基于管道爬行器的石油钻井套管形变检测技术研究

发布时间：2018-02-12 15:09

本文关键词： 形变检测管道爬行器轮廓提取位置检测　出处：《南京理工大学》2015年硕士论文　论文类型：学位论文

【摘要】：石油钻井套管的形变是造成油田钻井工作中存在安全隐患的重要因素,因此有必要研究石油钻井套管的形变检测技术。为了改进传统的机械检测方式在自动化程度方面的不足之处,本文设计了基于管道爬行器的管道形变图像检测系统。检测系统的设计以管道爬行器作为硬件基础,围绕管道爬行器,本文分析并搭建了管道形变图像检测系统的硬件平台和软件框架。系统的硬件平台主要由无线图像发射器与接收器、MTC-5308P摄像头与视频采集卡构建而成,其软件框架主要包括VS2008集成开发环境与OpenCV图像处理工具库。本文从管道形变的类型检测与位置检测两个方面对钻井套管的形变检测技术进行了研究。在形变的类型检测方面,通过分析管道形变的检测条件,从形变轮廓提取的角度分析并设计了相应的图像处理算法,包括经典边缘提取算法、数学形态学方法以及高级轮廓提取法。经实验证明,几种算法均能稳定且精确地检测出管道形变的类型。在形变的位置检测方面,本文分析了问题的关键在于如何获取形变在管道采集视频中所处的关键帧。按照这一思路,本文探索并提出了三种提取管道采集视频中形变所处关键帧的方法,分别是基于人工处理的关键帧提取法、基于模板匹配的关键帧提取法以及基于高级轮廓提取的关键帧提取法。通过理论与实际形变位置数据的比较和误差分析,发现几种算法所提取的视频关键帧均能有效确定管道形变的位置。
[Abstract]:Casing deformation in oil drilling is an important factor causing hidden safety problems in oil drilling. Therefore, it is necessary to study the deformation detection technology of oil drilling casing. A pipeline deformation image detection system based on pipeline crawler is designed in this paper. This paper analyzes and builds the hardware platform and software framework of the pipeline deformation image detection system. The hardware platform of the system is mainly composed of the wireless image transmitter and receiver MTC-5308P camera and video capture card. The software framework mainly includes VS2008 integrated development environment and OpenCV image processing tool library. This paper studies the deformation detection technology of drilling casing from two aspects of pipeline deformation type detection and location detection. By analyzing the detection conditions of pipeline deformation, the corresponding image processing algorithms are analyzed and designed from the point of view of deformation contour extraction, including classical edge detection algorithm, mathematical morphology method and advanced contour extraction method. Several algorithms can detect the type of pipeline deformation stably and accurately. In the aspect of deformation location detection, the key problem is how to obtain the key frame of deformation in pipeline video. In this paper, three methods of extracting the key frames of deformation in pipeline video are explored and proposed, which are based on manual processing. The key frame extraction method based on template matching and the key frame extraction method based on advanced contour extraction. It is found that the key frames extracted by several algorithms can effectively determine the position of pipeline deformation.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE931.2;TP391.41

【共引文献】