DamFormer: Generalizing Morphologies in Dam Break Simulations Using Transformer Model
This work addresses flood and tsunami disaster modeling for civil engineering and safety planning, but appears incremental as it applies a known transformer architecture to a specific domain without broad SOTA claims.
The paper tackled the problem of simulating wave interactions with different structural shapes like circles, triangles, and squares for flood defense applications, and introduced DamFormer, a transformer-based model that learned and simulated these complex dynamics using simulated data.
The interaction of waves with structural barriers such as dams breaking plays a critical role in flood defense and tsunami disasters. In this work, we explore the dynamic changes in wave surfaces impacting various structural shapes, e.g., circle, triangle, and square, by using deep learning techniques. We introduce the DamFormer, a novel transformer-based model designed to learn and simulate these complex interactions. The model was trained and tested on simulated data representing the three structural forms.