基于DEM-FEM耦合的滑坡作用下输气管道动力响应分析

油气长输管道输送距离长，途径地形地貌复杂多样，沿途地质灾害频发对油气管道的安全输送造成严重威胁与破坏。本文以联络线输气管道横向滑坡为例，通过离散元与有限元单向耦合方式，对管道进行滑坡作用下的动力响应分析。首先通过EDEM离散元软件模拟滑坡过程中土体颗粒对管道的动态冲击过程，分析在滑坡过程中土体颗粒与管道的接触变化以及受力变化，然后将管道所受最大滑坡推力作为荷载导入ANSYS中，模拟得到在滑坡及内压协同作用下管道的应力和位移分布，最后将模拟结果与线上监测数据进行对比。结果表明：由离散元模拟得到滑坡作用下输气管道受力时程变化曲线，在第3秒管道受到的滑坡推力最大，将其导入有限元中模拟得到滑坡段管道等效应力、剪切应力与位移分布云图，将等效应力与实际监测数据进行对比，计算得到相对误差在7.30%以内，因此表明该模型具有一定的可靠性。

Dynamic Response Analysis of Gas Pipeline under Landslide Based on DEM-FEM Coupling

The long-distance oil and gas pipeline has a long transportation distance, and the terrain is complex and diverse. The frequent occurrence of geological disasters along the way will cause serious threats and damage to the safe transportation of oil and gas pipelines. In this paper, taking the lateral landslide of a Lian-Luo gas pipeline as an example, the dynamic response of the pipeline under the action of the landslide is analyzed through the unidirectional coupling method of discrete element and finite element. Firstly, the EDEM discrete element software was used to simulate the dynamic impact process of soil particles on the pipeline during the landslide process, and the changes of contact state and stress between the soil particles and the pipeline in the process of landslide were analyzed. Then the maximum landslide thrust experienced by the pipeline was imported into ANSYS as the load, and the stress and displacement distribution of the pipeline under the synergistic effect of the landslide and internal pressure was simulated. Finally, the simulation results were compared with the online monitoring data. The results indicate that the stress time history curve of the gas transmission pipeline under the action of landslide was obtained by discrete element simulation, and the pipeline received the maximum landslide thrust in the third second. Imported it into the finite element simulation to obtain the equivalent stress, shear stress and displacement distribution cloud map of the landslide section pipeline. Comparing the equivalent stress with the actual monitoring data, the relative error was less than 7.30%, which shows that the model has certain reliability.