| China’s energy structure is mainly more coal and less gas-based.Jincheng anthracite coal is rich in Shanxi province,but its poor permeability seriously restricts the development and utilization of coalbed methane.The permeability mainly depends on the pore and fracture structure of coal.Due to the influence of geological characteristics,buried depth,multi-field coupling and other factors,the coal seam structure has changed in the process of coal mining,which results in the change of micro-pore and fracture structure in the coal seam,large pores,micro-fractures and fractures are the migration channels of coalbed methane.To study the spatial development,connectivity and distribution characteristics of coal seam structure and to improve permeability is the key point of efficient coal seam gas extraction.Because of the diversity and inhomogeneity of coal,the traditional methods,such as liquid nitrogen method,mercury injection method and electron microscope,are inadequate in the quantitative study of pore and fracture structure.In this paper,the non-destructive testing of high-precisionμCT225k VFCB scanning system,based on the combination of image processing and 3D visualization reconstruction technology,make up for the shortcomings of traditional research methods.Through pyrolysis experiments,the coal seam permeability is changed by increasing the internal temperature of coal sample and by the effect of temperature field on the microstructure of anthracite.Therefore,it is of great significance for the development of coalbed methane to study the evolution law of coalbed fracture structure with temperature under pyrolysis of Jincheng anthracite.In this paper,Jincheng anthracite is taken as the research object,based on pyrolysis and CT scanning experiments,using image processing technology to Jincheng anthracite CT image self-programming.Based on the traditional image processing segmentation algorithm,the coal and rock CT image segmentation algorithm is optimized and improved.By combining the advantages of CV and LBF algorithm,the computation time is shortened,and the high quality micro-structure diagram is obtained.The 3D fracture network of pyrolytic anthracite was reconstructed by 3D visualization technique,and the spatial distribution and development of3D single fracture surface were studied by establishing the main fracture surface tensor.The main contents and conclusions of the experiment are as follows:（1）pyrolysis anthracite test and CT scan testThe self-developed high-temperature atmosphere furnace was combined with high-precision CT microscope system,and the key parameter I=75ΜA was set up by heating at10°C/MIN in the experiment process of cycle Chuck fixing-high-temperature atmosphere furnace heating-scanning,through real-time scanning of anthracite samples at different temperatures,CT slices of coal samples at seven temperatures were obtained.It was found that the CT images contained minerals with high gray values,it was found that the fracture of coal sample expanded strongly in bedding plane,and the gray value of material boundary in CT image was obvious at room temperature and 100°C.With the increase of temperature,the structure of internal pore and fracture changes and the gray value of material changes.The total number of slices per sample is 1506,the size of image matrix of each slice is 2040×2040,the storage size is 15.8 mb,the format is“Bin”,the gray value is 0-255.（2）based on image processing,LGIF model segmentation algorithm is establishedBased on image processing and CT image of pyrolysis anthracite,the recognition of single-layer pore and fracture structure of anthracite at different temperatures is accomplished by self-programming.The wavelet function is used to initialize the CT image and remove the high threshold substance in the CT image.Through Graythresh function,the segmentation points of pore cracks and coal matrix are obtained,and the thresholds are written into the archive table,and the thresholds are converted from[0,1]to[0,255],and the CT images after impurity removal are normalized.The pore-fracture structure of pyrolysis anthracite was modeled by LGIF image segmentation algorithm.By combining CV model with LBF model,the optimized image segmentation algorithm based on LGIF model is found,which improves the iteration number of original CV model and processing efficiency.It can not only solve the difference of gray level,but also solve the sensitive problem of LBF to initial position.At the same time,the image of pyrolytic anthracite is segmented twice,which improves the segmentation precision of the inner hole and fissure of anthracite.The precise segmentation of the microstructure of pyrolytic anthracite is realized,and the spatial distribution of the microstructure of anthracite is clearly reflected.（3）analysis of two-dimensional morphological parameters of cracks in pyrolysis anthraciteBased on the two-dimensional image of meso-structure of pyrolytic anthracite,the evolution and development of typical fractures with different strike angles on bedding plane were studied by using image analysis technique.It is found that the initial fracture morphology of anthracite bedding is mainly composed of three stages:the first stage is at room temperature-200°C,where the pores are dominant in room temperature and 100°C bedding;The second stage at 200-300°C is the gentle stage,and the third stage at 300-600°C is the violent stage with the direction of 400-600°C fracture bedding.The transition of fracture geometry at 400°C is a key temperature point for fracture development in high temperature section.（4）three-dimensional reconstruction and quantitative characterization of meso-structure of pyrolytic anthraciteThere are a lot of fractures with different scales in coal rock.The results show that the number of fractures with equivalent diameter Deq<100μm is large and the volume is small by using the equivalent sphere model in the Quantitative analysis software The number of fractures with equivalent diameter Deq>1000μm is less,but the volume is larger,the connectivity is better,so the pore network model is established to observe the distribution of pores and fractures more clearly. |
