A Real-Time Monitoring Framework for Ship Pipeline Status Based on Ship-Shore Coordination

This paper establishes an intelligent pipeline system to predict the overall state of the corresponding pipeline. In order to implement the method, a pipe-aware network needs to be built. Then, use historical data to train machine learning models, obtain a pipeline state prediction model, and constr...

Full description

Saved in:
Bibliographic Details
Published in2024 8th International Conference on System Reliability and Safety (ICSRS) pp. 270 - 277
Main Authors Chen, Haonan, Fan, Shidong, Wang, Bin, Liu, Aihua, Zhu, Hanhua, Jiang, Pan, Wang, Zhanwei
Format Conference Proceeding
LanguageEnglish
Published IEEE 20.11.2024
Subjects
Accuracy
Data models
Deep learning
machine learning
Marine vehicles
Pipelines
Predictive models
real-time monitoring
Real-time systems
Reliability
ship pipelines
Stacking
Online AccessGet full text

Cover

More Information
Summary:This paper establishes an intelligent pipeline system to predict the overall state of the corresponding pipeline. In order to implement the method, a pipe-aware network needs to be built. Then, use historical data to train machine learning models, obtain a pipeline state prediction model, and construct an intelligent pipeline system for the pipeline. The test was conducted on the built pipeline platform, and the results show that this method has high accuracy in pipeline leakage prediction, with an R2 of 0.98. The LSTM method has slightly lower accuracy than this method. At the same time, the model can accurately express the leakage trend and play a certain role in the problem of difficult prediction of micro leakage states. Through experiments conducted on a pipeline platform, it has been proven that the intelligent pipeline system can accurately reflect the state of the pipeline when making real-time predictions.
DOI:10.1109/ICSRS63046.2024.10927472