Application Of Post-stack Seismic Attribute Analysis In Prediction Of Favorable Reservoirs In Turpan Area, Western Margin Of Taipei Depression

In order to solve the problems of oil exploration and development, Turpan-Hami Oilfield deployed Shengnan-Yanmuxi high-precision 3D exploration in Western Taibei Sag. Through research, the bottom of the Cretaceous sand mainly deposits fan delta frontal subfacies in the Turpan region, sand body varies fast horizontally and widely distributes, the reservoir is thin, it is difficult to distinguish. Current information and means can not meet the needs of development and evaluation, therefore, we carry out reservoir research and hydrocarbon detection by post stack seismic attribute analysis technology, Implementing the distribution of favorable reservoir in target area, to provide the basis for further exploration and development.Based on the theoretical study on seismic attributes, seismic attributes were applied to predict beneficial zones of the Cretaceous strata in Western Taibei Sag, Turpan-Hami Basin. Based on the 3D high precision seismic data, the seismic interpretation of the target layer is explained, and the structural features of the study area is understood through the structural interpretation and fault interpretation. The bottom of the Cretaceous in the whole study area is a ofnose-uplift structure from east to west, which uplift to the west. The entire study area mainly developed two sets of faults, a group of strike-slip fault is nearly nee, which controlled the distribution pattern of Turpan 2 structure, there is a row of small echelon faults and perpendicular to it, by which, the structure of Turpan 2 was divided into several parallel small fault block. Analysis of the seismic attribute obtained approximate distribution of sand body in the study area. By the RD inversion, the distribution of reservoirs have a more precise description, The sand body was striped across north-east. Then using the software extract the seismic attributes in target layer with Geoeast sections. By manual selection and automatic selection it will be optimized for seismic attributes to predict sand thickness and porosity prediction. The thickness and porosity of the reservoir are predicted by BP neural network, the basic error control in less than 4. Finally, the fluid detection predicted the hydrocarbon potential of the study area. By attribute analysis, inversion, predication methods for physical properties and detection of fluid distribution, favorable reservoir was predicted. Which is the favorable reservoir distribution areas in Turpan No.2 structure and more favorable reservoir distribution area in faulted anticline on the eastern side of Ya 2.