A welding penetration prediction model for laser welding process based on self-supervised learning using physics-informed neural networks

2026-06-24

AI Summary

This study addresses the challenge of predicting weld penetration status in laser welding, which is hindered by the scarcity of large-scale, high-quality annotated data. To overcome this limitation, the authors propose SimPhysNet, a novel model that, for the first time, integrates physical priors into a self-supervised contrastive learning framework by synergistically combining physics-informed neural networks (PINNs) with prototypical networks. This approach enables the extraction of physically meaningful features from unlabeled melt pool images and achieves high-accuracy few-shot classification. Remarkably, using only 200 labeled images—approximately 5% of a typical dataset—the method attains a classification accuracy of 96.06%, matching the performance of fully supervised baselines.

Abstract

The laser welding full-penetration is of critical importance, as it constitutes one of the fundamental factors in achieving defect-free welded joints. Accurate prediction of the penetration state is therefore essential for ensuring weld quality. To this end, this paper introduces SimPhysNet, a novel algorithm that achieves high classification accuracy in laser welding penetration prediction using only a limited number of labelled images. This approach effectively overcomes the limitations of supervised learning classification algorithms, which are hindered in industrial applications by their dependence on extensive, high-quality labelled data. The core of SimPhysNet is a unique self-supervised learning paradigm that embeds physical priors into a contrastive learning framework. By incorporating a physics-informed neural network (PINN), the model is guided to extract physically meaningful features of the molten pool and keyhole from a large set of unlabelled data, while three image augmentation tasks further enhance its generalization capabilities. Subsequently, a few-shot learning strategy, based on prototypical networks, enables robust classification by constructing class representations from a minimal set of labelled images. Experimental results demonstrate that SimPhysNet achieves a classification accuracy of 96.06% using only 200 labelled images (approximately 5% of the total labelled dataset), which is comparable to the performance of conventional supervised learning algorithms that utilize the entire labelled dataset. This work presents a new, efficient, and highly accurate method, providing the way for the intelligent automation of laser welding.

Problem

• laser welding
• penetration prediction
• limited labelled data
• weld quality
• self-supervised learning

Innovation

• self-supervised learning
• physics-informed neural networks
• few-shot learning
• laser welding penetration prediction
• contrastive learning