LM6-CastBench: A Geometry-Aware Simulation Dataset for Transient Thermal Prediction in LM6 Sand Casting

概要

Transient thermal fields during solidification connect casting process conditions and cooling behaviour to the formation of defect-prone regions in aluminum cast ings. Repeatedly solving this thermal problem across geometries and process conditions is computationally expensive, motivating data-driven surrogate models. However, existing machine-learning studies on casting thermal behaviour rely on study-specific simulation databases that do not support reproducible geometry-level benchmarking. We introduce LM6-CastBench, an open simulation dataset for geometry-aware learning of transient thermal fields in LM6 aluminum sand casting, combining three geometry families, ten variants per family, five pouring tempera tures, and three mould temperatures (450 geometry-process cases in total). Each case retains CAD geometry, unstructured mesh connectivity, spatial descriptors, process metadata, and time-resolved nodal temperatures. We benchmark a point wise MLP against a spatially structured U-Net under geometry-level data splits; the U-Net reduces temperature-field errors by a large margin, with the biggest gains on geometries with localized features and abrupt section changes. LM6-CastBench gives researchers a reproducible dataset and benchmark for geometry-aware surro gate modelling of casting thermal behaviour.

参考リンク

• Paper ↗

リンク先のライセンス・再現性・データ条件をご確認ください。自動分類や実装の目安は保証ではありません。

実装を始めるプロンプト

参考リンクを確認したうえで、次の内容をコピーして利用できます。

# 論文・記事
LM6-CastBench: A Geometry-Aware Simulation Dataset for Transient Thermal Prediction in LM6 Sand Casting

# 概要
Transient thermal fields during solidification connect casting process conditions and cooling behaviour to the formation of defect-prone regions in aluminum cast ings. Repeatedly solving this thermal…

# 原典
https://arxiv.org/abs/2611.00619

上記の手法を実装してください。実行前に依存関係とライセンスを確認してください。