BlastOFormer: Attention and Neural Operator Deep Learning Methods for Explosive Blast Prediction

TL;DR

BlastOFormer is a Transformer-based surrogate model that accurately predicts blast pressure fields rapidly, offering a real-time alternative to computationally intensive CFD simulations in complex environments.

Contribution

This work introduces BlastOFormer, a novel Transformer architecture utilizing SDF encoding for fast, accurate blast pressure prediction, outperforming CNNs and FNOs.

Findings

Achieves R2 score of 0.9516 on test data.

Inference time is only 6.4 milliseconds, over 600,000 times faster than CFD.

Outperforms CNNs and FNOs in accuracy and spatial coherence.

Abstract

Accurate prediction of blast pressure fields is essential for applications in structural safety, defense planning, and hazard mitigation. Traditional methods such as empirical models and computational fluid dynamics (CFD) simulations offer limited trade offs between speed and accuracy; empirical models fail to capture complex interactions in cluttered environments, while CFD simulations are computationally expensive and time consuming. In this work, we introduce BlastOFormer, a novel Transformer based surrogate model for full field maximum pressure prediction from arbitrary obstacle and charge configurations. BlastOFormer leverages a signed distance function (SDF) encoding and a grid to grid attention based architecture inspired by OFormer and Vision Transformer (ViT) frameworks. Trained on a dataset generated using the open source blastFoam CFD solver, our model outperforms convolutional neural networks (CNNs) and Fourier Neural Operators (FNOs) across both log transformed and unscaled domains. Quantitatively, BlastOFormer achieves the highest R2 score (0.9516) and lowest error metrics, while requiring only 6.4 milliseconds for inference, more than 600,000 times faster than CFD simulations. Qualitative visualizations and error analyses further confirm BlastOFormer's superior spatial coherence and generalization capabilities. These results highlight its potential as a real time alternative to conventional CFD approaches for blast pressure estimation in complex environments.