Multiphysics Bench
收藏资源简介:
Multiphysics Bench是一个专注于使用机器学习解决多物理场偏微分方程(PDEs)的数据集。它是迄今为止最全面的PDE数据集,拥有最广泛的耦合类型、最多的PDE公式和最大的数据集规模。数据集涵盖了六个典型的耦合场景,包括电磁场、传热、流体流动、固体力学、声学压力和质量传输。该数据集的创建旨在解决多物理场问题,并提供了机器学习在解决这些复杂问题方面的基准和见解。
Multiphysics Bench is a dataset dedicated to solving multiphysics partial differential equations (PDEs) via machine learning. It is the most comprehensive PDE dataset to date, featuring the widest range of coupling types, the largest number of PDE formulations, and the largest dataset scale. The dataset encompasses six typical coupling scenarios, including electromagnetic fields, heat transfer, fluid flow, solid mechanics, acoustic pressure, and mass transport. This dataset was developed to tackle multiphysics problems, and provides benchmarks and insights for applying machine learning to solve these complex problems.

- 1Multiphysics Bench: Benchmarking and Investigating Scientific Machine Learning for Multiphysics PDEs香港科技大学(广州)xLeaF实验室, 新加坡国立大学, 上海人工智能实验室 · 2025年



