Research On The Fracture Detection Technology Of Seismic Multi-attributes

Fracture is the important spaces and migration channel of tight and lowpermeable reservoir. Fractured detection by seismic data is one of the focus researchtopics in oil and gas exploration field. Research on seismic identification andprediction technology for fractured reservoir in home and abroad has been carried outin a few decades. But because the fractured causes and scale problems and thecomplex relationship between seismic response characteristics and the resolution,fracture prediction by seismic data is still in progress and subsequent developmentstage.The seismic attribute analysis is one of the effective methods in reservoirprediction. The research of seismic attribute started in1960’s, and has been widelyapplied in oil and gas exploration areas after several developed steps. In thepractical application of seismic attribute extraction, single attribute extraction methodis difficult to complete seismic and geological interpretation and analysis becauseeach approach has its different goals. Under this circumstance, comprehensiveprediction by multi-disciplines, multi-method and multi-parameter is the way toreduce multiplicity in fracture detection and prediction.This thesis aims at the analysis and use of detection and prediction technology offractures for which the tectonic deformation is dominant factors, and it is mainly usedthat the post-stack seismic attribute analysis for fracture identification and prediction.An effective fracture prediction flow of multi-attribute was set up throughcomprehensive prediction by using single attribute and multi-attributes combination.The paper also summarized an attribute optimization analysis method in the studyarea. The research results is applied to dolomite fractures in Fengcheng in Northwestedge of Xinjiang Oilfield which come from an actual undertaken project, and providesstrong technical support in the accurately oil and gas prediction in the area.