Work

Multimodal sensor fusion with SWIR imaging and audio for inline gas-tightness monitoring in laser-welded bipolar plates

Year: 2025

Type: article

Abstract: The gas-tightness of laser-welded bipolar plates (BPPs) is a key quality feature for the functionality of fuel cells. Pressure drop tests to check gas-tightness are very time-consuming and expensive, posing a challenge for large-scale production. This study proposes a novel sensor fusion approach combining short-wavelength infrared (SWIR) imaging and airborne audio signals, analyzed using recurrent neural networks, to enable real-time gas-tightness monitoring during the manufacturing process. The experimental results show a classification accuracy of 96 %, a substantial reduction in overkill rates compared to single-sensor evaluations, and the elimination of escapes that could lead to leaks. The new approach reduces the number of gas-tightness tests required by 47 %.

Source: Optics & Laser Technology

DOI: 10.1016/j.optlastec.2025.113441

Authors: Manuel Klaiber, Matthias Hartmann, Jan-Patrick Hermani, Andreas Jahn, Andreas Michalowski

Institutions: University of Stuttgart, TRUMPF (Germany)

Language: English

Topic: Industrial Vision Systems and Defect Detection

Subfield: Industrial and Manufacturing Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: hybrid

Estimated APC paid: $3,230