Application of a Multi?Layer Artificial Neural Network in a 3?D Global Electron Density Model Using the Long?Term Observations of COSMIC, Fengyun?3C, and Digisonde

The ionosphere plays an important role in satellite navigation, radio communication, and space weather prediction. However, it is still a challenging mission to develop model with high predictability that captures the horizontal-vertical features of ionospheric electrodynamics. In this study, multiple observations during 2005–2019 from space-borne global navigation system (GNSS) occultation (RO) systems (COSMIC FY-3C) Digisonde Global Ionosphere Radio Observatory are utilized completely three-dimensional electron density based on artificial neural network, namely ANN-TDD. correlation coefficients predicted profiles all exceed 0.96 for training, validation test datasets, minimum root-mean-square error residuals 7.8 × 104 el/cm3. Under quiet weather, accuracy ANN-TDD 30%–60% higher than IRI-2016 at Millstone Hill Jicamarca incoherent scatter radars. less capable predicting dynamic evolution under severe geomagnetic storms compared STORM option activated. Additionally, successfully reproduces large-scale electrodynamic features, including seasonal variation hemispheric asymmetries. These agree well structure revealed by RO derived FORMOSAT/COSMIC-2 mission. Furthermore, prominent regional patterns, equatorial ionization anomaly, Weddell Sea anomaly mid-latitude summer nighttime anomaly. new expected play application GNSS explanation physical mechanisms involved.