SE-MLP Model for Predicting Prior Acceleration Features in Penetration Signals

TL;DR

This work tackles the challenge of costly simulations by predicting penetration prior acceleration features directly from physical-condition parameters. It introduces SE-MLP, a lightweight three-layer MLP augmented with squeeze-and-excitation channel attention and residual fusion to enable robust nonlinear mappings to layer-wise acceleration features. Extensive four-fold cross-validation and ablation studies demonstrate superior accuracy, stability, and generalization compared with Transformer, XGBoost, and other baselines, with the SE and residual components contributing meaningfully to performance. Numerical simulations and range tests validate the approach, showing predictions within engineering tolerances and supporting rapid generation of prior features for fuze control and layer identification in penetration scenarios.

Abstract

Accurate identification of the penetration process relies heavily on prior feature values of penetration acceleration. However, these feature values are typically obtained through long simulation cycles and expensive computations. To overcome this limitation, this paper proposes a multi-layer Perceptron architecture, termed squeeze and excitation multi-layer perceptron (SE-MLP), which integrates a channel attention mechanism with residual connections to enable rapid prediction of acceleration feature values. Using physical parameters under different working conditions as inputs, the model outputs layer-wise acceleration features, thereby establishing a nonlinear mapping between physical parameters and penetration characteristics. Comparative experiments against conventional MLP, XGBoost, and Transformer models demonstrate that SE-MLP achieves superior prediction accuracy, generalization, and stability. Ablation studies further confirm that both the channel attention module and residual structure contribute significantly to performance gains. Numerical simulations and range recovery tests show that the discrepancies between predicted and measured acceleration peaks and pulse widths remain within acceptable engineering tolerances. These results validate the feasibility and engineering applicability of the proposed method and provide a practical basis for rapidly generating prior feature values for penetration fuzes.

Figures (10)

• Figure 1: Schematic diagram of the MLP module.
• Figure 2: Schematic diagram of the residual module.
• Figure 3: Structure diagram of the SE attention mechanism.
• Figure 4: Structure diagram of the squeeze and excitation multi-layer perceptron (SE-MLP).
• Figure 5: 1/4 symmetry model of the typical structure of the projectile-fuze system.