Research on underground pipeline illegal intrusion monitoring based on distributed optical fiber sensing technology

The transportation of oil and gas primarily relies on pipelines, but unauthorized construction and illegal excavation often damage these systems, leading to frequent leaks and significant economic and environmental repercussions. Existing methods for detecting third-party damage to pipelines are largely ineffective, susceptible to environmental interference, and often generate high false alarm rates. To enhance inspection accuracy and improve early warning capabilities, this paper presents a monitoring scheme based on Distributed Optical Fiber Sensing Technology (DOFS). This technology effectively monitors pipeline safety by leveraging its integrated sensing and transmission capabilities, long measurement distances, and high accuracy. The research utilizes Distributed Acoustic Sensing (DAS) to analyze backscattered signals and assess the maximum acoustic bandwidth of disturbances. Fast Fourier Transform (FFT) extracts spectral features to establish frequency and amplitude thresholds for event detection, while machine learning algorithms accurately identify and localize leak positions. Additionally, circumferentially distributed strain sensing optical fibers are deployed along the pipeline to measure circumferential deformation and internal pressure changes caused by excavation, drilling, or intentional damage. To reduce false alarms in detecting damage signals, the study combines short-term average energy with corrected zero-crossing rate standard deviation, enhancing detection accuracy. Ultimately, this research aims to establish a highly stable, high-precision, and integrated detection method for effective early warning and timely response to underground pipeline safety incidents, addressing the critical challenges in pipeline monitoring.