首页 | 本学科首页   官方微博 | 高级检索  
     检索      

利用人工神经网络预测电离层F2层骚扰
引用本文:柳文,焦培南.利用人工神经网络预测电离层F2层骚扰[J].地球物理学报,2001,44(1):24-30.
作者姓名:柳文  焦培南
作者单位:柳文(中国电波传播研究所,河南新乡 453003);焦培南(中国电波传播研究所,河南新乡 453003)
基金项目:电子科学院军事电子预研基金项目(DJ7.3.3.1).
摘    要:利用人工神经网络技术,提出预报离散随机的电离层骚扰事件的新方案.本文重点讨论了预报电离层骚扰的人工神经网络的构造,采用模糊理论和模式识别的思想构造了网络的输入层和输出层.将与电离层骚扰相关的日面现象如太阳耀斑、黑子等的日面位置、强度等参量作为网络的输入,该方案预报结果检验中,使传统方法难以预报的小型和中型电离层(骚扰达到80%以上)的预报准确率有所提高.最后还提出了利用人工神经网络识别单一型别骚扰事件的方案,预报准确率在95%以上。

关 键 词:人工神经网络,电离层骚扰,模糊集,隶属度.
文章编号:0001-5733(2001)01-0024-07
修稿时间:2000年5月30日

PREDICTION OF DISTURBANCES IN THE IONOSPHERE BY USING THE ARTIFICIAL NEURAL NETWORK
LIU WEN.PREDICTION OF DISTURBANCES IN THE IONOSPHERE BY USING THE ARTIFICIAL NEURAL NETWORK[J].Chinese Journal of Geophysics,2001,44(1):24-30.
Authors:LIU WEN
Abstract:A new method for predicting disturbances in the ionosphere by using the Artificial Neural Network (ANN) has been presented. We have inherited conventional prediction ideal and, at first, analyzed the solar terrestrial phenomena which are thought to be related with the ionospheric disturbances to define the out put and in put parameters of network. The prediction error is less than 20%. When putting the ANN into actual application, we should have good physical knowledge about the application, so that we can take better advantage of the Neural Network. In this paper we put our emphasis on how to construct the network, and use the ANN to recognize the single type ionosphere disturbance, the accuracy rate is 95%. All those work we have done can prove to some extent, that the method of the ANN for predicting disturbances in the ionosphere is effective and reasonable.
Keywords:Artificial Neural Network  Ionosphere disturbance  Fuzzy  
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号