摘要
针对当今海底天然气输送管道泄漏事件频频发生的现象,以多波束测深声纳的水体成像数据作为数据来源,设计了一种针对海底天然气管道的泄漏检测方法。该方法主要结合旁瓣抑制和海底底部检测以提高气体泄漏检测的稳定性和可靠性,通过水体成像数据的预处理、形态学处理、最大熵分割及连通区域目标标记等处理流程,得到泄露气体的疑似目标区域。水体成像数据采用真实水池测试数据,通过检测结果可证明该方法能够有效提取泄漏气体的疑似区域,并为之后水体目标检测、识别奠定基础,具有较好的工程应用前景。
Nowadays,the gas leaking events in subsea gas pipelines are frequently happened.In this paper,a detection algorithm of subsea gas pipeline leakage is proposed,using the multi-beam sonar to get water column imaging data as a data source.The algorithm mainly uses the sidelobe suppression and bottom detection to improve the stability and reliability of the task.The main processing includes pre-processing of water column data,morphological processing,maximum entropy segmentation and the connected area labeling.Then we can confirm the suspected area of the leaking gas.All the water column imaging data is real data of simulating the leaking gas in the pool.The results verify that the algorithm can effectively extract the suspected area of leakage gas and it has a good prospect in engineering application.
作者
张志刚
郭俊
杨嘉斌
蒲定
张宏伟
ZHANG Zhigang;GUO Jun;YANG Jiabin;PU Ding;ZHANG Hongwei(Acoustic Science and Technology Laboratory,Harbin Engineering University,Harbin 150001,China;Deep Water Engineering Construction Center,CNOOC Deep Development Co.,Ltd.,Shenzhen 518000,China;College of Mechanical Engineering,Tianjin University,Tianjin 300000,China)
出处
《应用科技》
CAS
2018年第6期12-16,共5页
Applied Science and Technology
基金
国家科技重大专项项目(2016ZX05057005)
关键词
气体泄漏检测
水体成像
测深声纳
旁瓣抑制
底部检测
海底管线
形态学
最大熵
gas leaking inspection
water column imaging
multi-beam echosounder
sidelobe suppression
bottom detection
subsea pipeline
morphology
maximum entropy
