Prediction of TEC with Artificial Intelligence Using Space Weather Data as Input

Abstract

Artificial Intelligence (AI) has significantly advanced the field of space weather analysis by enabling accurate prediction of ionospheric parameters from complex environmental data. The Prediction of Total Electron Content (TEC) with Artificial Intelligence Using Space Weather Data as Inputsystem is designed to predict TEC values by analyzing key space weather parameters such as Solar Flux (F10.7), Kp Index, Dst Index, Solar Wind Speed, Proton Density, and IMF Bz. The proposed system employs Machine Learning algorithms to learn patterns from historical space weather data and generate reliable TEC predictions. A user-friendly web application developed using Python, Streamlit, HTML, CSS, JavaScript, Scikitlearn, Pandas, NumPy, Matplotlib, and Joblib enables users to input space weather parameters and visualize predicted TEC values instantly. Compared to traditional statistical approaches, the proposed AI-based model improves prediction accuracy, reduces computational effort, and provides faster decision support for space weather monitoring. The system can assist researchers, communication engineers, and satellite navigation systems in understanding ionospheric behavior and mitigating the impact of space weather on modern communication and positioning technologies.