A Deep Surrogate Model for Robust and Generalizable Long-Term Blast Wave Prediction

TL;DR

This work tackles the challenge of robust, long-horizon blast-wave prediction in urban environments by introducing RGD-Blast, a deep surrogate that integrates a multi-scale feature extractor with dynamic-pressure and static-layout cues and a ConvGRU-based encoder-decoder to mitigate autoregressive error growth. The model jointly captures global flow patterns and local boundary interactions while fusing time-varying fields with static source and layout features, enabling strong out-of-distribution generalization. Empirical results on CFD-generated datasets show two orders of magnitude speedups over numerical methods while achieving high fidelity (e.g., RMSE < 0.01 and R^2 > 0.89 over 280 steps) and robust performance across unseen layouts, source locations, and charge weights; ablations confirm the necessity of the dynamic-static coupling and multi-scale temporal modeling. The approach offers practical impact for rapid structural damage assessment and urban-explosion risk analysis, with future work extending to full 3D urban environments and coupled structural response analyses.

Abstract

Accurately modeling the spatio-temporal dynamics of blast wave propagation remains a longstanding challenge due to its highly nonlinear behavior, sharp gradients, and burdensome computational cost. While machine learning-based surrogate models offer fast inference as a promising alternative, they suffer from degraded accuracy, particularly evaluated on complex urban layouts or out-of-distribution scenarios. Moreover, autoregressive prediction strategies in such models are prone to error accumulation over long forecasting horizons, limiting their robustness for extended-time simulations. To address these limitations, we propose RGD-Blast, a robust and generalizable deep surrogate model for high-fidelity, long-term blast wave forecasting. RGD-Blast incorporates a multi-scale module to capture both global flow patterns and local boundary interactions, effectively mitigating error accumulation during autoregressive prediction. We introduce a dynamic-static feature coupling mechanism that fuses time-varying pressure fields with static source and layout features, thereby enhancing out-of-distribution generalization. Experiments demonstrate that RGD-Blast achieves a two-order-of-magnitude speedup over traditional numerical methods while maintaining comparable accuracy. In generalization tests on unseen building layouts, the model achieves an average RMSE below 0.01 and an R2 exceeding 0.89 over 280 consecutive time steps. Additional evaluations under varying blast source locations and explosive charge weights further validate its generalization, substantially advancing the state of the art in long-term blast wave modeling.

Figures (16)

• Figure 1: Machine learning-based surrogate for Blast Wave Prediction
• Figure 2: Autoregressive forecasting process using a sliding window, illustrating the feedback loop and the tendency for error accumulation.
• Figure 3: Multi-scale feature extraction. The input integrates dynamic (pressure, temporal) and static (distance, layout) features into a multi-channel tensor. This rich input is fed into a multi-scale module at the beginning of the encoder.
• Figure 4: Overview of the RGD-Blast model, illustrating the encoder-decoder structure and the central ConvGRU for temporal modeling.
• Figure 5: Overview of spatio-temporal generalization tasks with different training and testing scenarios for RGD-Blast. (a) Generalization task for unseen layouts with random buildings. (b) Generalization task for unseen source locations, where the model is trained on known sources and tested on unknown ones. (c) Generalization task for unseen explosive charge weights, testing the model's adaptability to varying explosive charge weights.