IOT Enabled Oil Pipeline Leak Detection using Magnetic Flux Leakage and LSTM Neural Network

¹Vignesh S, ²Muhilan M, ³Vishalkirthik V, ⁴ M Sujitha, ⁵ V Sai Shanmuga Raja
1,2,3 Department Of Cse, Dr.M.G.R. Educational And Research Institute, Chennai , India.

Abstract— Despite being vital infrastructure, oil and gas pipelines frequently experience cracks and leaks due to aging, corrosion, and mechanical stress, whichcan result in financial and environmental losses.  Conventional leak detection methods rely on periodic manual inspections and threshold-based analysis, which lack real-time monitoring and struggle to interpret complex time-series magnetic signals. This studysuggests an Internet of Things (IoT)-enabled intelligent  pipeline monitoring system that combines Long Short Term Memory (LSTM) neural networks with MagneticFlux Leakage (MFL) sensing to automatically detect  leaks. Real-time data is collected by Hall-effect,temperature, and vibration sensors that are connected to an ESP32 controller. This data is then sent to a cloud platform for preprocessing and temporal analysis using LSTM. A laboratory prototype's experimental results demonstrate enhanced reliability, fewer false alarms,and efficient predictive maintenance with 96%detection accuracy and response latency under 2 seconds. The suggested system offers a scalable andaffordable way to monitor the integrity of smartpipelines.
Keywords— Deep Learning, Internet of Things (IoT), Leak Detection, Long Short-Term Memory (LSTM),Magnetic Flux Leakage (MFL), Pipeline Monitoring.2