关键词: 多分量地震数据, 模糊理论, 隶属度函数, 地震反演, 油气藏预测

Abstract: The multi-component seismic technology can provide more seismic and physical parameters information used for predicting hydrocarbon distribution.The seismic data are of some illegibility and indeterminacy for description of object under the influence of acquisition and processing errors and noise.A fuzzy-evidential fusion method was presented to make the best information.On the basis of the fuzzy theory,the Gauss subordinate functions of P-wave and S-wave velocities and Poisson's ratio were calculated.Taking the log subordinate function as model,the probable function was designed,then basic confidence level for the three parameters was obtained from seismic inversion.Three total basic confidence levels were obtained by fuzzy-evidential fusion and used to describe reservoirs with hydrocarbon or without hydrocarbon and indeterminacy area respectively.According to the maximal principle of total basic confidence level,the hydrocarbon distribution could be predicted.This method was applied to a gas field in Sichuan area.The predictions are consistent with the drilling results.

Key words: multi-component seismic data, fuzzy theory, subordinate function, seismic inversion, hydrocarbon reservoir prediction