基于神经网络算法的油田油气管道泄漏检测
|
摘要
长庆油田油气输送管道发展迅速,油气管道里程也越来越长。部分油气管道年代较长,管道腐蚀情况较严重,发生油气管道泄漏的危险系数很高。而油气管道一旦发生泄漏,将造成严重的事故,危害巨大。神经网络算法是目前最有效的分类识算法之一。采集长庆油田油气管道的正常压力波信号和管道泄漏时的压力波信号,并对信号进行预处理,分别对正常的压力波信号和泄漏时压力波信号进行小波变换分析,引入神经网络算法分析结果显示,此算法对长庆油田原油、天然气的传输管道泄漏问题,均具有有效的检测能力。当油气管道内的压力差变化达到5%以上时,油气管道泄漏的检测、预报精度接近100%。当油气管道内的压力差变化低于5%,需要根据实际情况分析,检测精度跟实验数据的样本数量密切相关。
Changqing oilfield oil and gas pipelines in the rapid development of oil and gas pipeline mileage is getting longer.Part of the oil and gas pipelines in the age of a long time,the pipeline corrosion is more serious,the risk of oil and gas pipeline leakage is very high.The oil and gas pipelines once the leak,will cause serious accidents,great harm.Neural network algorithm is one of the most effective algorithms.Acquisition of Changqing oilfield oil and gas pipelines of normal pressure wave signal and pipeline leakage of pressure wave signals,and the signals were pre processed,respectively to normal pressure wave signal and leakage when pressure wave signals were wavelet transform analysis,using neural network algorithm analysis results show,the algorithm of Changqing Oilfield crude oil and natural gas transmission pipeline leakage problem,is effective in detecting ability.When the pressure difference in the oil and gas pipeline reaches 5% or more,the detection accuracy of the oil and gas pipeline leakage is close to 100%.When the pressure difference in the oil and gas pipeline is less than 5%,according to the actual situation analysis,the detection accuracy is closely related to the number of samples of the experimental data.
Changqing oilfield oil and gas pipelines in the rapid development of oil and gas pipeline mileage is getting longer.Part of the oil and gas pipelines in the age of a long time,the pipeline corrosion is more serious,the risk of oil and gas pipeline leakage is very high.The oil and gas pipelines once the leak,will cause serious accidents,great harm.Neural network algorithm is one of the most effective algorithms.Acquisition of Changqing oilfield oil and gas pipelines of normal pressure wave signal and pipeline leakage of pressure wave signals,and the signals were pre processed,respectively to normal pressure wave signal and leakage when pressure wave signals were wavelet transform analysis,using neural network algorithm analysis results show,the algorithm of Changqing Oilfield crude oil and natural gas transmission pipeline leakage problem,is effective in detecting ability.When the pressure difference in the oil and gas pipeline reaches 5% or more,the detection accuracy of the oil and gas pipeline leakage is close to 100%.When the pressure difference in the oil and gas pipeline is less than 5%,according to the actual situation analysis,the detection accuracy is closely related to the number of samples of the experimental data.
引文
[1]贾宗贤.基于小波变换的天然气管道次声波信号分析[J].油气田地面工程,2011,30(11):13-15.
[2]贾莹.基于智能算法的油田天然气管道泄漏检测研究[D].大庆:东北石油大学,2014.
[3]丁冬,杨成禹.基于遗传算法优化的神经网络在长输管道泄漏检测系统中的应用[J].长春理工大学学报(自然科学版),2015(6):136-139.
[4]高丙坤,任秀菊.基于改进差分进化算法的神经网络管道泄漏检测[J].化工自动化及仪表,2014(1):14-18.
[5]贾莹,高丙坤.基于遗传算法优化神经网络的管道泄漏检测方法研究[J].化工自动化及仪表,2014(2):128-131,137.
[6]曾海龙.基于缺陷相关的腐蚀管道可靠性评价[D].兰州:兰州理工大学,2010.
[7]宿兰花.基于模糊理论的管道失效模式诊断模型与方法研究[D].北京:中国石油大学,2010.
[8]崔立元.关于吴起—延炼原油管道泄漏检测与定位系统的研究[D].西安:西安石油大学,2014.
[9]王正.基于物联网的管道远程泄漏检测系统设计[D].成都:西南石油大学,2015.
[10]王天瑜,吴宗之,王如君,等.基于改进灰关联分析法的埋地管道土壤腐蚀性评价[J].中国安全生产科学技术,2016(3):133-136.
[11]王少锋,仲济祥,王建国.立体管网微泄漏实时监测系统研究概述[J].河北科技大学学报,2016(2):130-138.
[12]周卫华,谢耀恒,吕玉宏,等.电气设备带电检测述评[J].湖南电力,2016(2):1-8.
[13]孙博,聂文.管道内检测技术在输气管道中的研究与应用[J].石油化工自动化,2016(2):58-61.
[14]薛斌.光缆在线检测技术在油气田行业中的应用[J].信息与电脑(理论版),2016(6):188-189.
[15]张媛,顾为敏.输气站场管道完整性检测评价技术体系[J].管道技术与设备,2016(2):47-50,57.
[2]贾莹.基于智能算法的油田天然气管道泄漏检测研究[D].大庆:东北石油大学,2014.
[3]丁冬,杨成禹.基于遗传算法优化的神经网络在长输管道泄漏检测系统中的应用[J].长春理工大学学报(自然科学版),2015(6):136-139.
[4]高丙坤,任秀菊.基于改进差分进化算法的神经网络管道泄漏检测[J].化工自动化及仪表,2014(1):14-18.
[5]贾莹,高丙坤.基于遗传算法优化神经网络的管道泄漏检测方法研究[J].化工自动化及仪表,2014(2):128-131,137.
[6]曾海龙.基于缺陷相关的腐蚀管道可靠性评价[D].兰州:兰州理工大学,2010.
[7]宿兰花.基于模糊理论的管道失效模式诊断模型与方法研究[D].北京:中国石油大学,2010.
[8]崔立元.关于吴起—延炼原油管道泄漏检测与定位系统的研究[D].西安:西安石油大学,2014.
[9]王正.基于物联网的管道远程泄漏检测系统设计[D].成都:西南石油大学,2015.
[10]王天瑜,吴宗之,王如君,等.基于改进灰关联分析法的埋地管道土壤腐蚀性评价[J].中国安全生产科学技术,2016(3):133-136.
[11]王少锋,仲济祥,王建国.立体管网微泄漏实时监测系统研究概述[J].河北科技大学学报,2016(2):130-138.
[12]周卫华,谢耀恒,吕玉宏,等.电气设备带电检测述评[J].湖南电力,2016(2):1-8.
[13]孙博,聂文.管道内检测技术在输气管道中的研究与应用[J].石油化工自动化,2016(2):58-61.
[14]薛斌.光缆在线检测技术在油气田行业中的应用[J].信息与电脑(理论版),2016(6):188-189.
[15]张媛,顾为敏.输气站场管道完整性检测评价技术体系[J].管道技术与设备,2016(2):47-50,57.