[1]高 帅.基于KPCA-GA-ELM的海底管道外腐蚀速率预测技术[J].焊管,2021,44(11):23-27.[doi:10.19291/j.cnki.1001-3938.2021.11.004]
 GAO Shuai.Prediction of Corrosion Rate of Submarine Pipeline based on KPCA-GA-ELM[J].,2021,44(11):23-27.[doi:10.19291/j.cnki.1001-3938.2021.11.004]
点击复制

基于KPCA-GA-ELM的海底管道外腐蚀速率预测技术()
分享到:

《焊管》[ISSN:1001-3938/CN:61-1160/TE]

卷:
第44卷
期数:
2021年第11期
页码:
23-27
栏目:
应用与开发
出版日期:
2021-11-28

文章信息/Info

Title:
Prediction of Corrosion Rate of Submarine Pipeline based on KPCA-GA-ELM
文章编号:
10.19291/j.cnki.1001-3938.2021.11.004
作者:
高 帅
中海石油(中国)有限公司深圳分公司, 广东 深圳 518000
Author(s):
GAO Shuai
CNOOC China Limited, Shenzhen Branch, Shenzhen 518000, Guangdong, China
关键词:
海底管道外腐蚀预测KPCAELMGA
Keywords:
submarine pipeline external corrosion prediction KPCA ELM GA
分类号:
TG174
DOI:
10.19291/j.cnki.1001-3938.2021.11.004
文献标志码:
A
摘要:
为了掌握海底管道外腐蚀情况,保证管道的安全运行,对海洋环境腐蚀因素进行了梳理,确定了8种影响海底管道外腐蚀速率的主要因素,通过核主成分分析(kernel principal components analysis, KPCA)对影响因素进行优选,并将优选后的样本放入极限学习机(extreme learning machine, ELM)进行训练,以影响因素为输入,腐蚀速率为输出,采用改进的遗传算法(genetic algorithm, GA)对ELM的输入权值和隐含层偏差进行优化。结果表明,KPCA-GA-ELM模型最大相对误差2.43%,MSE和MAPE分别为0.148 33和1.15,与其余三种模型相比,预测精度最高。研究结果可为提高海底管道完整性管理水平提供技术支持。
Abstract:
In order to master the corrosion situation of submarine pipeline and ensure the safe operation of pipeline, by sorting out the corrosion factors in the marine environment, eight main factors affecting the corrosion rate of submarine pipelines were determined. The influencing factors were optimized by KPCA, and the optimized samples were put into the extreme learning machine (ELM) for training. The influencing factors were taken as input and the corrosion rate as output. An improved genetic algorithm (GA) was used to optimize the input weight and hidden layer bias of ELM. The results show that the maximum relative error of KPCA-GA-ELM model is 2.43%, MSE and MAPE are 0.148 33 and 1.15, respectively. Compared with the other three models, the prediction accuracy is the highest. The research results can provide technical support for improving the integrity management level of submarine pipeline.

参考文献/References:

[1] 骆正山,袁宏伟. 基于误差补偿的GM-RBF海底管道腐蚀预测模型[J]. 中国安全科学学报,2018,28(3):96-101.[2] 俞树荣,李建华,李淑欣,等. 埋地管道腐蚀剩余寿命预测概率模型[J]. 中国安全科学学报,2008,18(6):11-15.[3] 唐晓时,春涛,曹光,等. 海岸土壤环境对油气管道局部腐蚀的影响[J]. 中国腐蚀与防护学报,2016,36(3):191-196.[4] 张新生,吕品品. 考虑随机效应的腐蚀管道贝叶斯退化分析[J]. 中国安全科学学报,2019,29(8):73-80.[5] 毕傲睿,骆正山,王小完,等. 基于土壤腐蚀主成分的金属管道退化维纳过程研究[J]. 材料保护,2018,51(1):37-42.[6] 王齐,胡林林. 基于BP神经网络的油气长输管道土壤腐蚀性预测[J]. 当代化工,2016,45(9):2198-2200.[7] 者娜,杨剑锋,刘文彬,等. 基于KPCA和SVM的工艺管道腐蚀速率预测[J]. 腐蚀与防护,2019,40(1):56-60.[8] 孙艳丽,杨娜,张正涛,等. 基于核主元分析和支持向量机的结构损伤识别研究[J]. 应用基础与工程科学学报,2018,26(4):888-900.[9] HUANG G B,ZHU Q Y,SIEW C K. Extreme learning machine:Theory and applications[J]. Neurocomputing,2006(70):489-501.[10] 范伟,林瑜阳,李钟慎. 遗传算法优化的BP神经网络压电陶瓷蠕变预测[J]. 电机与控制学报,2018,22(7):91-96.[11] SRINIVAS M, PATNAIK L M. Adaptive probabilities of crossover and mutation in genetic algorithms[J]. IEEE Transactions on Systems,Man,and Cybernetics,1994,24(4):656-667.

相似文献/References:

[1]蔡彬,郑福恩,李记科,等.海底油气管道延性断裂止裂CVN冲击功的计算[J].焊管,2011,34(12):39.[doi:1001-3938(2011)12-0039-04]
 CAI Bin,LI Ji-ke,LI Yun-long.CVN Impact Values Calculation for the Arrest of Ductile Fracture for Subsea Oil and Gas Pipeline[J].,2011,34(11):39.[doi:1001-3938(2011)12-0039-04]
[2]杜 伟,张京兵,魏亚秋,等.大壁厚海底管道用X65钢管性能均匀性研究[J].焊管,2013,36(2):32.[doi:1001-3938(2013)02-0032-03]
 DU Wei,ZHANG Jingbing,WEI Yaqiu,et al.Research on Inhomogeneity of X65 Steel Pipes Used for Submarine Pipeline with Heavy Wall Thickness[J].,2013,36(11):32.[doi:1001-3938(2013)02-0032-03]
[3]王海涛,池 强,李鹤林,等.海底油气输送管道材料开发和应用现状[J].焊管,2014,37(8):25.[doi:1001-3938(2014)08-0025-05]
 WANG Haitao,CHI Qiang,LI Helin,et al.Development and Application Status of Submarine Pipeline Materials for Oil and Gas Transportation[J].,2014,37(11):25.[doi:1001-3938(2014)08-0025-05]
[4]张海波,彭立山,赵志伟,等.内部冷却系统在海底管线用厚壁弯管制造中的应用[J].焊管,2014,37(9):60.[doi:1001-3938(2014)09-0060-04]
 ZHANG Haibo,PENG Lishan,ZHAO Zhiwei,et al.Application of Internal Cooling System in Manufacturing Heavy Thickness Bends for Subsea Pipeline[J].,2014,37(11):60.[doi:1001-3938(2014)09-0060-04]
[5]李全华,张 勇,曲兴刚,等.某海底天然气管线的内腐蚀直接评价[J].焊管,2016,39(8):58.[doi:10.19291/j.cnki.1001-3938.2016.08.014]
 LI Quanhua,ZHANG Yong,QU Xinggang,et al.Internal Corrosion Direct Assessment Method(ICDA) of Some Submarine Gas Pipeline[J].,2016,39(11):58.[doi:10.19291/j.cnki.1001-3938.2016.08.014]
[6]宋恩鹏,蔡 克,瞿婷婷,等.西部某压气站压缩机出口埋地管线外腐蚀原因分析[J].焊管,2019,42(3):59.[doi:10.19291/j.cnki.1001-3938.2019.3.012]
 SONG Enpeng,CAI Ke,QU Tingting,et al.Cause Analysis of Compressor Outlet Buried Pipeline External Corrosion in a Western Compressor Station[J].,2019,42(11):59.[doi:10.19291/j.cnki.1001-3938.2019.3.012]
[7]牛爱军,等.海底管线用管线钢及钢管的研发与应用[J].焊管,2019,42(6):1.[doi:10.19291/j.cnki.1001-3938.2019.6.001]
 NIU Aijun,BI Zongyue,et al.Development and Application of Pipeline Steel and Steel Pipe for Offshore Pipeline in China[J].,2019,42(11):1.[doi:10.19291/j.cnki.1001-3938.2019.6.001]
[8]胡松林.焊管在海底管道中的应用探讨[J].焊管,2019,42(11):1.[doi:10.19291/j.cnki.1001-3938.2019.11.001]
 HU Songlin.Discussion on the Application of Welded Pipe in Offshore Pipeline[J].,2019,42(11):1.[doi:10.19291/j.cnki.1001-3938.2019.11.001]
[9]史小东,谢建桥,张辉宇.海底管道典型缺陷磁记忆检测试验研究[J].焊管,2020,43(7):8.[doi:10.19291/j.cnki.1001-3938.2020.07.002]
 SHI Xiaodong,XIE Jianqiao,ZHANG Huiyu.Submarine Pipeline Typical Defect Detection Study Based on Magnetic Memory Testing[J].,2020,43(11):8.[doi:10.19291/j.cnki.1001-3938.2020.07.002]
[10]张天江,陈 亮,吴 员,等.基于API 1104标准的海底管道对接环焊缝超声波检验程序验证[J].焊管,2021,44(12):58.[doi:10.19291/j.cnki.1001-3938.2021.12.011]
 ZHANG Tianjiang,CHEN Liang,WU Yuan,et al.Verification of Ultrasonic Inspection Program for Butt Girth Welds Based on API 1104 Standard[J].,2021,44(11):58.[doi:10.19291/j.cnki.1001-3938.2021.12.011]

备注/Memo

备注/Memo:
收稿日期:2021-07-05作者简介:高 帅(1992—),男,陕西咸阳人,大学本科,工程师,2014年毕业于中国石油大学(华东)油气储运工程专业,现主要从事海洋油气开采及处理,海底管道运输、防腐及安全管理工作。
更新日期/Last Update: 2021-12-09