Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DamFormer: Generalizing Morphologies in Dam Break Simulations Using Transformer Model

Zhaoyang Mul, Aoming Liang, Mingming Ge, Dashuai Chen, Dixia Fan, Minyi Xu

Abstract

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.

DamFormer: Generalizing Morphologies in Dam Break Simulations Using Transformer Model

Abstract

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.

Paper Structure

This paper contains 11 sections, 3 equations, 7 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Dataset and solver Description
  4. Model introduction
  5. Learning Objective
  6. Results and Discussions
  7. Performance in the circle boundary
  8. Performance in the square boundary
  9. Performance in the triangle boundary
  10. Performance on transfer learning results on generalized boundary conditions
  11. Concluding remarks

Figures (7)

  • Figure 1: Physical boundary and initial condition are demonstrated through different boundary shapes.
  • Figure 2: Numerical solution data on a circular boundary, from left to right, corresponding sequentially to the evolution at 0s, 0.5s, 1s, 2s, 2.5s, and 3s.
  • Figure 3: The architecture of DamFormer
  • Figure 4: Comparison between the ground truth and output of Damformer
  • Figure 5: Comparison between the ground truth and output of Damformer
  • ...and 2 more figures