Application of Seismic Multi Attribute Fusion Technology in Reservoir Rediction of Baer Depression

Seismic attribute analysis is an efficient and effective method to predict reservoir, but the strong ambiguity of the seismic single attribute for reservoir prediction may reduce the accuracy of prediction obviously, especially for the fault basin which has characteristics of multisource, facies changing fast, multiple episodes of volcanic activities, and the complex ingredients of rocks. The reservoir prediction for the technology of seismic multi attribute fusion is given. Firstly to use seismic waveform classification technique to divide study area into different regions according to sedimentary characteristics, then count the correlation coefficient of the seismic multiple attributes and reservoir information for different regions, besides to use linear fitting. Finally, using the fitting results of each region to summarize the result of reservoir prediction in the whole region. In the reservoir prediction of Nantun baer sag in Hailar Basin, with using sand ratio data of 164 wells in the target stratum and 7 selected seismic attributes, adopting gradually linear regression method to do fitting in the whole area, the correlation coefficient is only 0.52.