【作者】 魏辉荣；

【导师】 熊金平；

【作者基本信息】 北京化工大学 ， 材料科学与工程， 2011， 硕士

【摘要】 随着油气工业的不断发展,金属材料在H2S/CO2体系中的腐蚀问题制约着油气的开发与生产,研究油气田用钢在服役环境中的腐蚀已经迫在眉睫。本文针对这一情况,研究了L360钢在H2S/CO2体系中的腐蚀规律及其机理,并开发了L360钢在酸性油气田中的腐蚀预测软件。本文利用静态挂片失重法和电化学技术,辅以扫描电镜(SEM)、能谱分析(EDS)和X射线衍射(XRD)技术,研究了溶液矿化度、元素硫含量和温度对L360钢H2S/C02腐蚀行为的影响以及L360钢-Crl3钢电偶对的电偶腐蚀行为,研究结果表明：(1)随着溶液矿化度升高,L360钢的腐蚀速率先升高后降低,当矿化度为67864mg/L时,腐蚀速率最大,其腐蚀产物均为四方晶系的FeS 1-x(Mackinawite),L360钢以H2S腐蚀为主。(2)元素硫的存在加速了L360钢的全面腐蚀,并导致严重的局部腐蚀；L360钢的腐蚀速率随着溶液中元素硫含量和温度的升高而增大,当元素硫含量达到20g/L时,其腐蚀速率最大,温度为70℃时腐蚀速率出现极大值点。(3)由L360钢和Cr13钢形成的电偶腐蚀电池中,L360钢成为电偶对的阳极,Crl3钢成为电偶对的阴极,随着阴阳极面积比增大,其电偶电流密度升高,电偶腐蚀效应越严重。本文运用人工神经网络技术建立了L360钢在H2S/CO2体系中的腐蚀预测模型,神经网络拓扑结构为5-4-1,网络模型训练成功以后,应用其对L360钢的H2S/C02腐蚀进行预测,预测结果表明,人工神经网络模型预测的结果与实验数据紧密相符,误差在14%以内。更多还原

【Abstract】 With the development of oil and gas industry, the corrosion of metal materials in H2S/CO2 system restricts the development and production. Therefore, it is imperative to study the corrosion behavior and mechanism of metal materials. In view of this situation, the corrosion behavior and mechanism of L360 steel was studied, and a corrosion prediction software in sour gas environment was developed.The influences of salinity, sulphur content and temperature on corrosion behavior of L360 steel in H2S/CO2 system and couple corrosion were studied by means of weight loss method, electrochemical techniques, SEM, EDS and XRD. The results are as follows:(1)The corrosion rate of L360 steel increases at first and then decreases with increasing of brine salinity. The peak of corrosion rate appears at 67864mg/L of brine salinity. Corrosion products- FeS1-x (Mackinaw) films are formed on the surfaces of the samples. H2S corrosion is the main one.(2)Elemental sulphur could accelerate the corrosion rate of L360 steel, and cause serious localized corrosion. The corrosion rate increased with the sulphur content and temperature increased. The peak of corrosion rate appears at 20g/L of elemental sulphur and 70℃.(3)Galvanic current density and galvanic effect between L360 steel and Cr13 steel increase with increasing the ratio of cathodic and anodic areas.An aritifical neueal network(ANN) model for the prediction of the corrosion rate of L360 steel in H2S/CO2 environment, neural network architecture is 5-4-1. After the succession of training, the model is applied to predict the corrosion rate of L360 steel in H2S/CO2 environment. The results show that the forecast results of network model check with the experimental date, the comparative error is less than 14%.更多还原