Prediction of Internal Corrosion Rate of Gas Field Gathering Pipelines Based on GRA-IFA-LSSVM Model
摘 要
针对气田集输管道的腐蚀问题,提出了一种基于GRA-IFA-LSSVM组合模型的内腐蚀速率预测算法。对GRA (灰色关联分析)模型、IFA (改进萤火虫)模型以及LSSVM (最小二乘支持向量机)模型理论进行了介绍,提出了组合模型的组合流程以及组合模型的评价指标;以我国某气田集输管道为例,对GRA-IFA-LSSVM组合模型的预测精度进行验证,同时,将其预测精度与其他常见预测模型的精度进行了对比。结果表明:温度、H2S含量、CO2含量、pH以及流速属于影响气田集输管道腐蚀的重要因素;使用GRA-IFA-LSSVM组合模型对气田集输管道内腐蚀速率进行预测时,其平均绝对误差为1.946%,均方根误差为1.496%,可决系数为97.53%,该组合模型的三项评价指标均小于其他常见预测模型。GRA-IFA-LSSVM组合模型对气田集输管道进行内腐蚀速率预测具有很强的准确性、鲁棒性及先进性,可以为气田集输管道的保护提供数据支持。
Abstract
Aiming at the corrosion problem of gas field gathering pipelines, a prediction algorithm of internal corrosion rate based on the GRA-IFA-LSSVM combined model was proposed. The theories of GRA (Gray Relational Analysis) model, IFA (Improved Firefly) model and LSSVM (Least Squares Support Vector Machine) model were introduced, and the combined process and evaluation indexes of the combined model were proposed. The prediction accuracy of the GRA-IFA-LSSVM combined model was verified, taking a domestic pipeline as an example, and was compared with those of the other common prediction models. The results show that temperature, H2S content, CO2 content, pH value and flow rate were important factors affecting the corrosion of gas field gathering pipelines. When the GRA-IFA-LSSVM combined model was used to predict the internal corrosion rate of gas field gathering pipelines, the average absolute error was 1.946%, the root mean square error was 1.496%, and the absolute coefficient was 97.53%. The three evaluation indexes of the combined model were all smaller than those of the other common prediction models. The GRA-IFA-LSSVM combined model had strong accuracy, robustness and advancement in the prediction of internal corrosion rate of gas field gathering pipelines, and could provide data support for the protection of gas field gathering pipelines.
中图分类号 TG172 DOI 10.11973/fsyfh-202208017
所属栏目 数值模拟
基金项目 陕西省科技统筹创新工程计划项目(2016KTZDGY07-05);西安石油大学研究生创新与实践能力培养计划(YCS21213161)
收稿日期 2021/12/6
修改稿日期
网络出版日期
作者单位点击查看
联系人作者王寿喜(swang@xsyu.com)
引用该论文: ZHOU Yang,WANG Shouxi. Prediction of Internal Corrosion Rate of Gas Field Gathering Pipelines Based on GRA-IFA-LSSVM Model[J]. Corrosion & Protection, 2022, 43(8): 86
共有人对该论文发表了看法,其中:
人认为该论文很差
人认为该论文较差
人认为该论文一般
人认为该论文较好
人认为该论文很好
参考文献
【1】刘彦麟,彭星煜,姚东池,等.考虑失效相关性的管道腐蚀故障树新算法 .油气储运,2019,38(1):31-39.
【2】顾锡奎,何鹏,陈思锭.高含硫长距离集输管道腐蚀监测技术研究 .石油与天然气化工,2019,48(1):68-73.
【3】张新生,叶晓艳.不同初始条件的UGM(1,1)管道腐蚀预测建模研究 .中国安全科学学报,2019,29(3):63-69.
【4】王文辉,骆正山,张新生.基于PSO-GRNN模型的埋地管道腐蚀剩余寿命预测 .表面技术,2019,48(10):267-275,284.
【5】单克,帅健,杨光,等.美国油气管道基本失效概率评估方法及启示 .油气储运,2020,39(5):530-535.
【6】曲志豪,唐德志,胡丽华,等.基于优化随机森林的H2S腐蚀产物类型及腐蚀速率预测 .表面技术,2020,49(3):42-49.
【7】骆正山,宋莹莹,王小完,等.凝析气田集输管线腐蚀预测研究 .中国安全科学学报,2019,29(11):135-140.
【8】RAJU J,BHALLA S,VISALAKSHI T.Pipeline corrosion assessment using piezo-sensors in reusable non-bonded configuration .NDT & E International,2020,111:102220.
【9】曾维国,李曙华,李岩,等.基于径向基函数神经网络预测模型评价油气水集输管道的均匀腐蚀缺陷 .腐蚀与防护,2020,41(10):50-56.
【10】梁金禄.高含硫气田集输管线内腐蚀预测研究 .粘接,2020,41(2):138-141.
【11】SOEPANGKAT B O P,NORCAHYO R,RUPAJATI P,et al.Multi-objective optimization in wire-EDM process using grey relational analysis method (GRA) and backpropagation neural network-genetic algorithm (BPNN-GA) methods .Multidiscipline Modeling in Materials and Structures,2019,15(5):1016-1034.
【12】RAJMOHAN T,PALANIKUMAR K,PRAKASH S.Grey-fuzzy algorithm to optimise machining parameters in drilling of hybrid metal matrix composites .Composites Part B:Engineering,2013,50:297-308.
【13】陆克中,孙俊.全局信息共享的自适应FA算法 .计算机工程与科学,2016,38(6):1164-1170.
【14】张兢,曾建梅,李冠迪,等.FA-SVM优化算法在情感识别中的应用 .激光杂志,2016,37(9):76-79.
【15】DEMESTICHAS P,GEORGANTAS N,TZIFA E,et al.Computationally efficient algorithms for location area planning in future cellular systems .Computer Communications,2000,23(13):1263-1280.
【16】CAI Y P,XU H,SUN X M,et al.Duple-EDA and sample density balancing .Science in China Series F:Information Sciences,2009,52(9):1640-1650.
【17】DHIFAOUI Z,KORTAS H,BENAMMOU S.Correlation dimension of fractional Gaussian noise:new evidence from wavelets .International Journal of Bifurcation and Chaos,2014,24(4):1450041.
【18】谈发明,王琪.CPSO-LSSVM算法在车载电池SOC预测中的应用 .实验室研究与探索,2018,37(8):110-114.
【19】徐南,马符讯,贾东振.智能优化LSSVM算法的混沌时间序列边坡变形预测模型 .测绘与空间地理信息,2015,38(2):9-11,17.
【20】李春祥,迟恩楠,何亮,等.基于时变ARMA和EMD-PSO-LSSVM算法的非平稳下击暴流风速预测 .振动与冲击,2016,35(17):33-38,51.
【21】MESBAH M,SOROUSH E,AZARI V,et al.Vapor liquid equilibrium prediction of carbon dioxide and hydrocarbon systems using LSSVM algorithm .The Journal of Supercritical Fluids,2015,97:256-267.
【22】SHAYEGHI H,GHASEMI A,MORADZADEH M,et al.Day-ahead electricity price forecasting using WPT,GMI and modified LSSVM-based S-OLABC algorithm .Soft Computing,2017,21(2):525-541.
【2】顾锡奎,何鹏,陈思锭.高含硫长距离集输管道腐蚀监测技术研究 .石油与天然气化工,2019,48(1):68-73.
【3】张新生,叶晓艳.不同初始条件的UGM(1,1)管道腐蚀预测建模研究 .中国安全科学学报,2019,29(3):63-69.
【4】王文辉,骆正山,张新生.基于PSO-GRNN模型的埋地管道腐蚀剩余寿命预测 .表面技术,2019,48(10):267-275,284.
【5】单克,帅健,杨光,等.美国油气管道基本失效概率评估方法及启示 .油气储运,2020,39(5):530-535.
【6】曲志豪,唐德志,胡丽华,等.基于优化随机森林的H2S腐蚀产物类型及腐蚀速率预测 .表面技术,2020,49(3):42-49.
【7】骆正山,宋莹莹,王小完,等.凝析气田集输管线腐蚀预测研究 .中国安全科学学报,2019,29(11):135-140.
【8】RAJU J,BHALLA S,VISALAKSHI T.Pipeline corrosion assessment using piezo-sensors in reusable non-bonded configuration .NDT & E International,2020,111:102220.
【9】曾维国,李曙华,李岩,等.基于径向基函数神经网络预测模型评价油气水集输管道的均匀腐蚀缺陷 .腐蚀与防护,2020,41(10):50-56.
【10】梁金禄.高含硫气田集输管线内腐蚀预测研究 .粘接,2020,41(2):138-141.
【11】SOEPANGKAT B O P,NORCAHYO R,RUPAJATI P,et al.Multi-objective optimization in wire-EDM process using grey relational analysis method (GRA) and backpropagation neural network-genetic algorithm (BPNN-GA) methods .Multidiscipline Modeling in Materials and Structures,2019,15(5):1016-1034.
【12】RAJMOHAN T,PALANIKUMAR K,PRAKASH S.Grey-fuzzy algorithm to optimise machining parameters in drilling of hybrid metal matrix composites .Composites Part B:Engineering,2013,50:297-308.
【13】陆克中,孙俊.全局信息共享的自适应FA算法 .计算机工程与科学,2016,38(6):1164-1170.
【14】张兢,曾建梅,李冠迪,等.FA-SVM优化算法在情感识别中的应用 .激光杂志,2016,37(9):76-79.
【15】DEMESTICHAS P,GEORGANTAS N,TZIFA E,et al.Computationally efficient algorithms for location area planning in future cellular systems .Computer Communications,2000,23(13):1263-1280.
【16】CAI Y P,XU H,SUN X M,et al.Duple-EDA and sample density balancing .Science in China Series F:Information Sciences,2009,52(9):1640-1650.
【17】DHIFAOUI Z,KORTAS H,BENAMMOU S.Correlation dimension of fractional Gaussian noise:new evidence from wavelets .International Journal of Bifurcation and Chaos,2014,24(4):1450041.
【18】谈发明,王琪.CPSO-LSSVM算法在车载电池SOC预测中的应用 .实验室研究与探索,2018,37(8):110-114.
【19】徐南,马符讯,贾东振.智能优化LSSVM算法的混沌时间序列边坡变形预测模型 .测绘与空间地理信息,2015,38(2):9-11,17.
【20】李春祥,迟恩楠,何亮,等.基于时变ARMA和EMD-PSO-LSSVM算法的非平稳下击暴流风速预测 .振动与冲击,2016,35(17):33-38,51.
【21】MESBAH M,SOROUSH E,AZARI V,et al.Vapor liquid equilibrium prediction of carbon dioxide and hydrocarbon systems using LSSVM algorithm .The Journal of Supercritical Fluids,2015,97:256-267.
【22】SHAYEGHI H,GHASEMI A,MORADZADEH M,et al.Day-ahead electricity price forecasting using WPT,GMI and modified LSSVM-based S-OLABC algorithm .Soft Computing,2017,21(2):525-541.
相关信息