[1]赵年峰,夏艳权,崔凯强,等.基于优化GM(1,N)模型的油气管道腐蚀速率预测[J].焊管,2023,46(8):38-44.[doi:10.19291/j.cnki.1001-3938.2023.08.006]
 ZHAO Nianfeng,XIA Yanquan,CUI Kaiqiang,et al.Corrosion Rate Prediction of Oil and Gas Pipeline based on Optimized GM (1, N) Model[J].,2023,46(8):38-44.[doi:10.19291/j.cnki.1001-3938.2023.08.006]
点击复制

基于优化GM(1,N)模型的油气管道腐蚀速率预测()
分享到:

《焊管》[ISSN:1001-3938/CN:61-1160/TE]

卷:
46
期数:
2023年第8期
页码:
38-44
栏目:
应用与开发
出版日期:
2023-08-19

文章信息/Info

Title:
Corrosion Rate Prediction of Oil and Gas Pipeline based on Optimized GM (1, N) Model
文章编号:
10.19291/j.cnki.1001-3938.2023.08.006
作者:
赵年峰夏艳权崔凯强孔淑颖解静梁昌晶
1.中国石油集团渤海石油装备制造有限公司石油机械厂,河北 任丘 062552;
2.河北华北石油工程建设有限公司,河北 任丘 062552;
3.国家管网集团西南管道有限责任公司 南宁输油气分公司,南宁 530200;
4.中国石油长庆油田分公司伴生气综合利用项目部,西安 710016
Author(s):
ZHAO NianfengXIA YanquanCUI KaiqiangKONG ShuyingXIE JingLIANG Changjing
1. Petroleum Machinery Factory, CNPC Bohai Petroleum Equipment Manufacturing Co., Ltd., Renqiu 062552, Hebei,China;
2. Hebei Huabei Petroleum Engineering Construction Co., Ltd., Renqiu 062552, Hebei, China;
3. Nanning Oil andGas Transportation Branch of State Pipe Network Group Southwest Pipeline Co., Ltd., Nanning 530200, China;
4. AssociatedGas Comprehensive Utilization Project Department of PetroChina Changqing Oilfield Company, Xi’an 710016, China
关键词:
GM(1N)模型油气管道腐蚀速率背景系数
Keywords:
GM (1 N) oil and gas pipeline corrosion rate background coefficient
分类号:
TG174
DOI:
10.19291/j.cnki.1001-3938.2023.08.006
文献标志码:
B
摘要:
为了提高油气管道腐蚀速率的预测精度,解决样本缺失情况下管道剩余寿命的预测问题,通过斯皮尔曼相关系数和随机森林算法寻找腐蚀因素中相关性较高的变量,去除相关性较高但重要性较低的变量,采用人工蜂群算法(ABC)对GM(1,N)模型的背景值进行动态优化,形成优化GM(1,N)模型,并对不同模型的预测结果进行了对比。结果表明,腐蚀因素中一些变量存在较强的相关性,筛选出重要性较高且相关性较小的5个变量,分别为土壤电阻率、含盐量、氧化还原电位、含水量和硫化物质量分数;与GM(1,10) 和GM(1,6)模型相比,优化的GM(1,6)模型在训练阶段和预测阶段的平均相对误差均大幅降低,分别为1.52%和2.03%,预测结果更接近实际值,说明了降低因素冗余性和对背景值进行优化的必要性。该优化模型可为油气管道腐蚀速率的准确预测提供实际参考和借鉴。
Abstract:
To improve the prediction accuracy of the corrosion rate of the oil and gas pipelines and solve the problem ofpredicting the remaining life of pipelines in the case of the missing samples. By spearman correlation coefficient and highcorrelation of random forest algorithm for corrosion factor variables, the correlation between high importances but lowervariables was removed. Using artificial colony algorithm (ABC) to the background value of GM (1, N) model for dynamicoptimization, optimized GM (1, N) model was formed, and the prediction results of different models were compared. Theresults showed that some variables of corrosion factors had strong correlation, and five variables with high importance andlow correlation were screened out, soil resistivity, salt content, oxidation reduction potential, water content and sulfidecontent. Compared with GM (1, 10) and GM (1, 6) models, the average relative errors of the optimized GM (1, 6) model inthe training and prediction stages are significantly reduced, which are 1.52% and 2.03%, respectively. The prediction resultsare closer to the actual values, which indicate the necessity of reducing the redundancy of factors and optimizing the background values. This optimization model can provide the practical reference for the accurate prediction of corrosion rate of the oil and gas pipelines.

参考文献/References:

[1]吴志平,陈振华,戴联双,等.油气管道腐蚀检测技术发展现状与思考[J].油气储运,2020,39(8):851-860.[2]吴明,谢飞,陈旭,等.埋地油气管道腐蚀失效研究进展及思考[J].油气储运,2022,41(6):712-722.[3]蔡利华,乐明,杨鹏,等.海上油气田CO2腐蚀预测模型研究[J].当代化工,2016,45(10):2356-2359.[4]李丽,李晓刚,邢士波,等.BP人工神经网络对国内典型地区碳钢土壤腐蚀的预测研究[J].腐蚀科学与防护技术,2013,25(5):372-376.[5]李昊燃,郑度奎,程远鹏,等.优化的灰色马尔科夫模型在油气集输管道腐蚀预测中的应用[J]. 腐蚀与防护,2021,42(5):42-46,58.[6]王庆锋,雷兴国.基于无偏灰色马尔可夫链组合模型的管线腐蚀剩余寿命预测[J]. 腐蚀与防护,2018,39(2):136-140,147.[7]张新生,赵梦旭,王小完.尾段残差修正GM(1,1)模型在管道腐蚀预测中的应用[J].中国安全科学学报,2017,27(1):65-70.[8]孙忠秋,高金萍,吴发云,等.基于机载激光雷达点云和随机森林算法的森林蓄积量估测[J]. 林业科学,2021,57(8):68-81.[9]杨笑源,赵晓刚,郭湛,等.基于GM(1,N)模型的管输油流静电数值预测[J].后勤工程学院学报,2016,32(2):87-90.[10]杨烨,肖传桃,刘彬.基于灰色系统GM(1,N)模型的脱碳装置效果分析[J].石油与天然气化工,2016,45(1):30-35.[11]叶廷宇,叶军,王晖,等.结合人工蜂群优化的粗糙K-means聚类算法[J].计算机科学与探索,2022,16(8):1923-1932.[12]南楠,严英占.基于改进人工蜂群算法的大数据优化[J].西南师范大学学报(自然科学版),2021,46(3):20-26.

备注/Memo

备注/Memo:
收稿日期:2022-10-08
作者简介:赵年峰(1984—),男,河北任丘人,本科,工程师,现从事油田采油装备、油气储运设备以及油田污水处理回注设备的设计与研究工作。
更新日期/Last Update: 2023-08-24