SE-MLP for Predicting Penetration Acceleration Features
Published:Dec 29, 2025 01:18
•1 min read
•ArXiv
Analysis
This paper addresses the computationally expensive nature of obtaining acceleration feature values in penetration processes. The proposed SE-MLP model offers a faster alternative by predicting these features from physical parameters. The use of channel attention and residual connections is a key aspect of the model's design, and the paper validates its effectiveness through comparative experiments and ablation studies. The practical application to penetration fuzes is a significant contribution.
Key Takeaways
- •Proposes an SE-MLP model for rapid prediction of acceleration features in penetration signals.
- •Integrates channel attention and residual connections for improved performance.
- •Demonstrates superior prediction accuracy, generalization, and stability compared to other models.
- •Validates the method's feasibility and engineering applicability for penetration fuzes.
Reference
“SE-MLP achieves superior prediction accuracy, generalization, and stability.”