Semantic Mining and Domain Correction Network for Online Defect Detection of Photovoltaic Data Streams With Nonstationary Distributions

The online defect detection for actual photovoltaic (PV) data streams is challenging, as the data stream from multiple production lines are unsupervised and unevenly distributed, while existing static defect detectors cannot adapt to such unsupervised nonstationary PV data stream. In this article, we propose a novel semantic mining and domain correction (SMDC) network, which can aid the static defect detector adapt to nonstationary PV data streams encountered in deployment environments in an online unsupervised manner. In SMDC, we proposes a consistency-based semantic-knowledge mining (CSM) module, which selects reliable low-confidence teacher predictions as pseudo labels for potential defective instances during the self-training process based on the teacher-student architecture, so as to better mine the potential semantic knowledge of unlabeled samples in the nonstationary data stream. Then, we introduce a domain correction contrastive learning (DCCL) module, which performs contrastive learning to measure and narrow the distance between defect-free features of test samples and prepreserved background prototypes in the feature space, thereby helping detector to learn robust domain-invariant background feature representations. Finally, extensive experiments based on the public PV dataset demonstrate that the proposed SMDC achieves the state-of-the-art performance for the online defect detection task of the nonstationary PV data stream, which is crucial for the continuous quality monitoring of PV products.