Neural network prediction of the topside electron content over the Euro-African sector derived from Swarm-A measurements

Abstract

This study presents the first prediction results of a neural network model for the vertical total electron content of the topside ionosphere based on Swarm-A measurements. The model was trained on 5 years of Swarm-A data over the Euro-African sector spanning the period 1 January 2014 to 31 December 2018. The Swarm-A data was combined with solar and geomagnetic indices to train the NN model. The Swarm-A data of 1 January to 30 September 2019 was used to test the performance of the neural network. The data was divided into two main categories: most quiet and most disturbed days of each month. Each category was subdivided into two subcategories according to the Swarm-A trajectory i.e. whether it was ascending or descending in order to accommodate the change in local time when the satellite traverses the poles. Four pairs of neural network models were implemented, the first of each pair having one hidden layer, and the second of each pair having two hidden layers, for the following cases: 1) quiet day-ascending, 2) quiet day-descending, 3) disturbed day-ascending, and 4) disturbed day-descending. The topside vertical total electron content predicted by the neural network models compared well with the measurements by Swarm-A. The model that performed best was the one hidden layer model in the case of quiet days for descending trajectories, with RMSE = 1.20 TECU, R = 0.76. The worst performance occurred during the disturbed descending trajectories where the one hidden layer model had the worst RMSE = 2.12 TECU, (R = 0.54), and the two hidden layer model had the worst correlation coefficient R = 0.47 (RMSE = 1.57).In all cases, the neural network models performed better than the IRI2016 model in predicting the topside total electron content. The NN models presented here is the first such attempt at comparing NN models for the topside VTEC based on Swarm-A measurements.

Financial Default Risk Prediction Algorithm Based on Neural Network under the Background of Big Data

“…We assume that A is the model's input independent variable and that the model consists of n layers of neural networks [16].…”

Prediction of Urban Scale Expansion Based on Genetic Algorithm Optimized Neural Network Model

“…Abuelezz's team proposed a model for predicting the vertical total electron content of the upper ionosphere based on neural network and compared it with the IRI2016 model. The results show that the IRI2016 model is inferior to this model in all cases [15]. Chen and other researchers established a film scoring prediction model based on the convolutional neural network.…”

Artificial deep neural network modeling of solar- and atmospheric-driven ground magnetic perturbations at mid-latitude

“…time-series) modeling as they allow us to consider independent parameters associated with physical sources contributing to the magnetic measurements. Neural networks have increasingly been used for spaceweather related applications including prediction of magnetic activity indices (e.g., Wu & Lundstedt, 1996;Kumluca et al, 1999;Stepanova & Pérez, 2000;Wintoft & Cander, 2000;Lundstedt et al, 2002;Uwamahoro & Habarulema, 2014;Shin et al, 2016;Zhelavskaya et al, 2017;Tebabal et al, 2018;Efitorov et al, 2018;Gruet et al, 2018;Jackson et al, 2020;Zou et al, 2020;Chakraborty & Morley, 2020;Myagkova et al, 2021;Abuelezz et al, 2021;Siciliano et al, 2021;Collado-Villaverde et al, 2021;Madsen et al, 2022;Zhang et al, 2022;Huang et al, 2022;Bernoux et al, 2022;Collado-Villaverde et al, 2023;Vladimirov et al, 2023). We will demonstrate the capability of our newly developed neural networks during quiet periods as well as disturbed periods and discuss future applications.…”