基于长短期记忆网络的扩展F事件短临预测  

Short-term forecasting of spread F using LSTM network

作　　者：孙树计[1] 徐彤[1] 班盼盼[1,2] 胡冉冉 陈春 SUN Shuji;XU Tong;BAN Panpan;HU Ranran;CHEN Chun(China Research Institute of Radiowave Propagation,Qingdao 266107,China;Xidian University,Xi’an 717071,China)

机构地区：[1]中国电波传播研究所,青岛266107 [2]西安电子科技大学,西安717071

出　　处：《电波科学学报》2023年第4期686-690,720,共6页Chinese Journal of Radio Science

基　　金：泰山学者工程(B992393001-5)。

摘　　要：考虑到我国区域电离层扩展F发生的物理机制和相关因素以及观测数据的可获取性,利用适宜于时间序列预测的长短期记忆网络建立了我国不同区域电离层扩展F事件的预测模型,对未来3 h是否会发生扩展F事件进行预测.以处于我国典型纬度地区的满洲里、北京、海口站为例,利用2015和2016年测试数据集对模型预测精度进行了检验,满洲里、北京、海口站平均准确率分别为92.4%、95.3%、96.0%,平均精确率分别为75.0%、61.2%、62.6%,平均召回率分别为73.0%、50.6%、31.5%.由此可见:在某些情况下,比如在低纬度地区的海口站,模型的召回率较低;除此之外,所建立的扩展F模型在多数情况下具有较高的预测能力.Considering the main mechanism of spread F and the available data related to the occurrence of spread F in our country,a model based on long short-term memory(LSTM)network is built to forecast spread F in 3 hours.The models for Manzhouli,Beijing,and Haikou were built and tested using data in 2015 and 2016.The mean accuracy rates in Manzhouli,Beijing and Haiou are 92.4%,95.3%,and 96.0%.The mean precision rates are 75.0%,61.2%,and 62.6%,and the mean recall rates are 73.0%,50.6%,and 31.5%,respectively.It is shown that the model has a high prediction ability in most cases,but it still needs to improve the ability of the model in some cases,especially in low latitude like Haikou station with a lower recall rate.

关 键 词：深度学习 扩展F 电离层 长短期记忆(LSTM)网络 预测模型 

分 类 号：P352.5[天文地球—空间物理学]