With the exploitation of oil and gas resources, especially in the context of the wide application of CO2 drive technology, the corrosion problem of oil and gas well tubular has become increasingly serious. By employing machine learning model ETR (Extra Tree Regression), this study establishes a nonlinear mapping relationship between environmental factors and corrosion rate. Meanwhile, model interpretable method SHAP is utilized to not only predict the corrosion rate under different conditions, but also provide the critical corrosion factors. The results show that Corrosion Inhibitor Concentration (CIC) and O2 content are the main factors affecting the corrosion rate of J55 tubing steel. By comparing the corrosion rates under different combinations of factors, this study summarizes the corrosion law of J55 tubular steel in a specific oil extraction well environment, which provides a scientific basis for anti-corrosion measures in Chang qing oilfield. Finally, experimental characterization is performed to analyze the corrosion law. The findings of this study are of great significance for optimizing the oilfield development strategy.
QiHLiYChenX, et al.Low-Temperature oxidation of light crude oil in oxygen-reduced air flooding. Pet Explor Dev2021; 48: 1393–1402.
2.
LiaoGYangHJiangY, et al.Applicable scope of oxygen-reduced air flooding and the limit of oxygen content. Pet Explor Dev2018; 45: 111–117.
3.
LiYYJiangZNZhangQH, et al.Unveiling the influential mechanism of O2 on the corrosion of N80 carbon steel under dynamic supercritical CO2 conditions. Corros Sci2022; 205: 110436.
4.
WangQWuWLiQ, et al.Under deposit corrosion of tubing served for injection and production wells of CO2 flooding. Eng Failure Anal2021; 127: 105540.
5.
ZhaoYLiuWDongB, et al.Damage analysis of rotation speed on corrosion film of 3Cr steel in the CO2 environment with silty sand. Eng Failure Anal2022; 133: 105995.
6.
LiuBLiuMLiuZ, et al.Nitrate-reducing-bacteria assisted hydrogen embrittlement of X80 steel in a near-neutral pH solution. Corros Sci2022; 202: 110317.
7.
ZhuGYLiYYHouBS, et al.Corrosion behavior of 13Cr stainless steel under stress and crevice in high pressure CO2/O2 environment. J Mater Sci Technol2021; 88: 79–89.
8.
LatosovEMaatenBSiirdeA,et al.The influence of O2 and CO2 on the possible corrosion on steel transmission lines of natural gas. Energy Proced2018; 147: 63–70.
9.
LinXLiuWWuF, et al.Effect of O2 on corrosion of 3Cr steel in high temperature and high pressure CO2-O2 environment. Appl Surf Sci2015; 329: 104–115.
10.
SinghAAnsariKRLinY, et al.Corrosion inhibition performance of imidazolidine derivatives for J55 pipeline steel in acidic oilfield formation water: electrochemical, surface and theoretical studies. J Taiwan Inst Chem Eng2019; 95: 341–356.
11.
WangXYangJChenX, et al.Synergistic inhibition properties and microstructures of self-assembled imidazoline and phosphate ester mixture for carbon steel corrosion in the CO2 brine solution. J Mol Liq2022; 357: 119140.
12.
OkaforPCLiuXZhengYG. Corrosion inhibition of mild steel by ethylamino imidazoline derivative in CO2-saturated solution. Corros Sci2009; 51: 761–768.
13.
YinZFZhangYLChangGR,et al.Corrosion behavior and characteristics of 3Cr steel in coexisting H2S- and CO2-containing solutions. J Mater Eng Perform2020; 29: 5442–5457.
14.
YinZFWangXZLiuL, et al.Characterization of corrosion product layers from CO2 corrosion of 13Cr stainless steel in simulated oilfield solution. J Mater Eng Perform2011; 20: 1330–1335.
15.
WangBDuMZhangJ,et al.Electrochemical and surface analysis studies on corrosion inhibition of Q235 steel by imidazoline derivative against CO2 corrosion. Corros Sci2011; 53: 353–361.
16.
LiuXOkaforPCZhengYG. The inhibition of CO2 corrosion of N80 mild steel in single liquid phase and liquid/particle TwoPhase flow by aminoethyl imidazoline derivatives. Corros Sci2009; 51: 744–751.
17.
FinšgarMJacksonJ. Application of corrosion inhibitors for steels in acidic Media for the oil and gas industry: a review. Corros Sci2014; 86: 17–41.
18.
LgazHSalghiRSubrahmanya BhatK, et al.Correlated experimental and theoretical study on inhibition behavior of novel quinoline derivatives for the corrosion of mild steel in hydrochloric acid solution. J Mol Liq2017; 244: 154–168.
19.
GuoXLiXLiuJ, et al.Fatigue failure mechanism of 3D tubing strings used in high-pressure, high temperature and high-yield curved gas wells. Eng Failure Anal2021; 128: 105536.
20.
LiuJWangHYuanZ. Forecast model for inner corrosion rate of oil pipeline based on PSO-SVM. Int J Simul Process Model2012; 7: 74–80.
21.
PengSZhangZLiuE, et al.A new hybrid algorithm model for prediction of internal corrosion rate of multiphase pipeline. J Nat Gas Sci Eng2021; 85: 103716.
22.
RenCQiaoWTianX. Natural gas pipeline corrosion rate prediction model based on BP neural network. Fuzzy Inf Eng Oper Res2012; 147: 449–455.
23.
XieMLiZZhaoJ,et al.A prognostics method based on back propagation neural network for corroded pipelines. Micromachines (Basel)2021; 12: 1568.
24.
LiaoKYaoQWuX,et al.A numerical corrosion rate prediction method for direct assessment of wet gas gathering pipelines internal corrosion. Energies2012; 5: 3892–3907.
25.
LiXJiaRZhangR, et al.A KPCA-BRANN based data-driven approach to model corrosion degradation of subsea oil pipelines. Reliab Eng Syst Saf2012; 219: 108231.
26.
OssaiCI. Corrosion defect modelling of aged pipelines with a feed-forward multi-layer neural network for leak and burst failure estimation. Eng Fail Anal2020; 110: 10439.
27.
AbbasMHNormanRCharlesA. Neural network modelling of high pressure CO2 corrosion in pipeline steels. Process Saf Environ2018; 119: 36–45.
28.
HernándezSNešićSWeckmanGR. Use of artificial neural networks for predicting crude oil effect on CO2 corrosion of carbon steels. Corrosion2005; 62: 467–482.
29.
OssaiCI. A data-driven machine learning approach for corrosion risk assessment-a comparative study. Big Data Cogn Comput2019; 3: 28.
30.
De MasiGGentileMVichiR, et al.Machine learning approach to corrosion assessment in subsea pipelines. Genova, Italy: OCEAN, 2015.
31.
SeghierMKeshtegarBKongFT. Prediction of maximum pitting corrosion depth in oil and gas pipelines. Eng Fail Anal2020; 112: 1–14.
32.
Al-AghbariMGujarathiAM. Hybrid optimization approach using evolutionary neural network & genetic algorithm in a real-world waterflood development. J Pet Sci Eng2022; 216: 110813.
33.
LuoZHuXGaoY. Corrosion research of wet natural gathering and transportation pipeline based on SVM. ICPTT, 2013. Trenchless Technology. https://doi.org/10.1061/9780784413142.102
34.
ZhangWDShenBAiYB,et al.Gas pipeline corrosion prediction based on modified support vector machine and unequal interval model. Appl Mech Mater2013; 373-375: 1987–1994.
ChengNLiMZhaoL, et al.Comparison and integration of computational methods for deleterious synonymous mutation prediction. Briefings Bioinf2020; 21: 970–981.
37.
ElamineBSMHöcheDZheludkevichM. Prediction of the internal corrosion rate for oil and gas pipeline: implementation of ensemble learning techniques. J Nat Gas Sci Eng2022; 99: 104425.
38.
DaiRZhangWTangW, et al.BBPpred: sequence-based prediction of blood-brain barrier peptides with feature representation learning and logistic regression. J Chem Inf Model2021; 61: 525–534.
39.
Ben SeghierMEAHöcheDZheludkevichM. Prediction of the internal corrosion rate for oil and gas pipeline: implementation of ensemble learning techniques. J Nat Gas Sci Eng2022; 99: 104425.
40.
XuFUszkoreitHDuY, et al.Natural language processing and Chinese computing. Lect Notes Comput Sci2019; 11839, Springer: 563–574.
41.
HuJL. Prediction of the internal corrosion rate for oil and gas pipelines and influence factor analysis with interpretable ensemble learning. Int J Press Vessels Pip2024; 212: 105329.
42.
WangZZhouTSundmacherK. Interpretable machine learning for accelerating the discovery of metal-organic frameworks for ethane/ethylene separation. Chem Eng J2022; 444: 136651.
43.
LiuKNiriMFApachiteiG, et al.Interpretable machine learning for battery capacities prediction and coating parameters analysis. Control Eng Pract2022; 124: 105202.
YinZFYuSZMaB, et al.Corrosion Evaluation and Mechanical Life Prediction of J55 Tubing Steel under Air-Foam Flooding Oil Well Conditions. J Mater Eng Perform2024. https://doi.org/10.1007/s11665-024-09780-1
46.
SongYWangQZhangX, et al.Interpretable machine learning for maximum corrosion depth and influence factor analysis. npj Mater Degrad2023; 7: 9.
47.
Ben SeghierMEAHöcheDZheludkevichM. Prediction of the internal corrosion rate for oil and gas pipeline: implementation of ensemble learning techniques. J Nat Gas Sci Eng2022; 99: 104425.
48.
DebKJainH. An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, part I: solving problems with box constraints. IEEE Trans Evol Comput2014; 18: 577–601.
49.
ChengRJinYOlhoferM,et al.A reference vector guided evolutionary algorithm for many-objective optimization. IEEE Trans Evol Comput2016; 20: 773–791.
50.
RoySChakrabortiN. Development of an Evolutionary Deep Neural Net for Materials Research. In: TMS 2020 149th Annual Meeting & Exhibition Supplemental Proceedings. The Minerals, Metals & Materials Series, Springer, Cham, 2020. https://doi.org/10.1007/978-3-030-36296-6_76
