aiqbench analysis
This article introduces PISCES, a physics-informed convolutional autoencoder designed to detect solar wind anomalies at the L1 Lagrange point without relying on labeled data. By integrating physical constraints such as magnetic field consistency and the Parker spiral angle directly into its loss function, the model provides interpretable anomaly scores that separate magnetic and plasma contributions. This approach advances space weather forecasting by enabling earlier warnings for sudden commencements while maintaining fidelity to underlying astrophysical laws.
PISCES: Physics-Informed Solar-wind Convolutional autoEncoder for Space-weather Anomaly Detection and Early Warning
arXiv:2609.28022v1 Announce Type: cross Abstract: Space weather early warning depends on detecting solar wind transients in in-situ measurements at the first Sun-Earth Lagrange point (L1), before they reach Earth. Fixed thresholds can miss combined magnetic and plasma structure, and many learning methods provide a single anomaly score. We present the Physics-Informed Solar-wind Convolutional autoEncoder for Space-weather (PISCES), a convolutional autoencoder trained without catalog labels on OMNI solar wind measurements under physics constraints. Its loss includes magnetic field consistency, an empirical relation between temperature and velocity, the Parker spiral angle, and penalties on changes between consecutive one-minute samples in derived quantities calculated from the reconstruction. At inference, PISCES separates the anomaly score into magnetic and plasma reconstruction errors, physics relations, and residual corrections, and reports the magnitude of each contribution. Attenuation of the skip connections, selected on validation data, improves average precision for the trained models, while the untrained scores remain nearly the same. The trained models also give a more consistent ordering of these physical contributions. After smoothing with a trailing median, the alarms can precede independently observed sudden commencements, including positive sudden impulses.
Abstract:Space weather early warning depends on detecting solar wind transients in in-situ measurements at the first Sun-Earth Lagrange point (L1), before they reach Earth. Fixed thresholds can miss combined magnetic and plasma structure, and many learning methods provide a single anomaly score. We present the Physics-Informed Solar-wind Convolutional autoEncoder for Space-weather (PISCES), a convolutional autoencoder trained without catalog labels on OMNI solar wind measurements under physics constraints. Its loss includes magnetic field consistency, an empirical relation between temperature and velocity, the Parker spiral angle, and penalties on changes between consecutive one-minute samples in derived quantities calculated from the reconstruction. At inference, PISCES separates the anomaly score into magnetic and plasma reconstruction errors, physics relations, and residual corrections, and reports the magnitude of each contribution. Attenuation of the skip connections, selected on validation data, improves average precision for the trained models, while the untrained scores remain nearly the same. The trained models also give a more consistent ordering of these physical contributions. After smoothing with a trailing median, the alarms can precede independently observed sudden commencements, including positive sudden impulses.
Poster presented at NASA 5th Eddy Cross-Disciplinary Symposium, May 2026. Available at: this https URL
Machine Learning (cs.LG); Solar and Stellar Astrophysics (astro-ph.SR); Artificial Intelligence (cs.AI); Space Physics (physics.space-ph)
arXiv:2609.28022 [cs.LG]
(or arXiv:2609.28022v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.28022 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 299 words
Related coverage
Forecasting space weather risks on power grids
Microsoft Research · Sep 30, 2026
Neural Operator based Multi-Field Reconstruction of Inner Solar Boundary State
arXiv · Aug 25, 2026
Discovering Dual-Origin Slow Wind from Solar Orbiter with Self-Supervised Contrastive Learning
arXiv · Aug 25, 2026
A Pre-trained Variational Autoencoder for Gyrokinetic Plasma Turbulence Surrogate Modeling
arXiv · Oct 1, 2026