What question did this study set out to answer?

The aim is to enhance prediction accuracy of Northeast China Cold Vortex (NCCV) trajectories using artificial intelligence techniques.

February 28, 2026Applied SciencesOpen Access

Prediction of Northeast China Cold Vortex Paths Based on Multi-Generator with Integrated Multimodal Features

View Full Paper

Ask AI

Bookmark

Authors

This research demonstrates improved forecasting accuracy of cold vortex paths in Northeast China, suggesting enhanced weather prediction methods.

The aim is to enhance prediction accuracy of Northeast China Cold Vortex (NCCV) trajectories using artificial intelligence techniques.
Constructed a 23-year multi-modal spatiotemporal dataset of NCCV using ERA5 reanalysis data.
Developed an improved generative adversarial network model with a multi-encoder architecture.
Introduced a multi-generator structure to enhance prediction capability.
Implemented a selector module for optimal path selection.
Conducted ablation experiments comparing single- to multi-modal data inputs.
Achieved a reduction in average prediction error by 67.96 km, a 34.0% improvement over previous models.
Improved 24-hour prediction error by 39.7%.
Demonstrated superior trajectory prediction accuracy for intervals of 6 h, 12 h, 18 h, and 24 h.
Reduced prediction distance errors by 21.4%, 29.2%, 34.0%, and 37.0% compared to LSTM models.

Jiao et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

February 28, 2026Applied SciencesOpen Access

View Full Paper

Ask AI

Bookmark

Authors

This research demonstrates improved forecasting accuracy of cold vortex paths in Northeast China, suggesting enhanced weather prediction methods.

The aim is to enhance prediction accuracy of Northeast China Cold Vortex (NCCV) trajectories using artificial intelligence techniques.
Constructed a 23-year multi-modal spatiotemporal dataset of NCCV using ERA5 reanalysis data.
Developed an improved generative adversarial network model with a multi-encoder architecture.
Introduced a multi-generator structure to enhance prediction capability.
Implemented a selector module for optimal path selection.
Conducted ablation experiments comparing single- to multi-modal data inputs.
Achieved a reduction in average prediction error by 67.96 km, a 34.0% improvement over previous models.
Improved 24-hour prediction error by 39.7%.
Demonstrated superior trajectory prediction accuracy for intervals of 6 h, 12 h, 18 h, and 24 h.
Reduced prediction distance errors by 21.4%, 29.2%, 34.0%, and 37.0% compared to LSTM models.

Jiao et al. (2026) studied this question.

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark