Pipeline oil and gas leakage detection method based on graph neural network and LSTM network

• Main Authors: WANG LEI, REN JIZHOU, HU CHAOHUI, YANG DONGYING, YAO GUANGLE, LYU BING, WANG YIRU, WANG XIANG, WANG HONGHUI
• Format: Patent
• Language: Chinese; English
• Published: 18.07.2023
• Subjects: BLASTING; HEATING; LIGHTING; MEASURING; MECHANICAL ENGINEERING; PHYSICS; PIPE-LINE SYSTEMS; PIPE-LINES; STORING OF DISTRIBUTING GASES OR LIQUIDS; TESTING; TESTING STATIC OR DYNAMIC BALANCE OF MACHINES ORSTRUCTURES; TESTING STRUCTURES OR APPARATUS NOT OTHERWISE PROVIDED FOR; WEAPONS

The invention discloses a pipeline oil and gas leakage detection method based on a graph neural network and an LSTM network. The method comprises the steps that optical fiber temperature data are collected and preprocessed, abnormal temperature marking is conducted on the data, and the data are cut into sample segments; converting the sample segments into graph structure data; and constructing a fusion network formed by the graph neural network and the LSTM network, and training by using the graph structure data to obtain a trained graph neural network and LSTM network fusion model for leakage detection of the to-be-detected region. According to the novel abnormal temperature positioning method, the graph neural network is adopted to construct the spatial relationship of all signal nodes and capture the spatial characteristics of the signals, and the LSTM network is adopted to capture the time domain characteristics of the signals, so that compared with the prior art, the method comprehensively considers the