Published In

IEEE Access

Document Type

Article

Publication Date

10-22-2025

Subjects

Battery Applications

Abstract

The battery tab-to-busbar welding process forms the primary electrical path in battery packs and is directly linked to both performance and safety. Variations in jig alignment, material tolerance, and forming quality can cause small interfacial gaps that lead to weak welds. An artificial intelligence (AI)-driven method was proposed in this study for predicting interfacial gaps in aluminum-copper overlap joints by integrating deep learning with multi-sensor data. A charge-coupled device (CCD) camera, spectrometer, and optical coherence tomography (OCT) sensors were employed to develop and validate deep learning models under varying gap conditions. The results revealed that the variation in melt-pool dimensions, changes in keyhole behavior, intensity variations at specific wavelengths, and keyhole depth derived from the OCT data enabled accurate gap classification. A 0.04-mm binary classification model achieved the highest accuracy of 99.33%. Among the sensors, the spectrometer was the most influential sensor, whereas the CCD and OCT sensors provided complementary inputs. The best performance was achieved on fusing all three sensors, which emphasizes the importance of sensor fusion for precise gap prediction.

Rights

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/

Locate the Document

10.1109/ACCESS.2025.3621683

DOI

10.1109/ACCESS.2025.3621683

Persistent Identifier

https://archives.pdx.edu/ds/psu/44197

Publisher

IEEE

Download

Share

COinS