What question did this study set out to answer?

The central aim is to develop a machine learning framework that accurately predicts welding-induced residual stresses in steel girders.

April 19, 2026Scientific ReportsOpen Access

A hybrid decoupled machine learning framework with physical constraints for predicting welding-induced residual stresses in steel girders

View Full Paper

Ask AI

Bookmark

Authors

JZJingwen ZhangXijing University WHWenbo HuangPowerChina (China)ZNZhipeng NiPowerChina (China)

Discussion

Member takes

Implication

Demonstrates a machine learning framework that predicts residual stresses in welded steel girders, indicating improved efficiency in design and fabrication.

Key Points

The central aim is to develop a machine learning framework that accurately predicts welding-induced residual stresses in steel girders.
Stage I uses Gaussian Process Regression to predict thermal descriptors from various parameters.
Stage II employs a gradient-boosted decision tree model to estimate tensile and compressive stress peaks.
A physics-constrained optimization ensures mechanical admissibility by applying material yield limits.
The framework is trained on thermomechanical finite element simulations and validated with independent tests.
Achieved high prediction accuracy in estimating residual stresses.
Significantly reduced computational time compared to traditional methods.
Successfully reproduced experimentally observed structural behavior in steel girders.

Cite This Study

Zhang et al. (2026) studied this question.

synapsesocial.com/papers/69e47282010ef96374d8e8e8 https://doi.org/10.1038/s41598-026-48635-8

View Full Paper

Ask AI

Bookmark

Implication

Demonstrates a machine learning framework that predicts residual stresses in welded steel girders, indicating improved efficiency in design and fabrication.

A hybrid decoupled machine learning framework with physical constraints for predicting welding-induced residual stresses in steel girders

Discussion

Member takes

Implication

Key Points

Cite This Study

A hybrid decoupled machine learning framework with physical constraints for predicting welding-induced residual stresses in steel girders

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider