Reservoir Parameter Predicted Method Research Based On Seismic Attribute Analysis

The research of seismic attribute started in 1960's, and now has been widely applied after several developed steps. Seismic attributes are specific measurements of geometric, kinematic, dynamic, or statistical features derived from seismic data. Seismic attributes indicate all information subsets in initial seismic data and the relation of all kinds of attributes is very complicated. Attributes analysis is to show the tendency model unseen in initial seismic data through deleting irrelevant information. This enlarges the information hidden in initial seismic data, and improves application value of seismic data.The seismic attributes extracted from the seismic data amount to several dozens, but the more parameters can't guarantee the better result of the reservoir prediction. The invalid parameters not only increase the workload and waste the limited resources, but also bring about the dimension disaster. To predict reservoir parameters effectively, we must optimize the parameters for making use of these seismic attributesFirstly, seismic attributes are more profoundly understood through studying signification, classification, extraction methods and influence factors of seismic attributes. Secondly, about attribute optimization, Principle Component Analysis (PCA) and Independent Component Analysis (ICA) about blind source of attributes are mainly applied. The information of multiple seismic attributes is concentrated on dominant-component images of three-dimension seismic attributes by the methods. It improves the efficiency of seismic attribute application. Thirdly, emphasis is played on the study of wavelet transform in singularity detection of seismic data. Another attribute (singularity attribute) is extracted. The attribute provides a layered model of subsurface and precise location of sediment interface. Singularity attribute is computed directly from migration seismic data as a single-channel process. No velocity information is required to generate this attribute. On the basis of singularity attribute, sedimentary cycle is classified and sedimentary interface of sand-gravel fan is detected by using the time-frequency localization characteristic of the wavelet transform. The desired result is achieved. At last, the seismic attribute is applied to the reservoir prediction. Amplitude, frequency, and phase attributes are extracted from the seismic data of a certain area in eastern China. The methods of attribute optimization and the cluster analysis of multi-attributes based on self-organizing map network are combined with for analyzing the seismic attributes. The desired result has been achieved by predicting the gas reservoir of this area.