Fault Diagnosis of Indium Tin Oxide Electrodes by Multi-channel S-parameter Patterns

Indium tin oxide (ITO) electrodes are integral components in a wide array of applications, from displays and sensors to solar cells. Ensuring the optimal performance and reliability of these devices necessitates effective fault diagnosis of ITO electrodes. Traditional visual inspection is challenging due to their transparency, and existing fault diagnosis methods often necessitate destructive evaluations and secondary material characterization techniques, limiting their root cause determination capabilities. In this study, we introduce a fault diagnosis approach utilizing scattering parameter (S-parameter) patterns, which provides early detection, remarkable diagnostic accuracy, and noise robustness. The proposed method incorporates Convolutional Neural Networks (CNNs) for the concurrent analysis of defect causes and severity. Experimental results demonstrate that combining different S-parameter channels significantly enhances diagnostic performance under additive noise conditions. This work not only provides an advanced fault diagnosis method for ITO electrodes but also demonstrates the importance of combining S-parameter channels to enhance fault detection.