Study On Defect Detection Technology Of Coal Mine Vertical Wellbore

The vertical shaft is the throat of the mine,bearing the heavy responsibility of connecting the underground mineral resources with the surface production system.The long-serving shaft barrels have different degrees of deformation and spalling and other defects under the action of various factors.At present,the inspection task of coal mine vertical shaft inspection mainly relies on manual work,which has the problems of low efficiency and poor accuracy.There is an urgent need to develop a wellbore inspection robot to realize the automated inspection of vertical wellbores.The main needs of current vertical wellbore inspection are focused on the detection of wellbore deformation and well wall spalling,etc.This paper combines the inspection scheme of the wellbore inspection robot and designs a corresponding inspection system to achieve wellbore deformation and spalling detection using 3D laser scanning technology,and detects cracks near the spalling area using the camera,thus providing a basis for subsequent grouting repair.The main research work in this paper is as follows.(1)An inspection scheme based on inspection robots was designed by combining the actual vertical shaft inspection working conditions.The different types of defects and their potential connections in coal mine vertical wellbores are first analyzed,followed by the design of the inspection robot and its accompanying inspection process,and the selection of key hardware such as sensors.This provides the basis for the next research on the detection techniques for wellbore deformation,spalling and cracking defects.(2)A 3D laser scanning-based method for detecting defects in vertical wellbores is studied.Firstly,the point cloud pre-processing algorithm is introduced,followed by the traditional point cloud alignment ICP algorithm and the feature point alignment algorithm based on the fast histogram description sub-features,based on which the improved Spin-ICP algorithm is proposed,taking into account the environment of the coal mine vertical shaft.Then the radial deformation detection method of coal mine vertical shaft shaft is studied,and the RANSAC algorithm is applied to extract the cylindrical fitted section deformation variables,and the least squares method is used to fit the vertical shaft circular section to obtain the overall deformation of the vertical shaft shaft.Finally,a wellbore spalling detection method based on the Euclidean clustering algorithm is proposed for the wellbore spalling detection problem.(3)A computer vision-based crack detection method for vertical wellbore surface is developed.Based on the inspection process when the laser detects deformation and spalling,the visual inspection technology is further used to detect the scene crack defect form.The detection method based on computer vision technology is proposed,and the image pre-processing algorithm process is introduced,illustrating the principles and processing effects of contrast enhancement,image smoothing,morphological processing,etc.The characteristics of wellbore crack images are analyzed,and the corresponding algorithms for boxed average and boundary point crack detection are designed.Two calculation methods for calculating the crack image width by the point boundary algorithm are studied,and the principles of the two algorithms and the process of calculating the crack width are described.(4)Experimental verification and analysis.A wellbore simulation experimental platform is designed to compare the alignment effects of the fast histogram descriptor and the rotating image descriptor,and the detection accuracy of the proposed deformation detection algorithm and crack detection algorithm is verified through comparative analysis.This thesis includes 64 figures,7tables,and 93 references.