Sedimentary Facies Division Based On Seismic Attributes Analysis

Sediment facies reveal the sediment environment of the interest interval, the origin and distribution rule of the reservoir. The reservoir engineers can build geological model of the reservoir according to the research of the sediment facies, and make a basis for the processing and interpretation of the seismic data. There are some relatively mature methods for quantitative study of sedimentary facies distribution using 3D seismic data and visualization techniques. For some large regions with scarce drilling data and only 2D seismic data, however, it is questionable how to study sedimentary pattern quantitatively using seismic information. But this study is important to selecting regional risk exploration targets and predicting effective reservoir rocks laterally.By taking Minfeng sag as a case in this paper, under the control of the region sedimentary environment, firstly, we analyze the well logging faces with well data. Secondly, we research the transverse spread of sand bodies and the divided of the seismic faces with seismic attributes on the base of logging analysis. This paper is used to recognize the tiny sandstone with studying the amplitude spectrum fractal attribute, and has certain effect in identification of sand body, which is more accurate than conventional attribute; In predicting sand body thickness,we got the result more accurately through the combinatorial optimization and multivariate linear regression with single attribute and compound attribute, which accuracy reached 0.8 contrasting the thickness of sand body in well; In seismic facies division, we study a method called gray level co-occurrence matrix, which can be named VCM. According to the theory of VCM, seismic amplitude object can be transformed into seismic facies classifications uniting neural network. The contrast between seismic amplitude object including VCM and seismic facies classifications indicates that VCM can seize the primary seismic facies. Example research introducts that the classification of facies not only can conform the large-scale geological faces, but also can show some information which is difficult to be recognized with commen seismic interpretation. The result is more objective and overcomes the artificial haphazard in the conventional interpretation. Finally, the seismic facies are transformed into the sedimentary facies integrating log, geology and seismic data.