基于磁记忆技术的钻具检测及修复评价

发布时间：2018-03-02 17:35

本文选题：磁记忆检测技术　切入点：再制造　出处：《东北石油大学》2015年硕士论文　论文类型：学位论文

【摘要】：钻井作业是石油钻采过程中必不可少的重要环节,而钻具作为钻井作业的重要部件起着至关重要的作用,一旦钻具发生失效,将会迫使钻井作业停止,造成巨大的经济损失。为了有效节约资源、促进循环经济发展,多数钻具在失效后都会经再制造技术修复后继续服役,为了筛选出具有再制造价值的旧钻具,对其进行损伤检测与评价显得尤为重要。磁记忆检测作为一门新兴的无损检测技术,不仅能够检测宏观损伤,还能有效评价钻具的应力集中与微观损伤,因此在钻具损伤检测和再制造修复方面具有广阔的应用前景。本文在钻具载荷分析和失效分析的基础上,对不同损伤类型的钻具进行了磁记忆现场检测,通过磁记忆信号对比可得钻具的易损部位即螺纹连接处的磁记忆信号特征明显,说明磁记忆技术可以有效检测钻具损伤部位和其损伤程度,但同时发现单一信号特征参数无法全面表征损伤度。考虑到钻具实际磁记忆信号样本的有限性和模糊性,本文利用支持向量机的小样本性和模糊隶属度的抗噪性,在提取八维磁记忆特征参数的基础上,建立了基于距离的隶属度模糊支持向量机多分类模型,对钻具损伤等级进行分类,结果表明采用径向基核函数的分类精度最高。根据损伤等级将钻具分为三类,即正常钻具、可修复钻具和直接报废钻具,将具有可修复价值的钻具筛选出来,利用再制造技术对钻具进行相应的再造修复并检测修复后的磁记忆信号,通过分析信号发现再造钻具的应力集中程度相比于修复前明显减小,但某些再造钻具由于再制造过程中的失误造成损伤程度加重,应力集中程度反而增大,磁记忆信号很好地反映了钻具的应力集中与损伤状态,为工程实际中钻具的检测评价与再制造修复质量鉴定提供了一种新的方法。
[Abstract]:Drilling operation is an essential and important link in the process of oil drilling and production. As an important part of drilling operation, drilling tool plays an important role. Once the drilling tool fails, it will force the drilling operation to stop. In order to effectively save resources and promote the development of circular economy, most drilling tools will continue to be in service after failure after being repaired by remanufacturing technology, in order to screen out old drilling tools with remanufacturing value. It is very important to detect and evaluate the damage. As a new nondestructive testing technique, magnetic memory detection can not only detect macroscopic damage, but also effectively evaluate the stress concentration and microcosmic damage of drilling tools. Therefore, it has a broad application prospect in drilling tool damage detection and remanufacture repair. Based on the load analysis and failure analysis of drilling tool, the field magnetic memory detection of different damage types of drill tool is carried out. By comparing the magnetic memory signals, the characteristics of the magnetic memory signals in the damaged parts of the drill tools, that is, the threaded joints, are obvious, which shows that the magnetic memory technology can effectively detect the damaged parts and the damage degree of the drill tools. But at the same time, it is found that the single signal characteristic parameter can not fully characterize the damage degree. Considering the limitation and fuzziness of the actual magnetic memory signal samples of drilling tools, this paper makes use of the small sample character of support vector machine and the anti-noise property of fuzzy membership degree. On the basis of extracting the characteristic parameters of 8-dimensional magnetic memory, the multi-classification model of fuzzy support vector machine based on distance is established to classify the damage grade of drilling tools. The results show that the classification accuracy of radial basis function is the highest. According to the damage grade, the drilling tools are divided into three categories, that is, the normal drilling tools, the repairable drilling tools and the direct abandoned drilling tools, and the repairable drilling tools are screened out. The remanufacturing technology is used to reconstruct the drill tool and detect the magnetic memory signal after repairing. By analyzing the signal, it is found that the stress concentration degree of the reconstructed drill tool is obviously smaller than that before the repair. However, the degree of damage and stress concentration of some regenerated drilling tools are aggravated due to the errors in the process of remanufacturing. The magnetic memory signal can well reflect the stress concentration and damage state of the drill tool. It provides a new method for the testing and evaluation of drilling tools in engineering practice and the quality evaluation of remanufacturing repair.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE92

【参考文献】