PDF下载

[1]宋福霖,赵弘,苗兴园.基于HMOGWO-RF的埋地管道点蚀深度机理-学习预测模型[J].油气储运,2024,43(11):1249-1259.[doi:10.6047/j.issn.1000-8241.2024.11.006]
　SONG Fulin,ZHAO Hong,MIAO Xingyuan.Mechanism-learning prediction model for pitting depth of buried pipeline based on HMOGWO-RF[J].Oil & Gas Storage and Transportation,2024,43(11):1249-1259.[doi:10.6047/j.issn.1000-8241.2024.11.006]

点击复制

基于HMOGWO-RF的埋地管道点蚀深度机理-学习预测模型

《油气储运》[ISSN:1000-8241/CN:13-1093/TE] 卷: 43 期数: 2024年11期页码: 1249-1259 栏目: 检测与完整性出版日期: 2024-11-25

Title:: Mechanism-learning prediction model for pitting depth of buried pipeline based on HMOGWO-RF

文章编号:: 1000-8241（2024）11-1249-11

作者:: 宋福霖; 赵弘; 苗兴园; 中国石油大学（北京）机械与储运工程学院

Author(s):: SONG Fulin; ZHAO Hong; MIAO Xingyuan; College of Mechanical and Transportation Engineering, China University of Petroleum (Beijing)

关键词:: 埋地管道; 点蚀深度; 腐蚀机理引导; 随机森林算法; 多目标灰狼优化算法; 特征重要性

Keywords:: buried pipeline; pitting depth; guided by corrosion mechanism; Random Forest (RF) algorithm; Hybrid Multi-Objective Grey Wolf Optimization (HMOGWO) algorithm; feature importance

分类号:: TE832

DOI:: 10.6047/j.issn.1000-8241.2024.11.006

文献标志码:: A

摘要:: 【目的】截至2025年，中国油气管网规模将达到24×104km，管道输送已然成为中国油气运输的重要方式之一。受管道周围土壤环境等因素影响，管道腐蚀现象时有发生，降低了管道的使用年限。为保障埋地管道的安全运行，需有效预测其所受腐蚀程度。【方法】将随机森林（Random Forest,RF）算法与多目标优化方法相结合，提出腐蚀机理引导下的埋地管道点蚀深度预测模型，将管道腐蚀的腐蚀机理知识引入机器学习模型中，提高模型的可解释性。根据特征变量之间的交互作用机制，构建新的特征变量，以更好地反映管道周围土壤环境的影响因素。通过随机森林算法中的基尼系数计算新特征空间中所有特征的重要性，利用混合多目标灰狼优化（Hybrid Multi-Objective Grey Wolf Optimization,HMOGWO）算法求解随机森林算法的最优超参数，并将特征选择融入多目标优化中。在多目标优化的过程中，综合考虑特征数量、预测准确率、模型稳定性3个优化目标，并设计综合评价指标，对比分析Pareto解集，以获取特征子集与最优超参数组合，得到最具代表性、优化性能最佳的特征子集，提高模型稳定性与预测准确性。【结果】模型设计完成后，采用实际埋地管道的点蚀数据集对模型进行验证，将三目标HMOGWO算法与RF模型相结合，模型的预测性能及稳定性远超三目标灰狼优化算法、双目标HMOGWO算法、双目标灰狼优化算法、单目标灰狼优化算法及单目标粒子群优化算法。【结论】该模型可以实现埋地管道最大点蚀深度的准确预测，所提出的腐蚀机理引导下的埋地管道点蚀深度预测模型可以提高管道腐蚀预测的可解释性与准确性，有助于延长管道的使用寿命，对于石油和天然气运输行业具有重要的实际应用意义。（图12，表2，参22）

Abstract:: [Objective] China’s oil and gas pipeline networks are expected to reach 24×104 km by 2025. Pipeline transportation has become one of the key means of transportation in the country. However, these pipelines are vulnerable to corrosion caused by the surrounding soil environment and other factors, which shortens their life in service. To ensure the safe operation of buried pipelines, accurately predicting the degree of corrosion is crucial. [Methods] This paper presents a prediction model for the pitting depth of buried pipelines, guided by the corrosion mechanism and combining a Random Forest (RF) algorithm with a Multi-Objective Optimization process. The incorporation of knowledge about the pipeline corrosion mechanism enhances the interpretability of the machine-learning (ML) model. By building on the interaction mechanisms among characteristic variables, new variables were created to better reflect the influencing factors of the surrounding soil environment. The Gini coefficients in the Random Forest algorithm were used to evaluate the importance of all features in the new characteristic space through calculations. Additionally, a Hybrid Multi-Objective Grey Wolf Optimization (HMOGWO) algorithm was adopted to determine the optimal hyperparameters of the RF algorithm. This feature selection approach was integrated with the multi-objective optimization process, considering three comprehensive optimization objectives: the number of features, prediction accuracy, and model stability. Using a defined comprehensive evaluation index, a comparative analysis of the Pareto solution set was conducted to obtain the optimal combination of feature subsets and hyperparameters. The resulting feature subsets, which are both representative and optimized for performance, contribute to improvements in model stability and prediction accuracy. [Results] The designed model was validated using a pitting dataset of real-world buried pipelines. By leveraging the combination of the three-objective HMOGWO algorithm and the RF model, it significantly outperformed the three-objective MOGWO algorithm, the two-objective HMOGWO algorithm, the two-objective MOGWO algorithm, as well as both the single-objective GWO algorithm and the single-objective PSO algorithm in terms of prediction performance and stability. [Conclusion] The proposed model has proven effective in accurately predicting the maximum pitting depth of buried pipelines. It is more interpretable and accurate in pipeline corrosion prediction, guided by the corrosion mechanism. This model is shown to be valuable in prolonging the service life of pipelines, highlighting its significance for practical applications in the oil and gas transportation sector. (12 Figures, 2 Tables, 22 References)

参考文献/References:

[1] 李秋扬，赵明华，任学军，王乐乐，冯学书，牛亚琨.中国油气管道建设现状及发展趋势[J].油气田地面工程，2019，38（增刊1）：14-17. 10.3969/j.issn.1006-6896.2019.z1.004. LI Q Y，ZHAO M H，REN X J，WANG L L，FENG X S，NIU Y K. Construction status and development trend of Chinese oil &gas pipeline[J]. Oil-Gas Field Surface Engineering，2019，38(S1):14-17.
[2] 吴志平，陈振华，戴联双，胡亚博，毕武喜.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. 10.6047/j.issn.1000-8241.2020.08.002. WU Z P，CHEN Z H，DAI L S，HU Y B，BI W X. Development status and thinking of oil and gas pipeline corrosion detection technology[J]. Oil & Gas Storage and Transportation，2020，39(8): 851-860.
[3] 骆正山，徐龙寅，骆济豪，王小完.改进SSA-LSSVM模型在埋地管道点蚀深度预测中的应用[J].安全与环境学报，2023，23（9）：3115-3122. 10.13637/j.issn.1009-6094.2022.0395. LUO Z S，XU L Y，LUO J H，WANG X W. Application of improved SSA-LSSVM model in prediction of pitting depth of buried pipelines[J]. Journal of Safety and Environment，2023，23(9): 3115-3122.
[4] SIM S，COLE I S，CHOI Y S，BIRBILIS N. A review of the protection strategies against internal corrosion for the safe transport of supercritical CO2 via steel pipelines for CCS purposes[J]. International Journal of Greenhouse Gas Control，2014，29: 185-199. DOI: 10.1016/j.ijggc.2014.08.010.
[5] CALEYO F，VEL?ZQUEZ J C，VALOR A，HALLEN J M. Probability distribution of pitting corrosion depth and rate in underground pipelines: a Monte Carlo study[J]. Corrosion Science，2009，51(9): 1925-1934. DOI: 10.1016/j.corsci. 2009.05.019.
[6] 史航.含点蚀埋地长输管道悬空沉降失效评估方法研究[D].大庆：东北石油大学，2023. SHI H. Study on failure evaluation method of suspended settlement of long-distance pipeline with pitting corrosion[D]. Daqing: Northeast Petroleum University，2023.
[7] LI X H，GUO M M，ZHANG R R，CHEN G M. A data-driven prediction model for maximum pitting corrosion depth of subsea oil pipelines using SSA-LSTM approach[J]. Ocean Engineering，2022，261: 112062. DOI: 10.1016/j.oceaneng.2022.112062.
[8] 吕林林，王杰，祁庆芳，郭策，贺蓉蓉，孙小伟.基于KPCA-IGOA-ELM的油气混输管道腐蚀速率预测模型[J].油气储运，2023，42（7）：785-792. 10.6047/j.issn.1000-8241.2023.07.007. LYU L L，WANG J，QI Q F，GUO C，HE R R，SUN X W. Corrosion rate prediction model of oil-gas mixed transportation pipelines based on KPCA-IGOA-ELM[J]. Oil & Gas Storage and Transportation，2023，42(7): 785-792.
[9] PENG S B，ZHANG Z，LIU E B，LIU W，QIAO W B. A new hybrid algorithm model for prediction of internal corrosion rate of multiphase pipeline[J]. Journal of Natural Gas Science and Engineering，2021，85: 103716.
[10] 闻亚星，吕坦，国滨，王锋，陈金忠，马义来.基于CART-熵权法的管道腐蚀状态评估及其应用[J].腐蚀与防护，2023，44（9）：16-21，100. 10.11973/fsyfh-202309003. WEN Y X，LYU T，GUO B，WANG F，CHEN J Z，MA Y L. Evaluation and its application of pipeline corrosion status based on CART-entropy weight method[J]. Corrosion and Protection，2023，44(9): 16-21，100.
[11] 吴明，谢飞，陈旭，王丹，孙东旭.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. 10.6047/j.issn.1000-8241.2022.06.013. WU M，XIE F，CHEN X，WANG D，SUN D X. Research progress and thinking on corrosion failure of buried oil and gas pipelines[J]. Oil & Gas Storage and Transportation，2022，41(6): 712-722.
[12] 陈一诺，董绍华.基于随机森林算法的管道缺陷预测方法[J].油气储运，2018，37（9）：975-979. 10.6047/j.issn.1000-8241. 2018.09.003. CHEN Y N，DONG S H. Pipeline defect prediction method based on Random Forests algorithm[J]. Oil & Gas Storage and Transportation，2018，37(9): 975-979.
[13] MIRJALILI S，SAREMI S，MIRJALILI S M，DOS S. COELHO L. Multi-objective grey wolf optimizer: a novel algorithm for multi-criterion optimization[J]. Expert Systems with Applications，2016，47: 106-119. DOI: 10.1016/j.eswa.2015.10.039.
[14] 黄文聪，张凤顺，胡滢，余文锦，常雨芳.基于IMOGWO算法的无线电能传输系统参数优化[J/OL].控制工程：1-9[2023-10-20]. https://doi.org/10.14107/j.cnki.kzgc.20220596. 10.14107/j.cnki.kzgc.20220596. HUANG W C，ZHANG F S，HU Y，YU W J，CHANG Y F. Parameter optimization of wireless power transfer system based on IMOGWO algorithm[J/OL]. Control Engineering of China: 1-9[2023-10-20]. https://doi.org/10.14107/j.cnki.kzgc.20220596.
[15] AL-KAYSSI A W. Use of water retention data and soil physical quality index S to quantify hard-setting and degree of soil compactness indices of gypsiferous soils[J]. Soil and Tillage Research，2021，206: 104805. DOI: 10.1016/j.still.2020.104805.
[16] DE MELO L B B，SILVA B M，PEIXOTO D S，CHIARINI T P A，DE OLIVEIRA G C，CURI N. Effect of compaction on the relationship between electrical resistivity and soil water content in oxisol[J]. Soil and Tillage Research，2021，208: 104876. DOI:10.1016/j.still.2020.104876.
[17] HUSSON O，HUSSON B，BRUNET A，BABRE D，ALARY K，SARTHOU J P，et al. Practical improvements in soil redox potential (Eh) measurement for characterisation of soil properties. Application for comparison of conventional and conservation agriculture cropping systems[J]. Analytica Chimica Acta，2016，906: 98-109. DOI: 10.1016/j.aca.2015.11.052.
[18] MIRJALILI S，MIRJALILI S M，LEWIS A. Grey wolf optimizer[J]. Advances in Engineering Software，2014，69: 46-61. DOI: 10.1016/j.advengsoft.2013.12.007.
[19] XIA Y Y，ZHANG C，WANG C X，LIU H J，SANG X X，LIU R，et al. Prediction of bending strength of glass fiber reinforced methacrylate-based pipeline UV-CIPP rehabilitation materials based on machine learning[J]. Tunnelling and Underground Space Technology，2023，140: 105319. DOI: 10.1016/j.tust. 2023.105319.
[20] 方晓彤，张立涛，闫秀霞，董春肖.基于BAS-BP神经网络的医药供应链协同绩效智能评价[J].计算机应用与软件，2023，40（8）：59-66，113. 10.3969/j.issn.1000-386x.2023.08.010. FANG X T，ZHANG L T，YAN X X，DONG C X. Intelligent evaluation of pharmaceutical supply chain collaborative performance based on BAS-BP neural network[J]. Computer Applications and Software，2023，40(8): 59-66，113.
[21] 梁海波，王怡，贾武升.基于机器学习的天然气钢质管道缺陷检测方法研究[J].安全与环境学报，2023，23（10）：3528-3537. 10.13637/j.issn.1009-6094.2022.1150. LIANG H B，WANG Y，JIA W S. Research on defect detection method for natural gas steel pipeline based on machine learning[J]. Journal of Safety and Environment，2023，23(10):3528-3537.
[22] VEL?ZQUEZ J C，CALEYO F，VALOR A，HALLEN J M. Predictive model for pitting corrosion in buried oil and gas pipelines[J]. Corrosion，2009，65(5): 332-342. DOI:10.5006/1.3319138.

相似文献/References:

[1]周祥.埋地钢质管道外涂层缺陷的安全质量分级评价[J].油气储运,2011,30(04):299.[doi:10.6047/j.issn.1000-8241.2011.04.017]
　Zhou Xiang.Safety quality grading evaluation on the external coating defects of buried steel pipeline[J].Oil & Gas Storage and Transportation,2011,30(11):299.[doi:10.6047/j.issn.1000-8241.2011.04.017]
[2]付吉强,吴明,杜明俊,等.埋地管道周围土壤湿热耦合相变过程的数值计算[J].油气储运,2011,30(01):30.[doi:10.6047/j.issn.1000-8241.2011.01.008]
　Fu Jiqiang,Wu Ming,Du Mingjun,et al.Numerical calculation on heat-moisture coupling and phase change process of the soil around buried pipeline[J].Oil & Gas Storage and Transportation,2011,30(11):30.[doi:10.6047/j.issn.1000-8241.2011.01.008]
[3]陈杨,王为民,陈伟聪,等.基于故障树与模糊理论的埋地管道风险评价[J].油气储运,2011,30(07):481.[doi:10.6047/j.issn.1000-8241.2011.07.001]
　Chen Yang,Wang Weimin,Chen Weicong,et al.Risk assessment of the buried pipeline based on the fault tree and fuzzy theory[J].Oil & Gas Storage and Transportation,2011,30(11):481.[doi:10.6047/j.issn.1000-8241.2011.07.001]
[4]胡士信,陆家榆,张本革,等.邻近强电线路管道交流干扰参数的测试方法[J].油气储运,2011,30(03):203.[doi:10.6047/j.issn.1000-8241.2011.03.012]
　Hu Shixin,Lu Jiayu,Zhang Benge,et al.Test method of AC interference parameters for pipeline adjacent power transmission lines[J].Oil & Gas Storage and Transportation,2011,30(11):203.[doi:10.6047/j.issn.1000-8241.2011.03.012]
[5]李朝阳,马贵阳,刘亮.埋地输油管道泄漏油品扩散模拟[J].油气储运,2011,30(09):674.[doi:10.6047/j.issn.1000-8241.2011.09.008]
　Li Zhaoyang,Ma Guiyang,Liu Liang,et al.Diffusion simulation on leaked oil for buried pipelines[J].Oil & Gas Storage and Transportation,2011,30(11):674.[doi:10.6047/j.issn.1000-8241.2011.09.008]
[6]王鸿膺,蒋涛,秦晓霞,等.川气东送管道土壤腐蚀埋片试验[J].油气储运,2010,29(10):769.[doi:10.6047/j.issn.1000-8241.2010.10.013]
　Wang Hongying,Jiang Tao,Qin Xiaoxia.Field Coupon Testing on the Soil Corrosivity of Sichuan-East Gas Pipeline[J].Oil & Gas Storage and Transportation,2010,29(11):769.[doi:10.6047/j.issn.1000-8241.2010.10.013]
[7]薛致远,张丰,毕武喜,等.东北管网阴极保护通电/断电电位测量与分析[J].油气储运,2010,29(10):772.[doi:10.6047/j.issn.1000-8241.2010.10.014]
　Xue Zhiyuan,Zhang Feng,Bi Wuxi.The Measurement and Analysis of On/Off-potential in Northeast Oil Pipeline Networks[J].Oil & Gas Storage and Transportation,2010,29(11):772.[doi:10.6047/j.issn.1000-8241.2010.10.014]
[8]刘仕鳌蒲红宇刘书文蒋洪.埋地管道应力分析方法[J].油气储运,2012,31(4):274.[doi:10.6047/j.issn.1000-8241.2012.04.009]
　Liu Shiao,Pu Hongyu,Liu Shuwen,et al.Stress analysis method of buried pipeline[J].Oil & Gas Storage and Transportation,2012,31(11):274.[doi:10.6047/j.issn.1000-8241.2012.04.009]
[9]吴峰,李向阳,张治军.埋地参数对管道同沟敷设传热的影响[J].油气储运,2010,29(11):835.[doi:10.6047/j.issn.1000-8241.2010.11.009]
　Wu Feng,Li Xiangyang,Zhang Zhijun.Influence of Buried Parameters for Heat Transfer of Pipelines Laid in One Ditch[J].Oil & Gas Storage and Transportation,2010,29(11):835.[doi:10.6047/j.issn.1000-8241.2010.11.009]
[10]赵雄,周卫军,郭瑞.SPAE评价法在管道防腐层性能评价中的应用[J].油气储运,2009,28(11):54.[doi:10.6047/j.issn.1000-8241.2009.11.015]
　ZHAO Xiong,ZHOU Weijun.Application of SPAE Method in Coating Performance Evaluation of Pipelines[J].Oil & Gas Storage and Transportation,2009,28(11):54.[doi:10.6047/j.issn.1000-8241.2009.11.015]
[11]宋福霖,赵弘,苗兴园.基于HMOGWO-RF的埋地管道点蚀深度机理-学习预测模型[J].油气储运,2024,43(11):1.
　Song Fulin,Zhao Hong,Miao Xingyuan.Mechanism-learning prediction model for pitting corrosion depth of buried pipeline based on HMOGWO-RF[J].Oil & Gas Storage and Transportation,2024,43(11):1.

备注/Memo

宋福霖，男，1998年生，在读博士生，2020年毕业于辽宁石油化工大学过程装备与控制工程专业，现主要从事无损检测方向的研究工作。地址：北京市昌平区府学路18号，102249。电话：13342127149。Email：2379991464@qq.com通信作者：赵弘，女，1975年生，教授，博士生导师，2003年博士毕业于西安交通大学机械电子工程专业，现主要从事无损检测方向的研究工作。地址：北京市昌平区府学路18号，102249。电话：13522390529。Email：hzhao_cn@163.com
基金项目：国家自然科学基金资助项目“海底油气管道封堵致振机理及气动减振控制方法研究”，51575528；中国石油大学（北京）前瞻导向及培育项目“油气管道及站场关键检测技术与实验方法研究”，2462022QEDX011。
· Received: 2023-11-22 · Revised: 2024-01-04 · Online: 2024-09-09

更新日期/Last Update: 2024-11-25