Construction of an LNG Carrier Port State Control Inspection Knowledge Graph by a Dynamic Knowledge Distillation Method

• Published in: Journal of marine science and engineering Vol. 13; no. 3; p. 426
• Main Authors: Gan, Langxiong, Yang, Qihao, Xu, Yi, Mao, Qiongyao, Liu, Chengyong
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 25.02.2025
• Subjects: Accident investigations; Accuracy; Artificial intelligence; Computational linguistics; Decision making; Deep learning; Distillation; Distilling; dynamic multi-teacher weight allocation; Gas detectors; Graphs; Inspection; Inspections; Knowledge acquisition; knowledge distillation; knowledge graph; Knowledge representation; Labels; Language; Language processing; Liquefied natural gas; Marine transportation; Methods; multilayer soft label fusion; Multilayers; Natural gas; Natural language interfaces; Neural networks; Performance evaluation; port state control inspection; Ports; Recognition; Ship accidents & safety; Shipping industry; Teachers; Traffic accidents & safety; Weight; China
• DOI: 10.3390/jmse13030426
• ISSN: 2077-1312

The Port State Control (PSC) inspection of liquefied natural gas (LNG) carriers is crucial in maritime transportation. PSC inspection requires rapid and accurate identification of defects with limited resources, necessitating professional knowledge and efficient technical methods. Knowledge distillation, as a model lightweighting approach in the field of artificial intelligence, offers the possibility of enhancing the responsiveness of LNG carrier PSC inspections. In this study, a knowledge distillation method is introduced, namely, the multilayer dynamic multi-teacher weighted knowledge distillation (MDMD) model. This model fuses multilayer soft labels from multi-teacher models by extracting intermediate feature soft labels and minimizing intermediate feature knowledge fusion. It also employs a comprehensive dynamic weight allocation scheme that combines global loss weight allocation with label weight allocation based on the inner product, enabling dynamic weight allocation across multiple teachers. The experimental results show that the MDMD model achieves a 90.6% accuracy rate in named entity recognition, which is 6.3% greater than that of the direct training method. In addition, under the same experimental conditions, the proposed model achieves a prediction speed that is approximately 64% faster than that of traditional models while reducing the number of model parameters by approximately 55%. To efficiently assist in PSC inspections, an LNG carrier PSC inspection knowledge graph is constructed on the basis of the recognition results to quickly and effectively support knowledge queries and assist PSC personnel in making decisions at inspection sites.