the prediction error for new thermal design input is about 2%.The result shows the BP artificial neural network is of higher precision and stability.Sensitivity analysis for thermal design parameters is carried through the neural network,which shows that the length of holder and the heat transfer coefficient of fitting surface of holder are more important for equipment's max temperature in the orbit.
引文
[1]林来兴.分布式小卫星系统的技术发展与应用前景[J].航天器工程,2010,19(1):60-66LIN Laixing.Technological Development and Application Prospects of Distributed Small Satell ite System[J].Spacecraft Engineering,2010,19(1):60-66
    [2]杨敏,孙洋,鲍凯.小卫星及卫星组网的战术应用研究[J].舰船电子工程,2014,34(9):4-7YANG Min,SUN Yang,BAO Kai.Tactical Application of Small Satellite and Constellation Newworking[J].ShipElectronic Engineering,2014,34(9):4-7
    [3]Rose T S,Janson S W,Lalumondiere S,et al.LEO to Ground Optical Communications From a Small Satellite Platform[C]//SPIE LASE.International Society for Optics and Photonics,2015:935401-935401
    [4]Wang C,Yang W,Feng L,et al.Thermal Design and Simulation of an Attitude-varied Space Camera[C]//Applied Optics and Photonics China.2015:96780U
    [5]Meng H,Sun L,Zhang C,et al.Thermal Design and Flight Validation for High Precision Camera[C]//Applied Optics and Photonics China.International Society for Optics and Photonics,2015:15490-155011
    [6]徐向华,程雪涛,梁新刚.圆形太阳同步轨道卫星的空间热环境分析[J].宇航学报,2012,33(3):399-404XU Xianghua,CHENG Xuetao,LIANG Xingang.Analysis on Space Thermal Environment for Circular SunSynchronous Orbit[J].Journal of Astronautics,2012,33(3):399-404
    [7]马伟,宣益民,韩玉阁.临近空间热环境分析及低速飞行器的热设计方法[J].宇航学报,2010,31(5):1272-1277MA Wei,XUAN Yimin,HAN Yuge.Principle of Stratospheric Vehicles[J].Journal of Astronautics,2010,31(5):1272-1277
    [8]赵振明,鲁盼,宋欣阳,“高分二号”卫星相机热控系统的设计与验证[J].航天返回与遥感,2015,36(4):34-40ZHAO Zhenming,LU Pan,SONG Xinyang.Thermal Design and Test for High Resolution Space Camera on GF-2Satellite[J].Spacecraft Recovery&Remote Sensing,2015,36(4):34-40
    [9]刘百麟,董艺,魏巍.卫星平台机热一体化设计探讨[J].航天器工程,2016,25(2):25-31LIU Bailin,DONG Yi,WEI Wei.Discussion on Mechanical and Thermal Integrated Design of Satellite Platform[J].Spacecraft Engineering,2016,25(2):25-31
    [10]孙大强,谢龙,翟载腾,等.百叶窗热性能优化设计研究[J].上海航天,2016,33(1):95-100SUN Daqiang,XIE Long,ZHAI Zaiteng etl.Study on Optimization Design of Thermal Louver Performance[J].Aerospace Shanghai,2016,33(1):95-100
    [11]张桂兰,周佐新.中星22号通信卫星热设计[J].航天器工程,2002(1):24-27ZHANG Guilan,ZHOU Zuoxin.Thermal Design of China22 Communication Satellite[J].Spacecraft Engineering,2002(1):24-27
    [12]李萍,曾令可,税安泽,等.基于MATLAB的BP神经网络预测系统的设计[J].计算机应用与软件,2008,25(4):149-150LI Ping,ZENG Lingke,SHUI Anze etl.Design of Forecast System of Back Propagation Neural Network Based on MATLAB[J].Computer Applications and Software,2008,25(4):149-150
    [13]李宇峰,裴旭东,黄聪明.BP神经网络实际应用中的若干问题[J].兵工自动化,1998(1):1-4LI Yufeng,PEI Xudong,HUANG Congming.Some Question of BP Neural Network In the Application[J].Ordnance Industry Automation,1998(1):1-4

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700