Satellite Surface Charging: AI-Based Detection via CNN and 19-Year Dataset

sorae Japan

Overview

Researchers from Kyushu Institute of Technology and JAXA developed a world-first AI method using Convolutional Neural Networks (CNN) to automatically detect surface charging on satellites. By analyzing 19 years of data from the DMSP F16 satellite, the team quantified how the balance between auroral electrons and ionospheric plasma determines charging levels. This advancement enables better risk assessment for satellite design and operational longevity.

In Depth

Key Findings

A research group from Kyushu Institute of Technology and JAXA has developed a method to automatically detect satellite surface charging using artificial intelligence. By applying this system to observation data from the U.S. military weather satellite DMSP F16 spanning 2005 to 2023, the team identified approximately 3,000 charging events. The analysis revealed that while high-energy electrons causing auroras increase satellite charging, the inflow of positive ions within the surrounding ionospheric plasma acts to suppress negative charging.

Technical Details

The detection system utilizes a Convolutional Neural Network (CNN), a type of AI typically used for image recognition, to automate the identification of charging phenomena. This replaces previous manual methods where human operators had to visually verify charging candidates extracted from observation data, a process that was too labor-intensive for long-term statistical analysis. The study quantitatively demonstrated that the intensity of satellite charging is determined by the balance between auroral electrons and the surrounding plasma.

Background and Context

Unlike ground-based electrical equipment, satellites cannot be grounded. Consequently, electrons in space can accumulate on the satellite surface. If significant potential differences develop between different areas of the surface, electrostatic discharges (sparks) can occur. These discharges pose a risk to satellite operations by potentially damaging surface materials and solar cells.

Significance and Outlook

The research, published in the journal “Space Weather” on October 27, 2026, provides a foundation for predicting the specific times and locations where satellites are most prone to charging. The research group intends to expand their analysis to previous periods of higher auroral activity to further refine these predictions. As the number of satellites in low Earth orbit increases for communication and Earth observation, this AI-driven detection method will support the design of charging-resistant satellites and the implementation of operational strategies to avoid hardware failure.

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.

• Your email and selected fields are used only to deliver the newsletter.
• We never share your information with third parties.
• You can unsubscribe anytime via the link in each email.

See our Privacy Policy for details.

Agree & Continue ▶

Takes about a minute · Unsubscribe anytime