Research On The Description Method Of Carbonate Caves Based On Gradient Structure Tensor

The exploration and development of fractured vuggy carbonate reservoirs in China belongs to the frontier of oil and gas geophysical exploration.The oil resource content of carbonate reservoirs in China is estimated to be150×10⁸t.Among them,the estimated scale of carbonate oil reserves in the middle and deep parts of Tahe oilfield is 13.2×10⁸t.Fracture vuggy carbonate rocks in Tahe oilfield have great potential for development and research.However,on the one hand,the carbonate reservoir has strong heterogeneity,on the other hand,the buried depth of carbonate reservoir is more than 4000 meters,and the fracture cave reflection characteristics are complex,so it is difficult to quantitatively detect.Therefore,due to the fragmentation of fracture cave development characteristics,the difference of scale size,the limitation of distribution range and the heterogeneity of medium,the identification,characterization and evaluation of karst caves are faced with great challenges.In response to the above problems,based on the forward response characteristics of geological modeling,the thesis first discusses the cave detection method based on post-stack attributes,and then focuses on the cave description technology based on the local gradient structure tensor.On this basis,the construction of a cave is completed.The series is based on the cave marking method to describe and quantitatively describe the parameters of karst caves,and is applied to the detection of karst caves in carbonate reservoirs in the actual study area.It has achieved good application effects and lays good ground for the accurate and quantitative evaluation of karst cave-type carbonate reservoirs.The main achievements of this paper are as follows:(1)The seismic response characteristics of carbonate caves were analyzed by numerical simulation and actual seismic data,and the seismic response characteristics of different cave models were established,and it was clear that the seismic wave"bead-like"reflection characteristics are the reservoirs of carbonate caves in the Ordovician in the Tarim Basin.The main reflective feature mark for layer recognition.(2)The system has combed the image texture description theory of gradient structure tensor and the application framework in seismic attribute analysis.On this basis,combined with the chaotic reflection characteristics of carbonate cave reservoirs and the"beaded"reflection characteristics,the framework is based on the gradient structure tensor Compared with conventional cave identification attributes,C_cave is a preferred attribute that can effectively separate the chaotic reflections and"beaded"reflection characteristics of acid salt cave reservoirs from the seismic response characteristics of other geological bodies.(3)On the basis of the cave recognition attribute C_caveof the gradient structure tensor,image processing techniques such as filtering,gray enhancement,image segmentation,and hole calibration are used to perform finer positioning and measurement of the cave target to obtain more accurate cave location parameters(The coordinates of the centroid of the cavern,the corner coordinates of the enveloped rectangle or the enveloped cube),the size parameters(the pixel volume,cross-sectional area,and height of the cavern),and the degree of cavern development(flatness)The design uses the cave location and size parameters to calculate the spatial distribution non-uniformity parameters of the cave(surface porosity,degree of aggregation).(4)Using the above-mentioned image processing technologies such as the cave detection attribute C_cave based on the gradient structure tensor and the hole calibration,the quantitative characterization and analysis of the middle-deep karst caves in the Ordovician system of a work area in Tahe Oilfield was carried out,and 221 effective target karst caves were identified,and all the karst caves were completed.The calculation of the position and scale parameters of the body and the surface porosity have verified the effectiveness and practicability of the method by combining the drilling productivity.This completes the quantitative description and evaluation of the development scale and reservoir heterogeneity of carbonate reservoirs.This thesis is innovative in the use of GST to construct cave detection operators and the design of multiple quantitative evaluation parameters for caves,and has achieved good application effects in production practice,which proves the rationality of the method design and the reliability of the results.It has good theoretical design and application promotion value.