Investigation of gas distribution pipes subjected to ground movements

管道是现代社会的生命线，为我们的生活提供水和能源。然而，管道事故，如泄漏和故障，可能会造成重大的经济和环境后果。能源管道泄漏会污染环境并引发火灾。地面运动已被确定为管道事故的原因之一。埋在移动地面中的管道承受的载荷会对管壁产生过大的应力和应变，从而导致泄漏和/或故障。然而，目前尚无可接受的设计方法来评估由于地面移动载荷引起的管壁应力和应变，特别是用于输配系统的小直径管道。当前研究的目标是开发受地面运动载荷影响的天然气分配管道的应力/应变评估方法和管道完整性评估框架，以识别高风险区域。将进行全面的实验室测试和土壤-管道相互作用的有限元（FE）建模，以开发用于评估天然气分配网络中管道的简化工具。实践工程师需要简化的设计工具来评估在役管道的运行状况，而详细的有限元建模被认为是昂贵和不方便的。测试将使用一种新的实验室测试方法来模拟现场预期的地面移动场景，以实验方式检查管道的行为。有限元建模技术将使用Abaqus有限元软件来模拟从试验中观察到的响应。提出了利用实验室试验数据和有限元模型，应用监督机器学习回归方法来开发简化的管道应变评估方法。这项研究的结果将促进对管道经历地面运动的认识，并将有助于减轻管道故障的风险。因此，研究结果将有助于有效和安全地向加拿大人提供天然气供应，从而保护加拿大环境免受天然气泄漏的影响。

Pipelines are lifelines of modern society, which are used to supply water and energy for our living. However,pipeline incidents, such as leaks and failure, can have significant economic and environmental consequences.Leaking from energy pipelines can contaminate the environment and cause a fire hazard. The groundmovement has been identified as one of the causes of pipeline incidents. Pipes buried in the moving groundexperience loads that induce excessive stresses and strains on the wall, causing leaks and/or failures. However,no acceptable design method is currently available for the assessment of pipe wall stresses and strains due tothe ground movement loads, particularly for small diameter pipes used in gas transmission and distributionsystems. The objectives of the current study are to develop methods for stress/strain assessment of gasdistribution pipes subjected to the ground movement loads and a pipeline integrity assessment framework toidentify high-risk areas. Full-scale laboratory testing and finite element (FE) modelling of soil-pipe interactionwill be performed to develop simplified tools for the assessment of the pipes in the gas distribution network. Asimplified design tool is required for practicing engineer for the assessment of the operational conditions ofpipes in service, where detailed FE modelling is considered costly and inconvenient. Tests will be conducted toexperimentally examine the behavior of the pipes using a novel laboratory test method to simulate groundmovement scenarios expected in the field. FE modelling techniques will be developed using Abaqus FEsoftware to simulate the observed responses from the tests. It is proposed to apply supervised machine learningregression methods using the data from the laboratory tests and FE modelling to develop the simplified methodof pipe strain assessment. The findings from this research will advance the knowledge on pipelinesexperiencing ground movements and would be useful for mitigating the risks of pipeline failures. Thus, theresearch findings will contribute to providing natural gas supply to the Canadians effectively and safely andthus, to the protection of the Canadian environment from gas leakage.