酸水系统碳钢冲蚀规律及预测方法研究

发布时间：2018-09-12 18:31

【摘要】：炼油装置酸水系统冲蚀是原油劣质化、装置大型化、运行工况苛刻化发展过程中普遍存在的失效形式。特别是随着高硫、高酸、含氯等劣质原油加工比例的增加，酸水处理系统的流动腐蚀失效问题极为突出，严重威胁装置的长周期运行。针对酸水系统频繁出现流动腐蚀失效的现象，如何揭示其内在的失效机制并建立定量的失效预测方法，一直是学术界和工程界关注的焦点之一。本文以酸水汽提塔顶空冷出口管道的失效为研究对象，运用解剖分析、工艺模拟、流动状态计算相结合的研究方法，分析管件失效形貌，获得空冷器出口管道内多相流物性参数及腐蚀性介质的平衡分布规律，结合流动状态揭示空冷器出口管道系统冲蚀失效机理；针对酸水系统的腐蚀性介质组成，配制多相流腐蚀性介质，运用旋转式冲蚀实验装置采用Tafel极化曲线和SEM形貌分析法开展典型管材碳钢冲蚀规律研究，测试获得了NH4HS浓度、温度、H2S分压、氯离子含量及壁面剪切应力影响因素下碳钢的腐蚀产物膜特性及冲蚀特性，形成冲蚀实验数据库；结合冲蚀实验数据库，基于BP神经网络构建酸水系统的冲蚀速率预测模型，实现对酸水系统冲蚀速率的准确预测。研究结果表明：1）典型工况下空冷出口管道系统的NH4HS浓度高，流速快是引发管壁冲蚀减薄的主要原因；2）出口管道配置不平衡使多相流分布不均造成弯头外侧及上游内侧局部腐蚀严重，腐蚀性介质以不连续液膜形式流动，使得壁面腐蚀形貌为坑蚀；3）碳钢的冲蚀速率受腐蚀反应及腐蚀介质扩散双重因素的影响；4）腐蚀产物膜疏松、表面多孔、粘结性差使得壁面剪切应力高于特定值时，产物膜脱离材质基体而失去保护作用；5）采用冲蚀预测模型对已有实验结果进行预测，，平均误差在8.63%以内，冲蚀模型预测精度高，可推广应用于其它条件下碳钢管材腐蚀速率预测。本文的创新之处在于：1）运用解剖分析、工艺仿真、流动状态计算手段，确定酸水系统冲蚀失效机理；2）结合产物膜特性和冲蚀特性研究，分析不同影响因素下冲蚀规律的内在影响机制；3）基于BP神经网络建立了酸水系统冲蚀预测模型。
[Abstract]:The erosion of acid water system in oil refining unit is a common failure form in the development process of crude oil deterioration, large-scale plant and harsh operation conditions. Especially with the increase of the processing proportion of crude oil with high sulfur, high acid and chlorine content, the flow corrosion failure of the acid water treatment system is very prominent, which seriously threatens the long period operation of the plant. In view of the frequent flow corrosion failure of acid water system, how to reveal its internal failure mechanism and establish a quantitative failure prediction method has been one of the focuses of academic and engineering circles. In this paper, the failure of the air-cooled outlet pipe at the top of the acid water stripping tower is studied. The failure morphology of the pipe is analyzed by using the methods of anatomical analysis, process simulation and flow state calculation. The physical parameters of multiphase flow and the equilibrium distribution of corrosive media in the outlet pipeline of air cooler are obtained. Combined with the flow state, the erosion failure mechanism of air cooler outlet pipeline system is revealed, and the corrosive medium composition of acid water system is discussed. The corrosive medium of multiphase flow was prepared. The erosion law of typical carbon steel was studied by means of Tafel polarization curve and SEM morphology analysis method. The concentration of NH4HS and the partial pressure of H _ 2S were obtained by measuring the concentration of NH4HS, the temperature and the partial pressure of H _ 2S. Under the influence of chloride ion content and wall shear stress, the corrosion product film and erosion characteristics of carbon steel are formed, the erosion experiment database is formed, the erosion rate prediction model of acid water system is constructed based on BP neural network, and the erosion experiment database is combined with the erosion experiment database. The erosion rate of acid water system can be predicted accurately. The results show that the NH4HS concentration of the air-cooled outlet pipeline system is high under typical operating conditions and the high velocity of flow is the main reason for the erosion thinning of the pipe wall. 2) the uneven distribution of the multiphase flow in the outlet leads to serious corrosion on the outside of the elbow and the inner side of the elbow, and the corrosive medium flows in the form of discontinuous liquid film, which makes the corrosion appearance of the wall as pit corrosion; 3) the erosion rate of carbon steel is affected by the double factors of corrosion reaction and corrosion medium diffusion. 4) when the corrosion product film is loose, the surface is porous, and the adhesion is poor, the wall shear stress is higher than the specific value. The erosion prediction model is used to predict the experimental results. The average error is less than 8.63%, and the prediction accuracy of the erosion model is high. It can be applied to predict corrosion rate of carbon steel pipes under other conditions. The innovation of this paper lies in the application of anatomical analysis, process simulation and flow state calculation to determine the erosion failure mechanism of acid-water system. Based on BP neural network, the erosion prediction model of acid-water system is established.
【学位授予单位】：浙江理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE986

【参考文献】