Pipeline Threat Event Identification Based On GAF Of Distributed Fiber Optic Signals

Jun Li,Ruixu Yao,Jiarui Zhang,Xinwei Zhang,Meiying Ren,Tian Ma

DOI: https://doi.org/10.1109/jsen.2023.3315933

IF: 4.3

2023-01-01

IEEE Sensors Journal

Abstract:This paper proposes a long distance pipeline threat event identification method based on Gramian Angular Field (GAF) and deep learning networks for phase sensitive Rayleigh distributed fiber optic vibration system. Eleven types of threat events of vibration signals along a 28.9 km natural pipeline are collected and converted to two-dimensional images using GAF. The performance of five deep learning networks: GoogLeNet, VGG, AlexNet, ShuffleNet, and DenseNet are compared for events classification. The results show that DenseNet achieved the highest classification accuracy within the eleven types of threat events. Moreover, the false alarm rates of manual digging and mechanical excavation among the eleven events were 0.31% and 0.18%, respectively. The proposed method is suitable for complicated pipeline safety monitoring niche and offers fast and accurate recognition solution for long distance pipeline safety monitoring.

engineering, electrical & electronic,instruments & instrumentation,physics, applied