Enhanced Crude Oil Pipeline Leakage Detection System Using Reinforcement Learning

Nkolika O. Nwazor; Lloyd E. Ogbondamati

Nkolika O. Nwazor Department of Electrical and Electronics Engineering, University of Port Harcourt, Rivers State, Nigeria
Lloyd E. Ogbondamati Department of Electrical and Electronics Engineering, University of Port Harcourt, Rivers State, Nigeria

Keywords: Reinforcement Learning, Pipeline, Detection, Leakage, Control, Artificial Intelligence

Abstract

This work used advanced reinforcement learning (RL) techniques to enhance crude oil pipeline leakage detection systems to address the frequent occurrence of undetected or misidentified leaks, which can lead to significant financial losses and safety risks. The existing pipeline monitoring system's performance was analysed, identifying the detection accuracy and response time limitations. Reinforcement learning algorithms were then integrated to optimize the system's ability to detect leaks and minimize false positives. The RL model was trained to enhance its performance through iterative learning and feedback, ultimately improving its accuracy to 100% and increasing precision to 100%. This was achieved by adjusting detection thresholds and refining control actions based on real-time data. When the system detected a pressure drop from 55 psi to 36 psi, reinforcement control measures were successfully implemented to restore pressure to 55 psi. Additionally, a leakage was accurately located at 600 meters along the pipeline, allowing for targeted intervention. The results also demonstrated the impact of flow rate and pressure variations on detection performance, emphasizing the importance of dynamic and responsive control strategies. The integration of RL techniques offers a significant advancement over traditional methods, providing a robust framework for managing pipeline integrity and ensuring environmental safety. This study sets a precedent for future developments in pipeline monitoring and management, advocating for the continuous incorporation of innovative technologies in maintaining infrastructure resilience.

References

Ahmed, S., Le Mouël, F., Stouls, N., & Lipeme Kouyi, G., (2023). Development and analysis of a distributed leak detection and localisation system for crude oil pipelines. Sensors, 23(9), 4298. doi: 10.3390/s23094298.
Kannan, P. & Deepan, C. (2021). Examining The Effectiveness of Wireless Sensor Network in Underwater Network Pipeline Management. International Journal of Production Technology and Management (IJPTM), 12(1), 1-10.
Khawar, R. & Faiza, N. (2017). Remote Pipeline Monitoring Using Wireless Sensor Networks. 2017 International Conference on Communication, Computing and Digital Systems (C-CODE).
Korlapati, N. V. S., Khan, F., Noor, Q., Mirza, S., & Vaddiraju, S. (2022). Review and analysis of pipeline leak detection methods. Journal of Pipeline Science and Engineering, 2(4), 100074. doi: 10.1016/j.jpse.2022.100074.
Masanobu, S., Pai. H., Chouad, S. K., Hong. R. K., Eunbae, Y., Hadil, M., Debasis, K. & Selim, P. (2020). Nondestructive Monitoring of a Pipe Network using a MEMS-Based Wireless Network.
Mathurine, G., Temidayo, J. O., Kehinde, O. O., Hilary, U. E. & Adedayo, B. (2017). Design and Implementation of a Compact Home Security System. FUOYE Journal of Engineering and Technology, 2(2), 17 -22.
Meribout, M. (2011). A wireless sensor network-based infrastructure for realtime and online pipeline inspection. IEEE Sensors Journal, 11(11), 2966-2972.
Milad, G., Aidin, G. A., & Mohammed, A. A. (2016). Pipeline Monitoring System by Using Wireless Sensor Network. IOSR Journal of Mechanical and Civil Engineering, 13(3), 43-53.
Mohamed, M. I., Wu, W. Y. & Moniri, M. (2011). Power Harvesting for Smart Sensor Networks in Monitoring Water Distribution System. International Conference on Networking, Sensing and Control Delft, the Netherlands, 304-308.
Mohamed, N., Jawhar, I., Al-Jaroodi, J. & Zhang, L. (2010). Monitoring underwater pipelines using sensor networks. In Proc. of 12th IEEE Int. Conf. on High Performance Computing and Communications, 346-353.
Mohammed, S. B., Syed, M. Q. & Abdulfattah, M. O. & Alberto, G. O. (2014). Review on Wireless Sensor Network for Water Pipeline Monitoring Applications.
Nicholas, N. T., Ikechi, R., & Judah, A. R. (2020). Design and Implementation of Intruder Detector System with SMS Alert. American Journal of Engineering Research (AJER), 9(11), 128-132.
Oseni, W. Y., Akangbe, O. A., & Abhulimen, K. (2023). Mathematical modelling and simulation of leak detection system in crude oil pipeline. Heliyon, 9(4), e15412. doi: 10.1016/j.heliyon.2023.e15412.