论文信息 - Performance Evaluation of Complex Neural Networks in Reservoir Characterization: Applied to Boonsville 3-D Seismic Data

Performance Evaluation of Complex Neural Networks in Reservoir Characterization: Applied to Boonsville 3-D Seismic Data

Summary Geophysical reservoir characterization requires building a nonlinear relation between seismic attributes and rock/fluid properties computed from well logs. With such a relation, the rock/fluid properties computed from well logs can be extended to inter-well points. Neural networks are powerful tools to obtain such nonlinear relation. In this study, radial basis function (RBF) neural networks are evaluated in the application of porosity prediction, and multilayer RBF networks are confirmed to outperform traditional single layer RBF networks. Also, centroid based multilayer perceptron (CMLP) network, hybrid of RBF and multilayer perceptron (MLP), turns out to have the best performance among all the tested network types. Porosity of Caddo Member in Boonsville field, Texas is predicted by the CMLP network which has the best prediction performance among tests.

Tao Zhao | Kumar Ramachandran

[1] A. W. Jayawardena,et al. Comparison of Multilayer Perceptron and radial Basis Function networks as tools for flood forecasting , 1996 .

[2] Mirko van der Baan,et al. Neural networks in geophysical applications , 2000 .

[3] D. E. Lancaster,et al. 3-D seismic evidence of the effects of carbonate karst collapse on overlying clastic stratigraphy and reservoir compartmentalization , 1996 .

[4] Dave Marsden. Correcting the Boonsville 3-D seismic data set check-shot surveys , 1999 .

[5] Quincy Chen,et al. Seismic attribute technology for reservoir forecasting and monitoring , 1997 .

[6] Victor R. Lahti,et al. The Atoka Group (Pennsylvanian) of the Boonsville Field Area, North-Central Texas , 1982 .

[7] Jukka Saarinen,et al. Centroid based Multilayer Perceptron Networks , 2004, Neural Processing Letters.

[8] Mehdi Eftekharifar,et al. 3D Petrophysical Modeling Using Complex Seismic Attributes And Limited Well Log Data , 2011 .

[9] J. Chao,et al. A multilayer RBF network and its supervised learning , 2001, IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222).