Otwarty dostęp

Digital Twin-Based Real-Time Monitoring and Intelligent Maintenance System for Oil and Gas Pipelines

Yihan Wang

Yihan Wang

College of Petroleum and Natural Gas Engineering, Southwest Petroleum UniversitySichuan, China
Wyszukaj tego autora
Sciendo|Google Scholar
Wang, Yihan
  
11 kwi 2025
Digital Twin-Based Real-Time Monitoring and Intelligent Maintenance System for Oil and Gas Pipelines's Cover Image
O artykule
Poprzedni artykuł
Następny artykuł
Zacytuj
Pobierz okładkę
Data publikacji: 11 kwi 2025
Otrzymano: 05 lis 2024
Przyjęty: 05 mar 2025
DOI: https://doi.org/10.2478/amns-2025-0849
Słowa kluczowe
© 2025 Yihan Wang, published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1.

Overall workflow of the digital twin-based monitoring and maintenance system. Data from pipeline sensors are incorporated into the digital twin via the EnKF, and the updated states are analyzed by a machine learning module. Anomalies trigger maintenance actions, and the entire process adapts over time through continual retraining and model refinement.
Overall workflow of the digital twin-based monitoring and maintenance system. Data from pipeline sensors are incorporated into the digital twin via the EnKF, and the updated states are analyzed by a machine learning module. Anomalies trigger maintenance actions, and the entire process adapts over time through continual retraining and model refinement.

Figure 2.

Experiment 1 (Baseline Condition): Average residual magnitude remains low and stable, indicating accurate alignment between sensor data and model forecasts.
Experiment 1 (Baseline Condition): Average residual magnitude remains low and stable, indicating accurate alignment between sensor data and model forecasts.

Figure 3.

Experiment 2 (Localized Corrosion): Wall thickness loss over time with the red dashed line indicating the point when the machine learning module issued a high-confidence anomaly alert.
Experiment 2 (Localized Corrosion): Wall thickness loss over time with the red dashed line indicating the point when the machine learning module issued a high-confidence anomaly alert.

Figure 4.

Experiment 3 (Combined Leak and Blockage): Timeline showing the onset and detection of two simultaneous anomalies. Dashed lines mark the true start of each anomaly, and dots indicate when the system raised alarms.
Experiment 3 (Combined Leak and Blockage): Timeline showing the onset and detection of two simultaneous anomalies. Dashed lines mark the true start of each anomaly, and dots indicate when the system raised alarms.
Język:
Angielski
Częstotliwość wydawania:
1 razy w roku
Dziedziny czasopisma:
Nauki biologiczne, Nauki biologiczne, inne, Matematyka, Matematyka stosowana, Matematyka ogólna, Fizyka, Fizyka, inne
Kanał RSS czasopisma