Construction and Application of a Cross Regional Complex Oil and Gas Pipeline Operation Monitoring and Sensing System Based on Digital Twins

As a crucial backbone of national energy supply, the oil and gas pipeline system is characterized by extensive scale, harsh and variable storage conditions, the presence of high-pressure flammable media, and multiple potential risks to its safe operation. For the perception and risk assessment of oil and gas leakage scenarios, this paper integrates leakage monitoring sensor data, existing pipeline production and operation data, and environmental monitoring data. Isolation forest algorithm and object detection algorithm are used for oil and gas leakage anomaly monitoring. The advantage of D-S evidence theory in handling uncertain information is utilized to perform decision level fusion on the results of object detection algorithm and anomaly detection algorithm, and output the final fused leakage diagnosis decision result. Establish standardized sensor specifications and heterogeneous fusion network architecture that meet the needs of ubiquitous perception, form multi-scale monitoring and global joint defense security assessment technology for the entire lifecycle of pipeline lines, ontology, and environment, build a digital twin application platform for pipeline intelligent sensing systems, and carry out demonstration applications in typical scenarios of strategic pipelines such as the China-Myanmar pipeline and the China-Russia pipeline.