Monitoring pipeline operating conditions is a vital component in pipeline safety and risk management. Although the SCADA system stores a large amount of operational data, the data lacks associated condition labels, making it difficult to mine. Furthermore, the operating circumstances of the multi-product pipeline vary often, and identification and monitoring by on-site employees are prone to error, so the pipeline’s operating conditions cannot be reliably identified. To address the aforementioned challenges, this chapter presents semi-supervised learning for operating condition identification. The findings show that semi-supervised learning has more stability and improved performance regardless of how the neural network is built. The suggested technique may be utilized as a decision-making tool for monitoring and identifying multi-product pipeline operating conditions.
Operation Condition Monitoring for Pipeline
Advanced Intelligent Pipeline Management Technology ; Kapitel : 5 ; 67-79
13.12.2023
13 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Operation Condition Prediction for Pipeline
Springer Verlag | 2023
|Use of pressure surge for pipeline condition monitoring
Tema Archiv | 2012
|An integrated system for pipeline condition monitoring and pig tracking
British Library Online Contents | 2009
|Elevator operation condition monitoring device and method
Europäisches Patentamt | 2024
|Elevator operation condition bad risk monitoring system
Europäisches Patentamt | 2023
|