Scholarlink

Abstract <p>Satellite communication and navigation systems are increasingly essential in modern society, making it crucial to understand the impact of solar activity on these technologies. Total electron content (TEC) significantly influences satellite performance, necessitating accurate forecasting to maintain operational reliability. This research focuses on predicting TEC during eleven distinct X-class solar flares that occurred in February, March, May, June, and August 2024, utilizing a long short-term memory (LSTM) model. The study employs a comprehensive dataset of TEC data sourced from the IONOLAB database, alongside important solar and geomagnetic parameters such as Kp, Ap, SSN, and F10.7 obtained from NASA OmniWeb. The model’s predictive performance was validated against the IRI-2017 model. Results demonstrate that the LSTM model effectively captures TEC variations during periods of extreme solar activity, consistently outperforming the IRI-2017 model. For instance, during significant solar events, the LSTM model achieved notable performance metrics, indicating its capability to provide precise TEC forecasts. This research contributes to the advancement of space weather forecasting models, enhancing the reliability of satellite-dependent systems critical for global communication and navigation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Ionospheric TEC Forecast Using LSTM during High-Intensity Solar Flares Occurred during the Year 2024 and Validation with IRI-2017

  • 期刊论文

  • 发表日期：

    2025年4月21日

查看全文
  • B. Raghavi,
  • R. Mukesh,
  • S. Muthamil,
  • S. Nivetha,
  • T. Muthu,
  • Sarat C. Dass,
  • S. Kiruthiga

摘要

Abstract

Satellite communication and navigation systems are increasingly essential in modern society, making it crucial to understand the impact of solar activity on these technologies. Total electron content (TEC) significantly influences satellite performance, necessitating accurate forecasting to maintain operational reliability. This research focuses on predicting TEC during eleven distinct X-class solar flares that occurred in February, March, May, June, and August 2024, utilizing a long short-term memory (LSTM) model. The study employs a comprehensive dataset of TEC data sourced from the IONOLAB database, alongside important solar and geomagnetic parameters such as Kp, Ap, SSN, and F10.7 obtained from NASA OmniWeb. The model’s predictive performance was validated against the IRI-2017 model. Results demonstrate that the LSTM model effectively captures TEC variations during periods of extreme solar activity, consistently outperforming the IRI-2017 model. For instance, during significant solar events, the LSTM model achieved notable performance metrics, indicating its capability to provide precise TEC forecasts. This research contributes to the advancement of space weather forecasting models, enhancing the reliability of satellite-dependent systems critical for global communication and navigation.