Jakarta, INTI - Indonesia’s National Research and Innovation Agency (BRIN) is leveraging artificial intelligence (AI) to improve the prediction of solar storms, aiming to provide earlier warnings and strengthen preparedness before space weather events impact Earth.
Forecasting Coronal Mass Ejections (CMEs), the massive bursts of solar plasma that can trigger solar storms, has long been a major challenge due to the highly dynamic and unpredictable nature of the solar wind. Conventional physics-based models alone often struggle to accurately estimate when these events will reach Earth.
Hybrid AI Model Enhances Solar Storm Forecasting
To address this challenge, Tiar Dani, a researcher at BRIN’s Space Research Center under the Aerospace Research Organization, and his team developed a hybrid approach that combines physics-based modeling with artificial intelligence.
The system integrates the Drag-Based Model (DBM), which simulates how solar wind resistance affects the movement of CME particles, with a Random Forest machine learning model trained on historical data collected from CME events observed over two solar activity cycles.
By learning from historical observations, the AI estimates the level of solar wind drag experienced by each CME during its journey toward Earth. This allows the system to combine established physical principles with data-driven learning, resulting in more accurate predictions of CME arrival times.
According to Tiar, the hybrid physics-AI DBM model can predict CME travel times with an average error of approximately 8.7 hours, a level of accuracy considered highly competitive in the field of space weather forecasting.
The new model also outperforms the conventional DBM, which assumes a constant level of solar wind resistance throughout a CME’s journey. By incorporating AI, the hybrid system adapts to changing space weather conditions, significantly improving prediction accuracy.
Supporting Indonesia’s Future Space Weather Early Warning System
The research is expected to support the development of an early warning system for space weather, enabling satellite operators, telecommunications providers, and managers of critical infrastructure to take preventive measures before solar storms disrupt technological systems.
Tiar said the project also strengthens the foundation for applying artificial intelligence to Indonesia’s space research initiatives. The innovation will become a key component of BRIN’s Research in AI for Space program, which aims to establish an independent, resilient, and continuously evolving space weather early warning framework capable of protecting future technological infrastructure.
Conclusion
By combining artificial intelligence with physics-based modeling, BRIN has taken an important step toward improving space weather forecasting in Indonesia. The hybrid system not only delivers more accurate predictions of solar storm arrivals but also lays the foundation for a resilient, AI-driven early warning framework that can help safeguard satellites, communications networks, and other critical infrastructure from future space weather events.
Read more: How AI Technology Assists Referees with Offside Decisions at the 2026 FIFA World Cup