Fujitsu and Nagoya University Automate Solar Energetic Particle Simulation: Multi-Spacecraft Bayesian Optimization Breakthrough

On October 21, 2026, a research consortium comprising Fujitsu, the Tokai National Higher Education and Research System (Nagoya University), JAMSTEC, and NICT announced a pioneering advance in computational space weather physics, published in Scientific Reports. The team has engineered the world's first automated technology capable of calibrating the complex governing physical parameters of Solar Energetic Particle (SEP) acceleration and interplanetary transport using simultaneous observation streams across five spacecraft.

High-energy solar particles generated during extreme coronal mass ejections represent a hazardous radiation risk to orbital electronics and human exploration missions, precipitating catastrophic single-event upsets in satellite microprocessors.

Solving a 19-Dimensional Parameter Space via Hybrid Black-Box Optimization

The propagation of SEPs through heliospheric magnetic flux tubes is governed by pitch-angle scattering and diffusion along the turbulent Parker spiral. Historically, deriving core governing parameters—such as the mean free path, radial diffusion coefficients, and shock acceleration injection efficiencies—required exhaustive, subjective trial-and-error by astrophysicists, resulting in severe variance and model discrepancies.

To establish an objective and reproducible calibration standard, the researchers developed a two-stage hybrid optimization pipeline. Low-discrepancy quasi-random Sobol' sequences initially explore the vast 19-dimensional parameter space, followed by targeted Bayesian optimization utilizing the open-source Optuna framework for rapid local convergence. Validated against the major solar event of October 9, 2021, the system successfully reproduced particle flux profiles across five spacecraft positioned at widely varying solar longitudes: Parker Solar Probe, BepiColombo, Solar Orbiter, STEREO-A, and Earth-orbiting satellites.

Operational Safeguards for Megaconstellations and Deep-Space Exploration

In contemporary aerospace engineering, the widespread deployment of commercial off-the-shelf (COTS) components in low-Earth orbit megaconstellations heightens vulnerability to radiation-induced memory corruption and destructive latch-ups. Deterministic SEP simulations provide satellite operators and flight directors with actionable early warning windows spanning several hours, allowing proactive spacecraft safe-moding before ionizing wavefronts arrive.

“Determining solar particle transport parameters has long been an artisanal endeavor hindered by human subjectivity. By framing heliospheric transport calibration as a reproducible mathematical optimization problem, we have transformed space weather modeling into a quantitative engineering discipline.”

This breakthrough establishes an empirical precedent for coupling multi-spacecraft telemetry with advanced algorithmic parameter estimation. As human spaceflight sets its sights on lunar base operations, automated space weather modeling transitions from an academic curiosity into an indispensable life-support asset.