磁暴期间电离层会出现强烈的扰动，它会严重干扰短波通信，还会引起卫星信号发生突变，导致不可预知的方位误差。对于这种常见的极端空间天气现象，本项目试图从一种新的角度来研究其过程，采用最新的深度学习算法构建新型电离层模型并探索性的研究其物理过程。具体说来，由于深度学习非常依赖于训练数据本身的特征，因此首先要利用白谱法来提取TEC数据中与磁暴相关的扰动作为训练数据，以及对应时刻的太阳和地磁活动指数作为条件，输入到条件深度卷积生成对抗网络中来建立暴时全球电离层TEC扰动模型。并用空间信息处理方法对模型的输出结果进行分析，提取其潜在的时空信息，深入研究电离层扰动与地磁活动在时空上定量关系。该项研究将有助于深入了解磁暴期间小尺度电离层扰动的动力学过程，发掘出深度学习在电离层物理中潜在应用能力，以及探究新的空间天气预报方法。本项目获得的模型将会提升与电离层相关应用在经济社会和国家安全方面的重要作用。

During the geomagnetic storm, strong disturbances will appear in ionosphere, which will seriously interfere with short-wave communication, and will also cause sudden changes in satellite signals, resulting in unpredictable azimuth errors. For this common extreme space weather phenomenon, this project attempts to study its process from a new perspective, using the latest deep learning algorithm to build a new ionospheric model and explore its physical process exploratoryly. Specifically, since deep learning is very dependent on the characteristics of the training data itself, the spectral whitening method will be first used to process the TEC data to obtain its disturbances which caused by the geomagnetic storm and then will be taken as the training data, and the solar and geomagnetic activity index at the corresponding time is used as a condition. Then inputing them into the conditional deep convolution generative adversarial networks to establish a storm-time global ionospheric TEC disturbance model. The spatial information processing method will be used to analyze the output of the model, and the potential spatio-temporal information will be extracted. The quantitative relationship between the ionospheric disturbance and the geomagnetic activity in space will deeply studied based on previous result. This research will contribute to deeply understanding the dynamic process of small-scale ionospheric disturbances during geomagnetic storms, uncover potential applications of deep learning in space physics, and explore new space weather prediction methods. The model obtained in this project will enhance the important role of ionospheric related applications in economic, social and national security.