ABSTRACT
The increasing deployment of industrial robots in manufacturing requires accurate fault diagnosis. Online monitoring data typically consist of a large volume of unlabeled data and a small quantity of labeled data. Conventional intelligent diagnosis methods heavily rely on supervised learning with abundant labeled data. To address this issue, this paper presents a semi-supervised Informer algorithm for fault diagnosis modeling, leveraging the Informer model's long- and short-term memory capabilities and the benefits of semi-supervised learning to handle the diagnosis of a small amount of labeled data alongside a substantial amount of unlabeled data. An experimental study is conducted using real-world industrial robot monitoring data to assess the proposed algorithm's effectiveness, demonstrating its ability to deliver accurate fault diagnosis despite limited labeled samples.
ABSTRACT
BACKGROUND: Sophora tonkinensis Gagnep is a traditional Chinese medical plant that is mainly cultivated in southern China. Drought stress is one of the major abiotic stresses that negatively impacts S. tonkinensis growth. However, the molecular mechanisms governing the responses to drought stress in S. tonkinensis at the transcriptional and posttranscriptional levels are not well understood. RESULTS: To identify genes and miRNAs involved in drought stress responses in S. tonkinensis, both mRNA and small RNA sequencing was performed in root samples under control, mild drought, and severe drought conditions. mRNA sequencing revealed 66,476 unigenes, and the differentially expressed unigenes (DEGs) were associated with several key pathways, including phenylpropanoid biosynthesis, sugar metabolism, and quinolizidine alkaloid biosynthesis pathways. A total of 10 and 30 transcription factors (TFs) were identified among the DEGs under mild and severe drought stress, respectively. Moreover, small RNA sequencing revealed a total of 368 miRNAs, including 255 known miRNAs and 113 novel miRNAs. The differentially expressed miRNAs and their target genes were involved in the regulation of plant hormone signal transduction, the spliceosome, and ribosomes. Analysis of the regulatory network involved in the response to drought stress revealed 37 differentially expressed miRNA-mRNA pairs. CONCLUSION: This is the first study to simultaneously profile the expression patterns of mRNAs and miRNAs on a genome-wide scale to elucidate the molecular mechanisms of the drought stress responses of S. tonkinensis. Our results suggest that S. tonkinensis implements diverse mechanisms to modulate its responses to drought stress.
