Abstract：Oil and gas sandstone reservoirs of Mesozoic in Block L can be characterized by large differences in vertical thickness，and thus accurate prediction of the reservoir is extremely challenging，causing huge risks in well deployment appraisal.Through the comprehensive analysis of data in this area，this paper proposes a multi-element cooperative reservoir prediction technique.On the basis of post-stack pseudo acoustic impedance inversion，AVO analysis，and pre-stack simultaneous inversion，combined with multiple seismic attributes under multi-level control step by step，the accuracy of reservoir prediction has been significantly improved.Firstly，constrained sparse pulse inversion，which can reflect the characteristics of reservoir distribution，and AVO analysis with high reservoir recognition are carried out.At the same time，multiple elastic data volumes are obtained by pre-stack simultaneous inversion based on fine calibration and model establishment.Secondly，multi-element collaborative inversion is carried out by using seismic multi-attribute body，AVO analysis，and pre-stack simultaneous inversion results under the multiple control from well data，and then the effective porosity body is obtained by fitting the nonlinear mapping relationship between well curve and seismic attribute through a neural network.This data body has the advantages of a seismic multi-attribute and post-stack inversion body with high vertical resolution，and it has a high fluid identification ability that comes from AVO analysis and pre-stack simultaneous inversion.Then through the analysis of the effective porosity body，high-precision prediction of the reservoir is achieved.The effectiveness and practicability of this method have been verified by well performance data in actual production，which greatly improves the accuracy of reservoir prediction for the reservoir with sand bar facies and provides a reliable reference for the deployment of appraisal wells.
杨文采.地球物理反演的理论与方法[M].北京: 地质出版社,1997,242-267. YANG Wencai.The Theory and Methods of Geophysical Inversion[M].Geology Press,Beijing,1997,242-267.
[2]
刘宪斌,林金逞,韩春明,等.地震储层研究的现状及展望[J].地球学报,2002,23(1): 73-78. LIU Xianbin,LIN Jincheng,HAN Chunming,et al.Present conditions and prospects of the researches on seismic reservoir[J].Acta Geoscientica Sinica,2002,23(1): 73-78.
[3]
仝敏波,景春利,刘英明,等.叠前地震直接定量反演方法在火成岩储层预测中的应用研究[J].地球物理学进展,2011,26(5): 1741-1747. TONG Minbo,JING Chunli,LIU Yingming,et al.The application of direct quantitative pre-stack seismic inversion method in igneous reservoir prediction[J].Progress in Geophysics,2011,26(5): 1741-1747.
[4]
郎晓玲,彭仕宓,康洪全,等.叠前同时反演方法在流体识别中的应用[J].石油物探,2010,49(2): 164-169. LANG Xiaoling,PENG Shimi,KANG Hongquan,et al.Application of prestack simultaneous inversion in fluid identification[J].Geophysical Prospecting for Petroleum,2010,49(2): 164-169.
[5]
张绍红.概率神经网络技术在非均质地层岩性反演中的应用[J].石油学报,2008,29(4): 549-552. ZHANG Shaohong.Application of probabilistic neural network technique to lithology inversion of heterogeneous stratum[J].Acta Petrolei Sinica,2008,29(4): 549-552.
[6]
杨志力,周路,彭文利,等.BP神经网络技术在声波测井曲线重构中的运用[J].西南石油大学学报(自然科学版),2008,30(1): 63-66. YANG Zhili,ZHOU Lu,PENG Wenli,et al.Application of BP neural network technology in sonic log data rebuilding[J].Journal of Southwest Petroleum University (Science & Technology Edition),2008,30(1): 63-66.
[7]
MASTERS T.Signal and Image Processing with Neural Networks[M].Wiley,Hoboken,1994.
[8]
HAMPSON D P,SCHUELKE J S,QUIREIN J A.Use of multiattribute transforms to predict log properties from seismic data[J].Geophysics,2001,66(1): 220-236.
