摘要
油气管道高后果区识别是完整性管理工作的关键环节,是后续风险评价、完整性评价、维修维护等的前提条件。评定过程中受到管道沿线建筑、交通、人口、环境、水文、地形地貌、空间等数据的影响,以这些数据为基础,人为进行打分和评分统计。基于BP神经网络的油气管道高后果区自动识别方法,结合GB32167油气输送管道完整性管理规范中高后果区识别规定,通过建立、训练、测试、验证高后果区识别BP神经网络模型,根据采集数据直接识别出高后果区等级。该方法的应用省去人工评判过程,很大程度上减少了实际高后果区识别评定的工作量。
High consequence area identification of oil and gas pipeline is the key process of integrity management, and the precondition of the following risk assessment, integrity assessment and maintenance. The assessment process is affected by many factors such as the construction, transportation, population, environment, hydrology, topography and spatial data along the pipeline, as it is rated and counted manually with these data. Based on the BP neural network's automatic identification methods for high consequence area of oil and gas pipeline, coupled with GB32167 oil and gas pipeline integrity management specification's high consequence area identification provisions, by establishing, training, testing and verifying the BP neural network high consequence area identification model, the high consequence area can be directly identified and rated with the automatic identification method according to the collected data. So the simplified method can save a lot of human rating and workload in the assessment and identification of high consequence area.
出处
《当代石油石化》
CAS
2019年第2期38-42,共5页
Petroleum & Petrochemical Today
关键词
油气管道
高后果区
BP神经网络
完整性管理
地理信息系统
oil and gas pipeline
high consequence area
BP neural network
integrity management
geographic information system
