Fracture And Fault Detection Based On Seismic Coherence Data

As a common geological structure,fractures and faults are important transportation channels and storage spaces for petroleum,natural gas and other geological resources,and their accurate description is of great significance before,during,and after the exploration of resources.Due to the complexity of geological conditions,the resolution of seismic data collected is usually low,and fractures and faults interpretations are susceptible to subjective cognitive factors.Traditional fractures and faults interpretation methods have low accuracy and reliability.Seismic coherence technology is a common seismic attribute enhancement technology.Through the similarity analysis of local neighborhoods of 3D seismic data,it can realize the enhanced display of fractures and faults in seismic data.Up to now,the third generation of seismic coherent algorithm has been developed.The first and second generations use cross-correlation algorithms to calculate similarity,and the third generation uses eigenvalue decomposition method which is more sensitive to fractures.The detection of fractures and faults based on seismic coherent volume data is a key step in the interpretation of seismic data and the construction of three-dimensional fractures and faults models,and it is also a hot topic at home and abroad.This thesis relies on horizontal scientific research projects,combined with edge detection technology in image processing,to study the algorithms and principles of automatic detection of fractures and faults in coherent data.The main research work of this thesis is as follows:(1)The principle of seismic curvature volume,variance volume and coherent volume algorithm was studied.The differences and relations between the first,second and third generation seismic coherence volume algorithms were analyzed,and their applicable conditions were given.(2)The noise characteristics of seismic coherent data were analyzed,and the denoising methods that protect the fractures and faults were studied.The differences and connections between them were compared experimentally,and the role of anisotropic diffusion filtering in detecting fractures and faults in seismic coherent volume data was analyzed.(3)The theory of fracture and fault detection based on the Hessian matrix was studied.In order to solve the problems of insufficient noise suppression and poor enhancement of fractures and faults in Jerman filtering,the noise suppression module and linear structure enhancement module in the Frangi operator were introduced to improve the fracture enhancement of tomography in low SNR data.(4)Aiming at the problem of low fracture contrast in seismic data and the easy detection of conventional space-based detection methods,a phase-consistency based fractures and faults detection method was introduced.By combining the Hessian matrix and phase-consistency fractures and faults detection results,the missing detection of fractures and faults was effectively reduced.(5)The mathematical morphology and skeleton refinement algorithms were studied.The structural characteristics of fractures and faults lines were analyzed.Aiming at the problem of more burrs in fractures and faults lines,a burr removal algorithm based on mathematical morphology analysis was proposed to improve fractures and faults lines.It improved the continuity and positioning accuracy of fractures and fault lines,and strengthened the geological significance of detection results.The validity and feasibility of this method were verified by processing and testing the actual seismic data in many areas.