Abstract: Because of the strong heterogeneities of carbonate reservoir, the research about it has become a worldwide hard task for a long time. Recently, with the technical development in seismic acquisition and seismic attribute, it is possible to research carbonate reservoir with seismic attribute. Taking Well Block Tazhong-6 as an sample, the author quantitatively predict the fractures and “Visual reservoir space”distribution in this area, with new theory of Visual reservoir space and “reservoir petrophysical facies”, using new method of fuzzy neural comprehensive estimation. Finally, the most favorable reservoir and fracture-developed areas are pointed out, with the quantitative determination of “Visual reservoir space” of carbonate reservoir and the evaluation of its quality.

Key words: central Tarim basin, carbonate rock, reservoir rock, fuzzy math, neural network, petrophysics, fracture, quantitative analysis